any recommendation for roaming ISPs (no NAT , ipsec , ssh pptp friendly).

[Boris shtrasman]

Hi,

I'm searching for a roaming ISP preferred using HSDPA or equivalent but not a
must.
I wish to be able to take my laptop and work for any urban area in IL
(Non GushDan areas), already checked few ISPs but they failed to
answer my needs (Pelehone  Cellcom):

Pelephone wasn't able to hold ipsec sessions, Cellcom provided me an
awful customer service (perhaps It's just my luck).

* The most important is to get not natted ip:
I need to be able to do ipsec + l2tp connections and lots of udp based
communication (zrtp/ srtp and plain old rtp).

* Most of the traffic is encrypted and going to a few selected servers
located in europe, it ain't a fixed rate (but not more then 1Gb per
day tops):
inner communication is :
A. rdp + vnc communication (as a client).
B. VoIP : sip + [zs]rtp heavy use  ~8 hours of talks on a daily basis and more.
C. version control (svn) : not more then 10mbs per day

I don't care about DPI , and privacy issues, but I do care about using
encrypted data transfer (as I use vpns ).

* preferred small delays but this is not a must - at least : download
rate  256 kbps  and upload 16 kbps.
(Kilo Byte)

* should support linux and windows without custom applications - I
must be able to communicate to the device using some public documented
way:
wvdial , mannual AT commands a kernel module already existing in the
kernel etc .. .
custom app for windows / linux is a big no-no for me.

Could anyone recommend something ?

___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

ppp/pptp disable multiple login

[Hetz Ben Hamo]

Hi,

I'm trying to find a simple solution to a small problem: I built a PPTP
service, but I want to disable the option to let a user login multiple times
at once.
I looked at poptop and other related packages and I cannot find anything,
which is quite surprising. Someone suggested me to look at PAM but that
didn't get me too far, I didn't see there any such option.

Any suggestion?

Thank,
Hetz
___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: ppp/pptp disable multiple login

[Oleg Goldshmidt]

Hetz Ben Hamo het...@gmail.com writes:

 Hi,

 I'm trying to find a simple solution to a small problem: I built a
 PPTP service, but I want to disable the option to let a user login
 multiple times at once.

 I looked at poptop and other related packages and I cannot find
 anything, which is quite surprising. Someone suggested me to look at
 PAM but that didn't get me too far, I didn't see there any such
 option.

 Any suggestion?

Disclaimer: I am no expert, just trying to apply common sense...

I suspect it is not within the scope of pptp/ppp/chap/whatever to
provide such an option - this stuff is way too low level for user
management or anything like that, so I am not surprised. 

Maybe you should look into combining poptop with something that
handles authentication, e.g., like RADIUS (FreeRADIUS?), and look for
options there?

-- 
Oleg Goldshmidt | p...@goldshmidt.org

___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: [Partial Resoltion] Re: PPTP issues and Barak over cable

[Ilya Konstantinov]

That's the basic thing they teach in all the Israeli Cable Internet FAQs :)

I even published a wrapper that does this automatically:
http://mirror.hamakor.org.il/archives/linux-il/10-2003/6117.html

On 1/30/07, Micha Feigin [EMAIL PROTECTED] wrote:


I appears that pptp (or pppd, don't know) set the gw for gre tunnel data
to the
tunnel itself causing a loopback of the data. This is one thing that
windows
seems to handle right but linux mixes up completely (it seems that at
least
some routing information is sent but either it is wrong or linux miss
handles
it).

Overriding the routing to the pptp server solved the problem.

RE: [Partial Resoltion] Re: PPTP issues and Barak over cable

[Micha Feigin]

I did it initially, but barak changed something, I'm not sure if the IPs of
their servers or something else and it got the whole thing messed up again. Took
me some time to figure out the reason (since the first time I used directions
that gave explicit IPs).
 
I think I got the overriding automatic now for the next time. Finding the right
gateway is more of a problem though.


  _  

From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf
Of Ilya Konstantinov
Sent: Thursday, February 01, 2007 4:37 PM
To: Micha Feigin
Cc: linux-il@linux.org.il
Subject: Re: [Partial Resoltion] Re: PPTP issues and Barak over cable


That's the basic thing they teach in all the Israeli Cable Internet FAQs :)

I even published a wrapper that does this automatically:
http://mirror.hamakor.org.il/archives/linux-il/10-2003/6117.html


On 1/30/07, Micha Feigin [EMAIL PROTECTED]  mailto:[EMAIL PROTECTED] 
wrote: 

I appears that pptp (or pppd, don't know) set the gw for gre tunnel data to the 
tunnel itself causing a loopback of the data. This is one thing that windows
seems to handle right but linux mixes up completely (it seems that at least
some routing information is sent but either it is wrong or linux miss handles 
it).

Overriding the routing to the pptp server solved the problem.





+++ 
This Mail Was Scanned By Mail-seCure System 
at the Tel-Aviv University CC.

PPTP issues and Barak over cable

[Micha Feigin]

Sorry for starting a new thread, due to connection issues I have patial net
access.

I connect to the internet over cable (hot) using barak with a pptp connection.

Thursday night they upgraded their system (don't know more details) and since
then I can't connect using linux anymore. Windows works fine, so it's not a
hardware issue. I tried two machines, one debian stable (powerpc) both with the
standard pptp package and the latest version compiled from source. The other is
debian unstable.

Both machines show the same symptoms.

The machine connects, bringing up the ppp connection and recieves an IP. It then
seems to send (according to ifconfig) about 300MB to 1GB over the course of a
minute and a half or so and the connection then dies.

I tried both sycronous and asyncronous connections but it didn't help

Ifconfig:
---

ppp0  Link encap:Point-to-Point Protocol  
  inet addr:89.0.121.41  P-t-P:172.26.255.17  Mask:255.255.255.255
  UP POINTOPOINT RUNNING NOARP MULTICAST  MTU:1500  Metric:1
  RX packets:4 errors:0 dropped:0 overruns:0 frame:0
  TX packets:884278 errors:0 dropped:0 overruns:0 carrier:0 collisions:0
txqueuelen:3 
  RX bytes:64 (64.0 b)  TX bytes:321621278 (306.7 MiB)

Tcpdump produces the following output (repeating over and over) which I don't
know how to decipher:

02:23:26.104538 IP 172.29.214.255  172.26.255.17: GREv1, call 63029, seq
885576, length 108: IP 172.29.214.255  172.26.255.17: GREv1, call 63029, seq
885562, length 72: IP [|ip]
02:23:26.104666 IP 172.29.214.255  172.26.255.17: GREv1, call 63029, seq
885577, length 144: IP 172.29.214.255  172.26.255.17: GREv1, call 63029, seq
885563, length 108: IP [|ip]
02:23:26.104790 IP 172.29.214.255  172.26.255.17: GREv1, call 63029, seq
885578, length 180: IP 172.29.214.255  172.26.255.17: GREv1, call 63029, seq
885564, length 144: IP [|ip]
02:23:26.104946 IP 172.29.214.255  172.26.255.17: GREv1, call 63029, seq
885579, length 432: IP 172.29.214.255  172.26.255.17: GREv1, call 63029, seq
885568, length 396: IP [|ip]
02:23:26.105275 IP 172.29.214.255  172.26.255.17: GREv1, call 63029, seq
885580, length 1480: IP truncated-ip - 36 bytes missing! 172.29.214.255 
172.26.255.17: GREv1, call 63029, seq 885569, length 1480: IP [|ip]
02:23:26.105381 IP 172.29.214.255  172.26.255.17: gre
02:23:26.107608 IP 172.29.214.255  172.26.255.17: GREv1, call 63029, seq
885581, length 72: IP 172.29.214.255  172.26.255.17: gre

The output in syslog:

Jan 30 02:21:16 litshi kernel: ppp_async: Unknown symbol crc_ccitt_table
Jan 30 02:21:37 litshi pptp[20157]: anon log[main:pptp.c:267]: The synchronous
pptp option is NOT activated 
Jan 30 02:21:37 litshi pptp[20160]: anon log[ctrlp_rep:pptp_ctrl.c:251]: Sent
control packet type is 1 'Start-Control-Connection-Request' 
Jan 30 02:21:37 litshi pptp[20160]: anon log[ctrlp_disp:pptp_ctrl.c:738]:
Received Start Control Connection Reply
Jan 30 02:21:37 litshi pptp[20160]: anon log[ctrlp_disp:pptp_ctrl.c:772]: Client
connection established.
Jan 30 02:21:38 litshi pptp[20160]: anon log[ctrlp_rep:pptp_ctrl.c:251]: Sent
control packet type is 7 'Outgoing-Call-Request' 
Jan 30 02:21:38 litshi pptp[20160]: anon log[ctrlp_disp:pptp_ctrl.c:857]:
Received Outgoing Call Reply.
Jan 30 02:21:38 litshi pptp[20160]: anon log[ctrlp_disp:pptp_ctrl.c:896]:
Outgoing call established (call ID 0, peer's call ID 63029). 
Jan 30 02:21:38 litshi pppd[20163]: pppd 2.4.4 started by root, uid 0
Jan 30 02:21:38 litshi pppd[20163]: using channel 2
Jan 30 02:21:38 litshi pppd[20163]: Using interface ppp0
Jan 30 02:21:38 litshi pppd[20163]: Connect: ppp0 -- /dev/pts/2
Jan 30 02:21:38 litshi pppd[20163]: sent [LCP ConfReq id=0x1 asyncmap 0x0
magic 0xc56c1737 pcomp accomp]
Jan 30 02:21:40 litshi kernel: Shorewall:net2fw:DROP:IN=eth0 OUT=
MAC=08:00:46:5b:70:62:00:05:00:e7:dd:9b:08:00 SRC=172.26.255.17
DST=172.29.214.255 LEN=54 TOS=0x00 PREC=0x00 TTL=253 ID=51790 PROTO=47 
Jan 30 02:21:40 litshi pptp[20158]: anon log[decaps_gre:pptp_gre.c:388]:
accepting packet 2
Jan 30 02:21:40 litshi pppd[20163]: rcvd [LCP ConfAck id=0x1 asyncmap 0x0
magic 0xc56c1737 pcomp accomp]
Jan 30 02:21:41 litshi pppd[20163]: sent [LCP ConfReq id=0x1 asyncmap 0x0
magic 0xc56c1737 pcomp accomp]
Jan 30 02:21:41 litshi pptp[20158]: anon log[decaps_gre:pptp_gre.c:388]:
accepting packet 3
Jan 30 02:21:41 litshi pppd[20163]: rcvd [LCP ConfAck id=0x1 asyncmap 0x0
magic 0xc56c1737 pcomp accomp]
Jan 30 02:21:42 litshi pptp[20158]: anon log[decaps_gre:pptp_gre.c:388]:
accepting packet 4
Jan 30 02:21:42 litshi pppd[20163]: rcvd [LCP ConfReq id=0x2 auth pap magic
0xd088f19e]
Jan 30 02:21:42 litshi pppd[20163]: sent [LCP ConfAck id=0x2 auth pap magic
0xd088f19e]
Jan 30 02:21:42 litshi pppd[20163]: sent [LCP EchoReq id=0x0 magic=0xc56c1737]
Jan 30 02:21:42 litshi pppd[20163]: sent [PAP AuthReq id=0x1 user=lipshi
password=hidden]
Jan 30 02:21:42 litshi pptp[20158]: anon log[decaps_gre:pptp_gre.c:388]:
accepting packet 5
Jan 30 02:21:42

RE: cable connection to barak (pptp)

[Micha Feigin]

Do you have a link by chance? I can't find it

I had a working connection according to all those manuals, the problem is that
barak did a major upgrade on their pptp server and my current configuration no
longer works. Now I need to know if it's possible to continue working with them
or whether I need to change providers.

Anyone else connecting a linux box directly to barak over cable around here?

Thanks 

 -Original Message-
 From: Baruch Shpirer [mailto:[EMAIL PROTECTED] 
 Sent: Friday, January 26, 2007 9:35 PM
 To: [EMAIL PROTECTED]
 Subject: Re: cable connection to barak (pptp)
 
 checkout pptp scripts in netvision site, i remember having 
 the same problem, you need to replace the ips to get the 
 packets to the tunnel entrence, just download the script and 
 compare and ull understand
 
 - Original Message -
 From: [EMAIL PROTECTED]
 To: Linux-IL,  linux-il@linux.org.il
 Sent: Friday, January 26, 2007 20:54
 Subject: cable connection to barak (pptp)
 
 
 I connect to the internet using cable (hot) through barak. 
 It appears that 
 last
  night they had a major upgrade of their pptp server and my 
 linux box no 
  longer
  connects.
 
  pptp connection comes up and recieves local and remote 
 addresses and dns
  address, but no packets pass over the pptp connection (the 
 remote server
  doesn't answer pings as well as any others).
 
  I tried connecting using a windows box and it connects 
 fine. I also tried
  upgrading pptp from 1.5 to 1.7 but to no avail.
 
  Does anyone know how to fix the connection or any way to 
 find out what is 
  wrong?
 
  Thanks
 
 
 
  
  This message was sent using IMP, the Internet Messaging Program.
 
 
  To 
  unsubscribe, send mail to [EMAIL PROTECTED] with
  the word unsubscribe in the message body, e.g., run the command
  echo unsubscribe | mail [EMAIL PROTECTED]
 
 
  !DSPAM:45ba5179232671935225532!
 
  
 
 
 
 
 +++
  This Mail Was Scanned By Mail-seCure System
  at the Tel-Aviv University CC.
 
 


=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

RE: cable connection to barak (pptp)

[Micha Feigin]

Ok, I found the scripts, they're at cables2.netvision.net.il, will report again
if it works 

 -Original Message-
 From: Baruch Shpirer [mailto:[EMAIL PROTECTED] 
 Sent: Friday, January 26, 2007 9:35 PM
 To: [EMAIL PROTECTED]
 Subject: Re: cable connection to barak (pptp)
 
 checkout pptp scripts in netvision site, i remember having 
 the same problem, you need to replace the ips to get the 
 packets to the tunnel entrence, just download the script and 
 compare and ull understand
 
 - Original Message -
 From: [EMAIL PROTECTED]
 To: Linux-IL,  linux-il@linux.org.il
 Sent: Friday, January 26, 2007 20:54
 Subject: cable connection to barak (pptp)
 
 
 I connect to the internet using cable (hot) through barak. 
 It appears that 
 last
  night they had a major upgrade of their pptp server and my 
 linux box no 
  longer
  connects.
 
  pptp connection comes up and recieves local and remote 
 addresses and dns
  address, but no packets pass over the pptp connection (the 
 remote server
  doesn't answer pings as well as any others).
 
  I tried connecting using a windows box and it connects 
 fine. I also tried
  upgrading pptp from 1.5 to 1.7 but to no avail.
 
  Does anyone know how to fix the connection or any way to 
 find out what is 
  wrong?
 
  Thanks
 
 
 
  
  This message was sent using IMP, the Internet Messaging Program.
 
 
  To 
  unsubscribe, send mail to [EMAIL PROTECTED] with
  the word unsubscribe in the message body, e.g., run the command
  echo unsubscribe | mail [EMAIL PROTECTED]
 
 
  !DSPAM:45ba5179232671935225532!
 
  
 
 
 
 
 +++
  This Mail Was Scanned By Mail-seCure System
  at the Tel-Aviv University CC.
 
 


=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

RE: cable connection to barak (pptp)

[Micha Feigin]

 -Original Message-
 From: [EMAIL PROTECTED] 
 [mailto:[EMAIL PROTECTED] On Behalf Of Micha Feigin
 Sent: Saturday, January 27, 2007 3:21 PM
 To: linux-il@linux.org.il
 Subject: RE: cable connection to barak (pptp)
 
 Ok, I found the scripts, they're at cables2.netvision.net.il, 
 will report again if it works 


Still no good. I tried the script. The pptp connection comes up, revieves ip and
dns but nothing is sent over the connection.
The log of running both pptp and pppd in debug mode (from syslog) is at the end
if anyone can make any sense of it.

Thanks
 
  -Original Message-
  From: Baruch Shpirer [mailto:[EMAIL PROTECTED]
  Sent: Friday, January 26, 2007 9:35 PM
  To: [EMAIL PROTECTED]
  Subject: Re: cable connection to barak (pptp)
  
  checkout pptp scripts in netvision site, i remember having the same 
  problem, you need to replace the ips to get the packets to 
 the tunnel 
  entrence, just download the script and compare and ull understand
  
  - Original Message -
  From: [EMAIL PROTECTED]
  To: Linux-IL,  linux-il@linux.org.il
  Sent: Friday, January 26, 2007 20:54
  Subject: cable connection to barak (pptp)
  
  
  I connect to the internet using cable (hot) through barak. 
  It appears that
  last
   night they had a major upgrade of their pptp server and my
  linux box no
   longer
   connects.
  
   pptp connection comes up and recieves local and remote
  addresses and dns
   address, but no packets pass over the pptp connection (the
  remote server
   doesn't answer pings as well as any others).
  
   I tried connecting using a windows box and it connects
  fine. I also tried
   upgrading pptp from 1.5 to 1.7 but to no avail.
  
   Does anyone know how to fix the connection or any way to
  find out what is
   wrong?
  
   Thanks
  

Jan 27 18:26:22 rice pppd[25839]: pppd 2.4.3 started by root, uid 0
Jan 27 18:26:22 rice pppd[25839]: Using interface ppp0
Jan 27 18:26:22 rice pppd[25839]: Connect: ppp0 -- /dev/ttyp1
Jan 27 18:26:22 rice pptp[25842]: anon log[main:pptp.c:276]: The synchronous
pptp option is NOT activated
Jan 27 18:26:22 rice pptp[25844]: anon log[ctrlp_rep:pptp_ctrl.c:251]: Sent
control packet type is 1 'Start-Control-Connection-Request'
Jan 27 18:26:22 rice pptp[25844]: anon log[ctrlp_disp:pptp_ctrl.c:738]: Received
Start Control Connection Reply
Jan 27 18:26:22 rice pptp[25844]: anon log[ctrlp_disp:pptp_ctrl.c:772]: Client
connection established.
Jan 27 18:26:23 rice pptp[25844]: anon log[ctrlp_rep:pptp_ctrl.c:251]: Sent
control packet type is 7 'Outgoing-Call-Request'
Jan 27 18:26:23 rice pptp[25844]: anon log[ctrlp_disp:pptp_ctrl.c:857]: Received
Outgoing Call Reply.
Jan 27 18:26:23 rice pptp[25844]: anon log[ctrlp_disp:pptp_ctrl.c:896]: Outgoing
call established (call ID 0, peer's call ID 33837).
Jan 27 18:26:23 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: accepting
packet 1
Jan 27 18:26:23 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: accepting
packet 2
Jan 27 18:26:23 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: accepting
packet 3
Jan 27 18:26:23 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: accepting
packet 4
Jan 27 18:26:23 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: accepting
packet 5
Jan 27 18:26:23 rice pppd[25839]: PAP authentication succeeded
Jan 27 18:26:23 rice pppd[25839]: kernel does not support PPP filtering
Jan 27 18:26:23 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: accepting
packet 6
Jan 27 18:26:23 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: accepting
packet 7
Jan 27 18:26:23 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: accepting
packet 8
Jan 27 18:26:23 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: accepting
packet 9
Jan 27 18:26:23 rice pppd[25839]: replacing old default route to eth0
[172.29.208.1]
Jan 27 18:26:23 rice pppd[25839]: Cannot determine ethernet address for proxy
ARP
Jan 27 18:26:23 rice pppd[25839]: local  IP address 89.0.9.77
Jan 27 18:26:23 rice pppd[25839]: remote IP address 172.26.255.17
Jan 27 18:26:23 rice pppd[25839]: primary   DNS address 212.150.49.10
Jan 27 18:26:23 rice pppd[25839]: secondary DNS address 62.90.42.110
Jan 27 18:26:24 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: accepting
packet 10
Jan 27 18:26:45 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: accepting
packet 11
Jan 27 18:27:05 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: accepting
packet 12
Jan 27 18:27:26 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: accepting
packet 13
Jan 27 18:27:34 rice dnsmasq[442]: reading /etc/resolv.conf
Jan 27 18:27:34 rice dnsmasq[442]: using nameserver 62.90.42.110#53
Jan 27 18:27:34 rice dnsmasq[442]: using nameserver 212.150.49.10#53
Jan 27 18:27:34 rice dnsmasq[442]: using local addresses only for domain
luna.local
Jan 27 18:27:46 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: accepting
packet 14
Jan 27 18:27:51 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388

RE: cable connection to barak (pptp)

[Omer Zak]

From the connection script, it looks to me like the remote side
disconnects from you after the connection has been established.

The log of pptp is not complete.  There are supposed to be some messages
by pptp before starting pppd.

There is something strange with the numbers of sent and received bytes.
You seem to have sent 199MB and received 164 bytes while your connection
was alive.  If this is real data transfer, then I'd look for some
process which clogged the PPP channel with its own junk data, preventing
any other data from passing.

   --- Omer

On Sat, 2007-01-27 at 17:04 +0200, Micha Feigin wrote:
  -Original Message-
  From: [EMAIL PROTECTED] 
  [mailto:[EMAIL PROTECTED] On Behalf Of Micha Feigin
  Sent: Saturday, January 27, 2007 3:21 PM
  To: linux-il@linux.org.il
  Subject: RE: cable connection to barak (pptp)
  
  Ok, I found the scripts, they're at cables2.netvision.net.il, 
  will report again if it works 
 
 
 Still no good. I tried the script. The pptp connection comes up, revieves ip 
 and
 dns but nothing is sent over the connection.
 The log of running both pptp and pppd in debug mode (from syslog) is at the 
 end
 if anyone can make any sense of it.
 
 Thanks
  
   -Original Message-
   From: Baruch Shpirer [mailto:[EMAIL PROTECTED]
   Sent: Friday, January 26, 2007 9:35 PM
   To: [EMAIL PROTECTED]
   Subject: Re: cable connection to barak (pptp)
   
   checkout pptp scripts in netvision site, i remember having the same 
   problem, you need to replace the ips to get the packets to 
  the tunnel 
   entrence, just download the script and compare and ull understand
   
   - Original Message -
   From: [EMAIL PROTECTED]
   To: Linux-IL,  linux-il@linux.org.il
   Sent: Friday, January 26, 2007 20:54
   Subject: cable connection to barak (pptp)
   
   
   I connect to the internet using cable (hot) through barak. 
   It appears that
   last
night they had a major upgrade of their pptp server and my
   linux box no
longer
connects.
   
pptp connection comes up and recieves local and remote
   addresses and dns
address, but no packets pass over the pptp connection (the
   remote server
doesn't answer pings as well as any others).
   
I tried connecting using a windows box and it connects
   fine. I also tried
upgrading pptp from 1.5 to 1.7 but to no avail.
   
Does anyone know how to fix the connection or any way to
   find out what is
wrong?
   
Thanks
   
 
 Jan 27 18:26:22 rice pppd[25839]: pppd 2.4.3 started by root, uid 0
 Jan 27 18:26:22 rice pppd[25839]: Using interface ppp0
 Jan 27 18:26:22 rice pppd[25839]: Connect: ppp0 -- /dev/ttyp1
 Jan 27 18:26:22 rice pptp[25842]: anon log[main:pptp.c:276]: The synchronous
 pptp option is NOT activated
 Jan 27 18:26:22 rice pptp[25844]: anon log[ctrlp_rep:pptp_ctrl.c:251]: Sent
 control packet type is 1 'Start-Control-Connection-Request'
 Jan 27 18:26:22 rice pptp[25844]: anon log[ctrlp_disp:pptp_ctrl.c:738]: 
 Received
 Start Control Connection Reply
 Jan 27 18:26:22 rice pptp[25844]: anon log[ctrlp_disp:pptp_ctrl.c:772]: Client
 connection established.
 Jan 27 18:26:23 rice pptp[25844]: anon log[ctrlp_rep:pptp_ctrl.c:251]: Sent
 control packet type is 7 'Outgoing-Call-Request'
 Jan 27 18:26:23 rice pptp[25844]: anon log[ctrlp_disp:pptp_ctrl.c:857]: 
 Received
 Outgoing Call Reply.
 Jan 27 18:26:23 rice pptp[25844]: anon log[ctrlp_disp:pptp_ctrl.c:896]: 
 Outgoing
 call established (call ID 0, peer's call ID 33837).
 Jan 27 18:26:23 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: 
 accepting
 packet 1
 Jan 27 18:26:23 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: 
 accepting
 packet 2
 Jan 27 18:26:23 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: 
 accepting
 packet 3
 Jan 27 18:26:23 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: 
 accepting
 packet 4
 Jan 27 18:26:23 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: 
 accepting
 packet 5
 Jan 27 18:26:23 rice pppd[25839]: PAP authentication succeeded
 Jan 27 18:26:23 rice pppd[25839]: kernel does not support PPP filtering
 Jan 27 18:26:23 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: 
 accepting
 packet 6
 Jan 27 18:26:23 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: 
 accepting
 packet 7
 Jan 27 18:26:23 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: 
 accepting
 packet 8
 Jan 27 18:26:23 rice pptp[25842]: anon log[decaps_gre:pptp_gre.c:388]: 
 accepting
 packet 9
 Jan 27 18:26:23 rice pppd[25839]: replacing old default route to eth0
 [172.29.208.1]
 Jan 27 18:26:23 rice pppd[25839]: Cannot determine ethernet address for proxy
 ARP
 Jan 27 18:26:23 rice pppd[25839]: local  IP address 89.0.9.77
 Jan 27 18:26:23 rice pppd[25839]: remote IP address 172.26.255.17
 Jan 27 18:26:23 rice pppd[25839]: primary   DNS address 212.150.49.10
 Jan 27 18:26:23 rice pppd[25839]: secondary DNS address 62.90.42.110
 Jan 27 18:26:24 rice

cable connection to barak (pptp)

[michf]

I connect to the internet using cable (hot) through barak. It appears that last
night they had a major upgrade of their pptp server and my linux box no longer
connects.

pptp connection comes up and recieves local and remote addresses and dns
address, but no packets pass over the pptp connection (the remote server
doesn't answer pings as well as any others).

I tried connecting using a windows box and it connects fine. I also tried
upgrading pptp from 1.5 to 1.7 but to no avail.

Does anyone know how to fix the connection or any way to find out what is wrong?

Thanks




This message was sent using IMP, the Internet Messaging Program.


To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

PPtP in Slax LiveCD

[Ancient1]

Hello List   :-) 



Latest, SLAX  5.1.6, 


Motorolla SB4200 on USB ETH0. Detected OK

012 script - using precompiled pptp-linux - Installs OK.


Cable-Start  = IFCONFIG = everthing looks ok : eth0 , ppp0 .. = 
nothing but HOT !?



012 support said to manually change /etc/ppp/pptp.conf  make this 
connection default i.e MakeDefault=NO 

The result of this manual change is that the cable-start command , after 
stating that PAP authentication succeeded , doesn't state that it didn't 
change the default gateway (which is the one assigned automatically by 
HOT) .



I get ppp0 IP and everthing looks normal to me ... but I don't 
understand much in these matters .



does anyone know how to approach this issue beside demanding from 012 
no-login connection (which is not solving the issue at all) .



did anyone try L2TP with 012  ?   (probably modding Actcom script )


Thanx



=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: Question : Connect to barak 013 through cable (PPTP)

[Ilya Konstantinov]

Ilia K. wrote:


Some time ago I tried to get a PPTP connection work,
but as you could saw, sometimes one pptp packet is lost and this cause
lots of problems in Linux. The driver tries to buffer subsequental
packets and wait for the lost one until a timeout occurs.
It's generally always a good idea to use your system's traffic shaping 
tools to artificially limit your upstream to what _you_ know to be the 
cap set by your ISP. I'm saying what _you_ know since, unlike with old 
dial-ups, on cables/ADSL your computer gets no hints about the upstream 
offered to you.
On Linux, you can do this easily with the 'tc' tool (use a TBF queuing 
discipline) on the pppN interface. On Windows 2000, you may use its QoS 
features, though admittedly it's a bit more complicated to set up.
After doing that, your system will never send more packets through the 
interface than the interface can handle, and more important, it could 
schedule more important packets (VoIP, pings...) to go out first.

1. Install the latest kernel (I heard, some changes were done recently
regarding PPTP)
Minor remark: PPTP support is done entirely in user space. The kernel is 
not involved here. The kernel only deals with PPP.




=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: Question : Connect to barak 013 through cable (PPTP)

[Ilia K.]

On 5/23/06, ILAN [EMAIL PROTECTED] wrote:

Hi

I just found out this thread using Google.

I'm connected to 012 on Suse 10.1 using their login scripts.
I'm using Motorola SB4200 on USB .

I have a lot of packet losses (?) etc.  My download tests vs HOT are OK but vs
012 are 50% !

Example from System Log :
anon log[decaps_gre:pptp_gre.c:407]: buffering packet 55280 (expecting 55279,
lost or reordered)

Did you discover the cause of the problem   ?
012 Support claim its the problem of the Linux USB and tell me to use NIC
instead which doesn't fit my needs .

Thank you

Ilan



Hi Ilan!
I actually have no ADSL or Cable connection now (I connect through
Technion LAN). Some time ago I tried to get a PPTP connection work,
but as you could saw, sometimes one pptp packet is lost and this cause
lots of problems in Linux. The driver tries to buffer subsequental
packets and wait for the lost one until a timeout occurs. Probably
this timeout exeeds some TCP/IP one, so a TCP/IP level thinks, that
lots of packets were lost and tries to retransmit them all. I suppose,
windows PPTP driver just passes all data to upper levels, so TCP/IP
level sees that only one packet was lost and only this packet is
retransmitted. So IMHO the problem of lost packets persists in Windows
too, but it's less notable. Also I don't think, that the USB modem is
what causes your problems. I've seen the same with ethernet-connected
modems.
So I can advise you to try the following:

1. Install the latest kernel (I heard, some changes were done recently
regarding PPTP)

2. Look for all kernel options in PPTP section. Probably you can
disable/minimise the time, which takes PPTP level to decide, that some
packets are lost (and not just reordered).

3. Try a different ISP. You can ask some ISP to give you 1 day trial
without any obligations (you don't have to unsubscribe from 012
service for that). If you find some ISP which isn't suffer from this
problem, just switch to it. If the problem persists with all ISPs you
should talk with HOT/Bezeq tech.support and insist on sending their
technician to your home to check the connection. Probably you'll have
to do it several times, before they actually fix something.

5. Change to PPPoE protocol. This may require to change the modem /
internet provider / connecting method (cable/adsl). AFAIK PPPoE causes
less problems with linux.

6. If you will be very strong, you can demand from your ISP the
connection without a dialer, which means all you should do to
coonect' to the internet (to get a real IP) is just to make DHCP
request. No annoying tunnels any more! Most ISP will claim at first,
that they don't offer such a service. Don't beleve, they do. This
require a little bit more work from their tech. stuff, but it isn't
hard to make this possible. Probably most salesmen don't know about
such option, so you can ask to talk with their headman. Some ISPs can
tell you, that such a service require additional pay, but some won't.
You can also say to 012, that you will discontinue to use their
service, if they don't give you such connection. So, after some
fighting you have a chance to get very suitable connection.

And some common tip: subscribe to linux-il mailing list (send the word
subscribe to  [EMAIL PROTECTED]), so you can discuss
your problem with more, than one person. I'm CC'ing this email to
Linux-IL. There are people in this list, which probably have more
experience with ADSL/Cable connecting, than I have (and they probably
check their mailboxes more frequently, than me). Also, if you discover
some solution, others will be able to find it in maillist archives and
benefit from this too.

Currently I have to prepare for some exams, so I don't think I'll have
a time to check this mailbox (and make replies) in the near future :(

Good luck,
Ilia.

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: Question : Connect to barak 013 through cable (PPTP)

[amir]

Hi,
  One option which was not mentioned: try to connect by L2TP.  See:
http://support.actcom.net.il/support/tips/cableen/cableRedhat_l2tp.php

Amir

On May 27, 13:10, Ilia K. wrote:
} Subject: Re: Question : Connect to barak 013 through cable (PPTP)
 On 5/23/06, ILAN [EMAIL PROTECTED] wrote:
  Hi
 
  I just found out this thread using Google.
 
  I'm connected to 012 on Suse 10.1 using their login scripts.
  I'm using Motorola SB4200 on USB .
 
  I have a lot of packet losses (?) etc.  My download tests vs HOT are OK but 
  vs
  012 are 50% !
 
  Example from System Log :
  anon log[decaps_gre:pptp_gre.c:407]: buffering packet 55280 (expecting 
  55279,
  lost or reordered)
 
  Did you discover the cause of the problem   ?
  012 Support claim its the problem of the Linux USB and tell me to use NIC
  instead which doesn't fit my needs .
 
  Thank you
 
  Ilan
 
 
 Hi Ilan!
 I actually have no ADSL or Cable connection now (I connect through
 Technion LAN). Some time ago I tried to get a PPTP connection work,
 but as you could saw, sometimes one pptp packet is lost and this cause
 lots of problems in Linux. The driver tries to buffer subsequental
 packets and wait for the lost one until a timeout occurs. Probably
 this timeout exeeds some TCP/IP one, so a TCP/IP level thinks, that
 lots of packets were lost and tries to retransmit them all. I suppose,
 windows PPTP driver just passes all data to upper levels, so TCP/IP
 level sees that only one packet was lost and only this packet is
 retransmitted. So IMHO the problem of lost packets persists in Windows
 too, but it's less notable. Also I don't think, that the USB modem is
 what causes your problems. I've seen the same with ethernet-connected
 modems.
 So I can advise you to try the following:
 
 1. Install the latest kernel (I heard, some changes were done recently
 regarding PPTP)
 
 2. Look for all kernel options in PPTP section. Probably you can
 disable/minimise the time, which takes PPTP level to decide, that some
 packets are lost (and not just reordered).
 
 3. Try a different ISP. You can ask some ISP to give you 1 day trial
 without any obligations (you don't have to unsubscribe from 012
 service for that). If you find some ISP which isn't suffer from this
 problem, just switch to it. If the problem persists with all ISPs you
 should talk with HOT/Bezeq tech.support and insist on sending their
 technician to your home to check the connection. Probably you'll have
 to do it several times, before they actually fix something.
 
 5. Change to PPPoE protocol. This may require to change the modem /
 internet provider / connecting method (cable/adsl). AFAIK PPPoE causes
 less problems with linux.
 
 6. If you will be very strong, you can demand from your ISP the
 connection without a dialer, which means all you should do to
 coonect' to the internet (to get a real IP) is just to make DHCP
 request. No annoying tunnels any more! Most ISP will claim at first,
 that they don't offer such a service. Don't beleve, they do. This
 require a little bit more work from their tech. stuff, but it isn't
 hard to make this possible. Probably most salesmen don't know about
 such option, so you can ask to talk with their headman. Some ISPs can
 tell you, that such a service require additional pay, but some won't.
 You can also say to 012, that you will discontinue to use their
 service, if they don't give you such connection. So, after some
 fighting you have a chance to get very suitable connection.
 
 And some common tip: subscribe to linux-il mailing list (send the word
 subscribe to  [EMAIL PROTECTED]), so you can discuss
 your problem with more, than one person. I'm CC'ing this email to
 Linux-IL. There are people in this list, which probably have more
 experience with ADSL/Cable connecting, than I have (and they probably
 check their mailboxes more frequently, than me). Also, if you discover
 some solution, others will be able to find it in maillist archives and
 benefit from this too.
 
 Currently I have to prepare for some exams, so I don't think I'll have
 a time to check this mailbox (and make replies) in the near future :(
 
 Good luck,
 Ilia.
 
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]
 

:~v


=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: Connect to barak 013 through cable (PPTP)

[guy keren]

you missed one important point - how is the windows installed on the same
hardware (if there is any) does with regards to internet connection? does
it also show the same problem, or it manages to work properly?

--guy

On Fri, 19 Aug 2005, Ilia K. wrote:

 Hi All again!

 Thanks to all guys, who have replied me!
 However, nobody have told me yet, whether I've missed some point in
 the configuration.
 I want to emphasize the fact, that even under very low load (one ssh
 connection only), _half_ of pings are lost when pinging the PPTP
 server or some public servers.
 Under no circumstances this can be treated as normal cable operation.

 Please, someone who has working cable connection, send me output of:
 ifconfig; route -n

 This could help me decide, whether my configuration is OK or buggy
 (and if it's OK, I'll apply to the cable company).

 Thanks.

 On 8/17/05, Ilia K. [EMAIL PROTECTED] wrote:
  Hi, All!
 
  I've convinced one girl to install linux instead of winXP. She is in
  Jerusalem and I am in Haifa, but we managed to install SuSe 9.2 and
  connect it to barak's cable internet through PPTP. Now I can ssh to
  the computer. Unfortunately, the connection is _very_ unstable and
  slow. Half of the packets are seems to be lost. I suppose, the problem
  can be somewhere in routing (this is my weak point, so the
  probability for the mistakes is high).
  I very want, that this girl will stay with linux, but internet
  connection seems to be the bottle neck. I'll appreciate any
  feedback, especially helpful one :)
 
  PPTP is set up by executing:
  linux:~# pppd call barak
 
  The settings (after PPTP is up) are:
 
  modem's address (in modem-eth0 network) seems to be 172.25.192.1
 
  linux:~ # ifconfig eth0; ifconfig ppp0
  eth0Link encap:Ethernet  HWaddr 00:50:8D:62:7C:5B
  inet addr:172.25.212.201  Bcast:255.255.255.255 Mask:255.255.224.0
  inet6 addr: fe80::250:8dff:fe62:7c5b/64 Scope:Link
  UP BROADCAST NOTRAILERS RUNNING MULTICASTMTU:1500  Metric:1
  RX packets:253799 errors:0 dropped:0 overruns:0 frame:0
  TX packets:14236 errors:0 dropped:0 overruns:0 carrier:0
  collisions:0 txqueuelen:1000
  RX bytes:20082953 (19.1 Mb)TX bytes:2413726 (2.3 Mb)
  Interrupt:10 Base address:0xb000
 
  ppp0Link encap:Point-to-Point Protocol
  inet addr:85.64.160.99  P-t-P:172.26.255.17  Mask:255.255.255.255
  UP POINTOPOINT RUNNING NOARP MULTICASTMTU:1000  Metric:1
  RX packets:203 errors:0 dropped:0 overruns:0 frame:0
  TX packets:153 errors:0 dropped:0 overruns:0 carrier:0
  collisions:0 txqueuelen:3
  RX bytes:14557 (14.2 Kb)  TX bytes:13792 (13.4 Kb)
 
  linux:~ # route -n
  Kernel IP routing table
  Destination   Gateway Genmask Flags Metric RefUse Iface
  172.26.255.17 172.25.192.1255.255.255.255 UGH   0  00 eth0
  172.26.255.17 0.0.0.0   255.255.255.255 UH0  00 ppp0
  172.25.192.0  0.0.0.0 255.255.224.0   U 0  00 eth0
  169.254.0.0   0.0.0.0 255.255.0.0 U 0  00 eth0
  127.0.0.0 0.0.0.0 255.0.0.0 U 0  00 lo
  0.0.0.0   172.26.255.17   0.0.0.0 UG0  00 ppp0
 
  Note: The first route (172.26.255.17/32 via 172.25.192.1) was manualy
  added before running pppd/pptp.
 
  linux:~ # cat /var/lib/dhcpcd/dhcpcd-eth0.info
  IPADDR=172.25.212.201
  NETMASK=255.255.224.0
  NETWORK=172.25.192.0
  BROADCAST=255.255.255.255
  GATEWAY=172.25.192.1
  HOSTNAME='linux'
  DNS=192.168.101.102,192.168.101.101
  DHCPSID=213.57.35.2
  DHCPGIADDR=172.25.160.1
  DHCPSIADDR=0.0.0.0
  DHCPCHADDR=00:50:8D:62:7C:5B
  DHCPSHADDR=00:05:00:E7:CD:A6
  DHCPSNAME=''
  LEASETIME=426054
  RENEWALTIME=213027
  REBINDTIME=372797
  INTERFACE='eth0'
  CLASSID='Linux 2.6.8-24-default i686'
  CLIENTID=00:50:8D:62:7C:5B
 
  linux:~ # rpm -q pptp
  pptp-1.5.0-2
 
  linux:~ # cat /etc/ppp/peers/barak
  linkname barak
  noauth
  user SOME_USERNAME
  pty /usr/sbin/pptp 172.26.255.17 --nolaunchpppd
  lock
  noipdefault
  usepeerdns
  nobsdcomp
  nodeflate
  lcp-echo-failure 10
  lcp-echo-interval 20
  defaultroute
  #mtu 1460
  mtu 1000
  mru 1000
  debug
 
 
  In /var/log/messages I see lots of pptp's messages like these:
  Aug 17 20:41:52 linux pptp[10933]: anon
  log[decaps_gre:pptp_gre.c:404]: buffering packet 4098 (expecting 4082,
  lost or reordered)
  Aug 17 20:41:52 linux pptp[10933]: anon
  log[decaps_gre:pptp_gre.c:404]: buffering packet 4099 (expecting 4082,
  lost or reordered)
  Aug 17 20:41:52 linux pptp[10933]: anon
  log[decaps_gre:pptp_gre.c:404]: buffering packet 4100 (expecting 4082,
  lost or reordered)
  Aug 17 20:41:52 linux pptp[10933]: anon
  log[decaps_gre:pptp_gre.c:404]: buffering packet 4101 (expecting 4082,
  lost or reordered)
  Aug 17 20:41:52 linux pptp[10933]: anon
  log[decaps_gre:pptp_gre.c:404]: buffering packet 4102 (expecting 4082

Re: Connect to barak 013 through cable (PPTP)

[Ilia K.]

You are right, Guy. This is an important point, I haven't reported. 
Currently, there is no Win on the computer, so I can't test it, but
the girl, who owns the comp says it has been working well under
windows. She doesn't remember download speed etc., but she says, that
browsing experience was definitely better, than now.

On 8/19/05, guy keren [EMAIL PROTECTED] wrote:
 
 you missed one important point - how is the windows installed on the same
 hardware (if there is any) does with regards to internet connection? does
 it also show the same problem, or it manages to work properly?
 
 --guy
 
 On Fri, 19 Aug 2005, Ilia K. wrote:
 
  Hi All again!
 
  Thanks to all guys, who have replied me!
  However, nobody have told me yet, whether I've missed some point in
  the configuration.
  I want to emphasize the fact, that even under very low load (one ssh
  connection only), _half_ of pings are lost when pinging the PPTP
  server or some public servers.
  Under no circumstances this can be treated as normal cable operation.
 
  Please, someone who has working cable connection, send me output of:
  ifconfig; route -n
 
  This could help me decide, whether my configuration is OK or buggy
  (and if it's OK, I'll apply to the cable company).
 
  Thanks.
 
  On 8/17/05, Ilia K. [EMAIL PROTECTED] wrote:
   Hi, All!
  
   I've convinced one girl to install linux instead of winXP. She is in
   Jerusalem and I am in Haifa, but we managed to install SuSe 9.2 and
   connect it to barak's cable internet through PPTP. Now I can ssh to
   the computer. Unfortunately, the connection is _very_ unstable and
   slow. Half of the packets are seems to be lost. I suppose, the problem
   can be somewhere in routing (this is my weak point, so the
   probability for the mistakes is high).
   I very want, that this girl will stay with linux, but internet
   connection seems to be the bottle neck. I'll appreciate any
   feedback, especially helpful one :)
  
   PPTP is set up by executing:
   linux:~# pppd call barak
  
   The settings (after PPTP is up) are:
  
   modem's address (in modem-eth0 network) seems to be 172.25.192.1
  
   linux:~ # ifconfig eth0; ifconfig ppp0
   eth0Link encap:Ethernet  HWaddr 00:50:8D:62:7C:5B
   inet addr:172.25.212.201  Bcast:255.255.255.255
 Mask:255.255.224.0
   inet6 addr: fe80::250:8dff:fe62:7c5b/64 Scope:Link
   UP BROADCAST NOTRAILERS RUNNING MULTICASTMTU:1500  Metric:1
   RX packets:253799 errors:0 dropped:0 overruns:0 frame:0
   TX packets:14236 errors:0 dropped:0 overruns:0 carrier:0
   collisions:0 txqueuelen:1000
   RX bytes:20082953 (19.1 Mb)TX bytes:2413726 (2.3 Mb)
   Interrupt:10 Base address:0xb000
  
   ppp0Link encap:Point-to-Point Protocol
   inet addr:85.64.160.99  P-t-P:172.26.255.17 
 Mask:255.255.255.255
   UP POINTOPOINT RUNNING NOARP MULTICASTMTU:1000  Metric:1
   RX packets:203 errors:0 dropped:0 overruns:0 frame:0
   TX packets:153 errors:0 dropped:0 overruns:0 carrier:0
   collisions:0 txqueuelen:3
   RX bytes:14557 (14.2 Kb)  TX bytes:13792 (13.4 Kb)
  
   linux:~ # route -n
   Kernel IP routing table
   Destination   Gateway Genmask Flags Metric RefUse
 Iface
   172.26.255.17 172.25.192.1255.255.255.255 UGH   0  00
 eth0
   172.26.255.17 0.0.0.0   255.255.255.255 UH0  00
 ppp0
   172.25.192.0  0.0.0.0 255.255.224.0   U 0  00
 eth0
   169.254.0.0   0.0.0.0 255.255.0.0 U 0  00
 eth0
   127.0.0.0 0.0.0.0 255.0.0.0 U 0  00 lo
   0.0.0.0   172.26.255.17   0.0.0.0 UG0  00
 ppp0
  
   Note: The first route (172.26.255.17/32 via 172.25.192.1) was manualy
   added before running pppd/pptp.
  
   linux:~ # cat /var/lib/dhcpcd/dhcpcd-eth0.info
   IPADDR=172.25.212.201
   NETMASK=255.255.224.0
   NETWORK=172.25.192.0
   BROADCAST=255.255.255.255
   GATEWAY=172.25.192.1
   HOSTNAME='linux'
   DNS=192.168.101.102,192.168.101.101
   DHCPSID=213.57.35.2
   DHCPGIADDR=172.25.160.1
   DHCPSIADDR=0.0.0.0
   DHCPCHADDR=00:50:8D:62:7C:5B
   DHCPSHADDR=00:05:00:E7:CD:A6
   DHCPSNAME=''
   LEASETIME=426054
   RENEWALTIME=213027
   REBINDTIME=372797
   INTERFACE='eth0'
   CLASSID='Linux 2.6.8-24-default i686'
   CLIENTID=00:50:8D:62:7C:5B
  
   linux:~ # rpm -q pptp
   pptp-1.5.0-2
  
   linux:~ # cat /etc/ppp/peers/barak
   linkname barak
   noauth
   user SOME_USERNAME
   pty /usr/sbin/pptp 172.26.255.17 --nolaunchpppd
   lock
   noipdefault
   usepeerdns
   nobsdcomp
   nodeflate
   lcp-echo-failure 10
   lcp-echo-interval 20
   defaultroute
   #mtu 1460
   mtu 1000
   mru 1000
   debug
  
  
   In /var/log/messages I see lots of pptp's messages like these:
   Aug 17 20:41:52 linux pptp[10933]: anon
   log[decaps_gre:pptp_gre.c:404]: buffering packet 4098 (expecting 4082,
   lost or reordered)
   Aug 17 20:41:52

Re: Connect to barak 013 through cable (PPTP)

[Ilia K.]

Your story is very useful. As I've answered to Guy, I still can't be
sure, that the problem is in cable, but I'll keep in mind this case
(and the way, they take care of it).
I think, next time I'll be asked to help someone to connect to the
internet, I'll prefer ADSL to cables.

On 8/19/05, Tzahi Fadida [EMAIL PROTECTED] wrote:
 Ready yourself for a long process.
 The first, second and maybe third guys are usually contract techies who
 usually replace your modem or check wall connections and maybe change in
 house
 cables. Only after they can't solve your problem the expert guy will
 come
 to check your stairs and building connections. If that doesn't work they
 will call
 another expert guy with permission to check your building juncture.
 For me it took about 2 month to get it right, however I also had
 disconnection
 to back me up. Yours might be more subtle and harder to make them fix.
 In addition, you have to rule out any software or ISP problems. How you
 are going
 to do that I don't know. Perhaps you should keep a closer eye on the
 cable modem
 3rd led and see if sometimes it blinks (i.e. disconnection of the modem)
 and if so,
 its definitely a cable problem.
 Also, note what time of day is worse and what is the weather outside.
 broken lines
 and problematic connectors are often susceptible to temperature changes.
 You need to understand that even the most sophisticated equipment they
 have sometimes
 can't find your problem as was in my case where they found a cable
 underground 
 1 meter before the main juncture which was glued with duck tape! :) 
 Hope, u'll solve your problem.
 
 Regards,
   tzahi.
 
  -Original Message-
  From: [EMAIL PROTECTED] 
  [mailto:[EMAIL PROTECTED] On Behalf Of Ilia K.
  Sent: Friday, August 19, 2005 1:45 AM
  To: Noam Meltzer
  Cc: Linux-IL
  Subject: Re: Connect to barak 013 through cable (PPTP)
  
  
  Hi!
  
  On 8/18/05, Noam Meltzer [EMAIL PROTECTED] wrote:
   Hi,
   One year ago in my previous apartment I had connection 
  problems to the 
   cables. After lot of misery I have found that the problem had 
   consisted of two
   elements:
   
   1. The cable connection is not very stable. In their underlying 
   protocol they lose many packets, resulting sometimes in 
  losing *your* 
   packets. I had cable technicians to my place at least 3 
  times, messing 
   around with the signals strength of the Rx and Tx until they got to 
   something which is similar to a stabler connection.
  
  What did you tell to cable company, so they come? Did it cost 
  you money? Is my case, half of the packets are lost even when 
  traffic is very low (only one ssh connection, sometimes also 
  ping test). How can I be sure, that the problem is not 
  inside the computer?
  
  
   2. The PPP configuration I had was too sensitive. I had to 
  change the 
   LCP configuration to do more LCP requests before declaring the 
   connection as dead (I think 5 lost packets in sequence) and to 
   enlarge the time interval between every LCP packet (to 30 sec. if i 
   recall correctly).
  The main problem is connection speed (packets loosing), 
  rather than unexpected termination. But I'll check lcp-echo-* 
  options too.
  
   
   After these two elements were fixed, I did not suffer from any 
   noticeable disruptions in my connection.
   
   Not sure that this is your problem, but maybe it would help...
  
  Probably, p.1 will help me!
  
   Noam
   
   On 8/17/05, Ilia K. [EMAIL PROTECTED] wrote:
Hi, All!

I've convinced one girl to install linux instead of 
  winXP. She is in 
Jerusalem and I am in Haifa, but we managed to install 
  SuSe 9.2 and 
connect it to barak's cable internet through PPTP. Now I 
  can ssh to 
the computer. Unfortunately, the connection is _very_ 
  unstable and 
slow. Half of the packets are seems to be lost. I suppose, the 
problem can be somewhere in routing (this is my weak 
  point, so the 
probability for the mistakes is high). I very want, that 
  this girl 
will stay with linux, but internet connection seems to be the 
bottle neck. I'll appreciate any feedback, especially 
  helpful one 
:)

PPTP is set up by executing:
linux:~# pppd call barak

The settings (after PPTP is up) are:

modem's address (in modem-eth0 network) seems to be 172.25.192.1

linux:~ # ifconfig eth0; ifconfig ppp0
eth0  Link encap:Ethernet  HWaddr 00:50:8D:62:7C:5B
  inet addr:172.25.212.201  Bcast:255.255.255.255
   Mask:255.255.2240
  inet6 addr: fe80::250:8dff:fe62:7c5b/64 Scope:Link
  UP BROADCAST NOTRAILERS RUNNING MULTICAST  
  MTU:1500  Metric:1
  RX packets:253799 errors:0 dropped:0 overruns:0 frame:0
  TX packets:14236 errors:0 dropped:0 overruns:0 carrier:0
  collisions:0 txqueuelen:1000
  RX bytes:20082953 (19.1 Mb)  TX bytes:2413726 (2.3 Mb)
  Interrupt:10 Base

Re: Connect to barak 013 through cable (PPTP)

[Ilia K.]

Hi, Geoffrey!

On 8/19/05, Geoffrey S. Mendelson wrote:
 On Fri, Aug 19, 2005 at 01:30:18PM +0300, Ilia K. wrote:
  Hi, Geoffrey and thanks for your post!
  
  I understand, that 212.143.205.162 is your PPTP server address. But
  what is 212.143.205.253?  Which of these IP's are you dialing with
  pppd/pptp?
 
 From my /etc/hosts:
 
   212.143.205.253 cable.netvision.net.il

 It's netvision's pptp host. 

Oops. So what is 212.143.205.162?

 I'm also enclosing my cablestart script with annotations. It came from
 the linux list a long time ago and grew, :-) 

Looking for my problem's solution I've seen several scripts and
several HOWTOs, but I couldn't know, which of them works and which is
outdated. In addition, there are no description for configuring SuSe
in these HOWTOs, so I'll go your way, which IMHO is a hack and should
better be implemented through ip-up.d / ip-down.d scripts...

Regarding your script:
1. All route commands is very useful for me. I'll try to proceed with
connection in the similar way you do and see, whether it will solve my
problem.
2. Why do you need nice --19 for pptp?
3. You don't really need to restart ntpd. It can be started during
system boot. It tries to connect to the external NTP server(s)
periodically and makes no harm if the server can't be reached.

The girl, who owns a computer is out for the weekend, so I'll be able
to test your settings only on Sunday.
Thank you.

 N.B. I've written hundreds of thousands of lines of code in other
 languanges
 (PL/I and REXX) so my shell scripts, PERL and C code follow the standard
 I am used to (and no-one else). 
 
   #!/bin/bash
 
   USERNAME=xx
 
   CPS=`ps -ax | grep cablestart | grep -cv grep `
 
   if [ 2 == $CPS ] ; then
   echo ok starting cable connection
   else
   echo cablestart is already running.
   exit
   fi
 
 This prevent multiple cablestarts from running for example if I start
 it manualy and cron also starts it.
 
 
   PS=`ps -ax | grep dhclient | grep eth1 | awk {print \\$1}`
 
   if [   !=  $PS ] ; then
   echo killing dhcp client for eth1 psid  $PS
   kill -KILL $PS
   else
   echo no dhcp client running for eth1 (good)
   fi
 
   PS=`ps -ax | grep cable.netvision.net.il | grep pptp-linux | awk {print
 \\$1}`
 
   if [   !=  $PS ] ; then
   echo killing pptp client for netvision psid  $PS
   kill -KILL $PS
   else
   echo no pptp client running for netvision (good)
   fi
 
 Ok, so I've cleaned up any old processes and clients left that
 refused to die on their own.
 
   echo shutting down eth1
   /sbin/ifdown eth1
   echo restarting eth1
   /sbin/ifup eth1
 
   INET=`ifconfig eth1 | grep inet | grep addr | awk {print \\$2} | awk 
 -F:
 {print \\$2}`
 
   echo IP address from cable modem  is  $INET
 
 I should have a check for 192.x.x.x address here, as that means the
 cable modem is up, but the connection to the head end is not.
 
   if [   ==  $INET ] ; then
   PS=`ps -ax | grep dhclient | grep eth1 | awk {print \\$1}`
   if [   !=  $PS ] ; then
   echo killing dhcp client for eth1 psid  $PS
   kill -KILL $PS
   echo giving up
   exit 1
   fi  
   fi
 
 Didn't get an IP address, cleaning up, and dying so that cron can restart
 it.
 
 
   /sbin/route add -host cable.netvision.net.il  dev eth1
 
 Pretty critical.
 
   echo starting pptp
   /bin/nice --19 /usr/sbin/pptp-linux cable.netvision.net.il line split
   user $USERNAME  debug mtu 1452 mru 1452 noauth
 
 N.B. note that is a nice level of -19 (the highest priority) not 19 (the
 lowest)
 
   echo waiting 30 seconds
   sleep 30
 
 I could never get a good way to tell, so I just wait. It has taken more
 than
 15 seconds at times, but never more than 30.
 
   NEWGW=$(/sbin/ifconfig ppp0 | grep inet | cut -d: -f3 |  line split
   tail -1 | cut -d  -f1)
   echo deleting old default route
   /sbin/route del default dev eth1
 
   if [  $NEWGW ==   ] ; then
   echo PPTP connection to netvision failed. Giving up.
   exit
   fi
 If it worked, I have a ppp0 and it has an IP address, If not, go away
 and try again later.
 
 
   echo seting default route to $NEWGW
   /sbin/route add default gw $NEWGW
 
 That's the most critical part of it.
 
   echo starting firewall
   /etc/rc.d/init.d/fwstart
 
 Just a shell script to set up an ip filter. Includes blocking some ports
 to the outside world and clamping mss.
 
 
   NEWLOCAL=$(/sbin/ifconfig ppp0 | grep inet | line split
   cut -d: -f2 | tail -1 | cut -d  -f1)
  
   echo Please wait while I resync the clock
   /etc/rc.d/init.d/ntpd stop

Re: Connect to barak 013 through cable (PPTP)

[Offer Kaye]

On 8/17/05, Ilia K. [EMAIL PROTECTED] wrote:

 I very want, that this girl will stay with linux, but internet
 connection seems to be the bottle neck. I'll appreciate any
 feedback, especially helpful one :)
 
 PPTP is set up by executing:
[...snip...]

I don't know a thing about routing, instead I will recommend the
program I use- pptpconfig, the graphical frontend to the pptpclient[1]
program. Setting up pptpconfig and getting connected to the Net using
pptpconfig was a breeze, even without knowing anything about PPTP and
routing...
There are suse-9.2 specific instructions on the site for getting
pptpclient installed and running[2]. Make sure you un-check the
Require MPPE checkbox in the Encryption tab, it is on by default but
I assume the connection uses a user+passwd instead, like mine does
(I'm connected through Netvision).

BTW, let me recommend that instead of Suse, you let your friend try
out Mepis. It's a live-CD (so she can try it before installing it on
her HD), it's *so* easy to install that it's not even funny, it's
newbie-friendly and its Debian based so it uses apt-get ;) Most
importantly for us, it comes with pptpconfig pre-installed and working
out-of-the-box.

[1] http://pptpclient.sourceforge.net/
[2] http://pptpclient.sourceforge.net/howto-suse-92.phtml

BTW I also live in Haifa so if you want a copy of Mepis instead of
downloading it yourself, feel free to ping me.

Regards,
-- 
Offer Kaye

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: Connect to barak 013 through cable (PPTP)

[Noam Meltzer]

Hi,
One year ago in my previous apartment I had connection problems to the cables.
After lot of misery I have found that the problem had consisted of two elements:

1. The cable connection is not very stable. In their underlying
protocol they lose many packets, resulting sometimes in losing *your*
packets.
I had cable technicians to my place at least 3 times, messing around
with the signals strength of the Rx and Tx until they got to something
which is similar to a stabler connection.

2. The PPP configuration I had was too sensitive. I had to change the
LCP configuration to do more LCP requests before declaring the
connection as dead (I think 5 lost packets in sequence) and to
enlarge the time interval between every LCP packet (to 30 sec. if i
recall correctly).

After these two elements were fixed, I did not suffer from any
noticeable disruptions in my connection.

Not sure that this is your problem, but maybe it would help...
Noam

On 8/17/05, Ilia K. [EMAIL PROTECTED] wrote:
 Hi, All!
 
 I've convinced one girl to install linux instead of winXP. She is in
 Jerusalem and I am in Haifa, but we managed to install SuSe 9.2 and
 connect it to barak's cable internet through PPTP. Now I can ssh to
 the computer. Unfortunately, the connection is _very_ unstable and
 slow. Half of the packets are seems to be lost. I suppose, the problem
 can be somewhere in routing (this is my weak point, so the
 probability for the mistakes is high).
 I very want, that this girl will stay with linux, but internet
 connection seems to be the bottle neck. I'll appreciate any
 feedback, especially helpful one :)
 
 PPTP is set up by executing:
 linux:~# pppd call barak
 
 The settings (after PPTP is up) are:
 
 modem's address (in modem-eth0 network) seems to be 172.25.192.1
 
 linux:~ # ifconfig eth0; ifconfig ppp0
 eth0  Link encap:Ethernet  HWaddr 00:50:8D:62:7C:5B
   inet addr:172.25.212.201  Bcast:255.255.255.255  Mask:255.255.2240
   inet6 addr: fe80::250:8dff:fe62:7c5b/64 Scope:Link
   UP BROADCAST NOTRAILERS RUNNING MULTICAST  MTU:1500  Metric:1
   RX packets:253799 errors:0 dropped:0 overruns:0 frame:0
   TX packets:14236 errors:0 dropped:0 overruns:0 carrier:0
   collisions:0 txqueuelen:1000
   RX bytes:20082953 (19.1 Mb)  TX bytes:2413726 (2.3 Mb)
   Interrupt:10 Base address:0xb000
 
 ppp0  Link encap:Point-to-Point Protocol
   inet addr:85.64.160.99  P-t-P:172.26.255.17  Mask:255.255.255.255
   UP POINTOPOINT RUNNING NOARP MULTICAST  MTU:1000  Metric:1
   RX packets:203 errors:0 dropped:0 overruns:0 frame:0
   TX packets:153 errors:0 dropped:0 overruns:0 carrier:0
   collisions:0 txqueuelen:3
   RX bytes:14557 (14.2 Kb)  TX bytes:13792 (13.4 Kb)
 
 linux:~ # route -n
 Kernel IP routing table
 Destination Gateway Genmask Flags Metric RefUse Iface
 172.26.255.17   172.25.192.1255.255.255.255 UGH   0  00 eth0
 172.26.255.17   0.0.0.0 255.255.255.255 UH0  00 ppp0
 172.25.192.00.0.0.0 255.255.224.0   U 0  00 eth0
 169.254.0.0 0.0.0.0 255.255.0.0 U 0  00 eth0
 127.0.0.0   0.0.0.0 255.0.0.0   U 0  00 lo
 0.0.0.0 172.26.255.17   0.0.0.0 UG0  00 ppp0
 
 Note: The first route (172.26.255.17/32 via 172.25.192.1) was manualy
 added before running pppd/pptp.
 
 linux:~ # cat /var/lib/dhcpcd/dhcpcd-eth0.info
 IPADDR=172.25.212.201
 NETMASK=255.255.224.0
 NETWORK=172.25.192.0
 BROADCAST=255.255.255.255
 GATEWAY=172.25.192.1
 HOSTNAME='linux'
 DNS=192.168.101.102,192.168.101.101
 DHCPSID=213.57.35.2
 DHCPGIADDR=172.25.160.1
 DHCPSIADDR=0.0.0.0
 DHCPCHADDR=00:50:8D:62:7C:5B
 DHCPSHADDR=00:05:00:E7:CD:A6
 DHCPSNAME=''
 LEASETIME=426054
 RENEWALTIME=213027
 REBINDTIME=372797
 INTERFACE='eth0'
 CLASSID='Linux 2.6.8-24-default i686'
 CLIENTID=00:50:8D:62:7C:5B
 
 linux:~ # rpm -q pptp
 pptp-1.5.0-2
 
 linux:~ # cat /etc/ppp/peers/barak
 linkname barak
 noauth
 user SOME_USERNAME
 pty /usr/sbin/pptp 172.26.255.17 --nolaunchpppd
 lock
 noipdefault
 usepeerdns
 nobsdcomp
 nodeflate
 lcp-echo-failure 10
 lcp-echo-interval 20
 defaultroute
 #mtu 1460
 mtu 1000
 mru 1000
 debug
 
 
 In /var/log/messages I see lots of pptp's messages like these:
 Aug 17 20:41:52 linux pptp[10933]: anon
 log[decaps_gre:pptp_gre.c:404]: buffering packet 4098 (expecting 4082,
 lost or reordered)
 Aug 17 20:41:52 linux pptp[10933]: anon
 log[decaps_gre:pptp_gre.c:404]: buffering packet 4099 (expecting 4082,
 lost or reordered)
 Aug 17 20:41:52 linux pptp[10933]: anon
 log[decaps_gre:pptp_gre.c:404]: buffering packet 4100 (expecting 4082,
 lost or reordered)
 Aug 17 20:41:52 linux pptp[10933]: anon
 log[decaps_gre:pptp_gre.c:404]: buffering packet 4101 (expecting 4082,
 lost or reordered)
 Aug 17 20:41:52 linux pptp[10933]: anon
 log[decaps_gre:pptp_gre.c:404

Re: Connect to barak 013 through cable (PPTP)

[Aaron]

Hey this is exactly my problem, I called the cable company and they couldn't 
help
so I suffer from my connection getting bumped a number of times a day.

so how do I change the LCP configuration? 

I am using debian sarge.

Thanks
Aaron
On Thu, Aug 18, 2005 at 09:21:22AM +0300 or thereabouts, Noam Meltzer wrote:
 Hi,
 One year ago in my previous apartment I had connection problems to the cables.
 After lot of misery I have found that the problem had consisted of two 
 elements:
 
 1. The cable connection is not very stable. In their underlying
 protocol they lose many packets, resulting sometimes in losing *your*
 packets.
 I had cable technicians to my place at least 3 times, messing around
 with the signals strength of the Rx and Tx until they got to something
 which is similar to a stabler connection.
 
 2. The PPP configuration I had was too sensitive. I had to change the
 LCP configuration to do more LCP requests before declaring the
 connection as dead (I think 5 lost packets in sequence) and to
 enlarge the time interval between every LCP packet (to 30 sec. if i
 recall correctly).
 
 After these two elements were fixed, I did not suffer from any
 noticeable disruptions in my connection.
 
 Not sure that this is your problem, but maybe it would help...
 Noam
 
 On 8/17/05, Ilia K. [EMAIL PROTECTED] wrote:
  Hi, All!
  
  I've convinced one girl to install linux instead of winXP. She is in
  Jerusalem and I am in Haifa, but we managed to install SuSe 9.2 and
  connect it to barak's cable internet through PPTP. Now I can ssh to
  the computer. Unfortunately, the connection is _very_ unstable and
  slow. Half of the packets are seems to be lost. I suppose, the problem
  can be somewhere in routing (this is my weak point, so the
  probability for the mistakes is high).
  I very want, that this girl will stay with linux, but internet
  connection seems to be the bottle neck. I'll appreciate any
  feedback, especially helpful one :)
  
  PPTP is set up by executing:
  linux:~# pppd call barak
  
  The settings (after PPTP is up) are:
  
  modem's address (in modem-eth0 network) seems to be 172.25.192.1
  
  linux:~ # ifconfig eth0; ifconfig ppp0
  eth0  Link encap:Ethernet  HWaddr 00:50:8D:62:7C:5B
inet addr:172.25.212.201  Bcast:255.255.255.255  Mask:255.255.2240
inet6 addr: fe80::250:8dff:fe62:7c5b/64 Scope:Link
UP BROADCAST NOTRAILERS RUNNING MULTICAST  MTU:1500  Metric:1
RX packets:253799 errors:0 dropped:0 overruns:0 frame:0
TX packets:14236 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:20082953 (19.1 Mb)  TX bytes:2413726 (2.3 Mb)
Interrupt:10 Base address:0xb000
  
  ppp0  Link encap:Point-to-Point Protocol
inet addr:85.64.160.99  P-t-P:172.26.255.17  Mask:255.255.255.255
UP POINTOPOINT RUNNING NOARP MULTICAST  MTU:1000  Metric:1
RX packets:203 errors:0 dropped:0 overruns:0 frame:0
TX packets:153 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:3
RX bytes:14557 (14.2 Kb)  TX bytes:13792 (13.4 Kb)
  
  linux:~ # route -n
  Kernel IP routing table
  Destination Gateway Genmask Flags Metric RefUse 
  Iface
  172.26.255.17   172.25.192.1255.255.255.255 UGH   0  00 eth0
  172.26.255.17   0.0.0.0 255.255.255.255 UH0  00 ppp0
  172.25.192.00.0.0.0 255.255.224.0   U 0  00 eth0
  169.254.0.0 0.0.0.0 255.255.0.0 U 0  00 eth0
  127.0.0.0   0.0.0.0 255.0.0.0   U 0  00 lo
  0.0.0.0 172.26.255.17   0.0.0.0 UG0  00 ppp0
  
  Note: The first route (172.26.255.17/32 via 172.25.192.1) was manualy
  added before running pppd/pptp.
  
  linux:~ # cat /var/lib/dhcpcd/dhcpcd-eth0.info
  IPADDR=172.25.212.201
  NETMASK=255.255.224.0
  NETWORK=172.25.192.0
  BROADCAST=255.255.255.255
  GATEWAY=172.25.192.1
  HOSTNAME='linux'
  DNS=192.168.101.102,192.168.101.101
  DHCPSID=213.57.35.2
  DHCPGIADDR=172.25.160.1
  DHCPSIADDR=0.0.0.0
  DHCPCHADDR=00:50:8D:62:7C:5B
  DHCPSHADDR=00:05:00:E7:CD:A6
  DHCPSNAME=''
  LEASETIME=426054
  RENEWALTIME=213027
  REBINDTIME=372797
  INTERFACE='eth0'
  CLASSID='Linux 2.6.8-24-default i686'
  CLIENTID=00:50:8D:62:7C:5B
  
  linux:~ # rpm -q pptp
  pptp-1.5.0-2
  
  linux:~ # cat /etc/ppp/peers/barak
  linkname barak
  noauth
  user SOME_USERNAME
  pty /usr/sbin/pptp 172.26.255.17 --nolaunchpppd
  lock
  noipdefault
  usepeerdns
  nobsdcomp
  nodeflate
  lcp-echo-failure 10
  lcp-echo-interval 20
  defaultroute
  #mtu 1460
  mtu 1000
  mru 1000
  debug
  
  
  In /var/log/messages I see lots of pptp's messages like these:
  Aug 17 20:41:52 linux pptp[10933]: anon
  log[decaps_gre:pptp_gre.c:404]: buffering packet 4098 (expecting 4082,
  lost or reordered)
  Aug 17 20:41:52 linux pptp

Re: Connect to barak 013 through cable (PPTP)

[Ilia K.]

Hi!

On 8/18/05, Noam Meltzer [EMAIL PROTECTED] wrote:
 Hi,
 One year ago in my previous apartment I had connection problems to the
 cables.
 After lot of misery I have found that the problem had consisted of two
 elements:
 
 1. The cable connection is not very stable. In their underlying
 protocol they lose many packets, resulting sometimes in losing *your*
 packets.
 I had cable technicians to my place at least 3 times, messing around
 with the signals strength of the Rx and Tx until they got to something
 which is similar to a stabler connection.

What did you tell to cable company, so they come? Did it cost you money?
Is my case, half of the packets are lost even when traffic is very low
(only one ssh connection, sometimes also ping test). How can I be
sure, that the problem is not inside the computer?


 2. The PPP configuration I had was too sensitive. I had to change the
 LCP configuration to do more LCP requests before declaring the
 connection as dead (I think 5 lost packets in sequence) and to
 enlarge the time interval between every LCP packet (to 30 sec. if i
 recall correctly).
The main problem is connection speed (packets loosing), rather than
unexpected termination. But I'll check lcp-echo-* options too.

 
 After these two elements were fixed, I did not suffer from any
 noticeable disruptions in my connection.
 
 Not sure that this is your problem, but maybe it would help...

Probably, p.1 will help me!

 Noam
 
 On 8/17/05, Ilia K. [EMAIL PROTECTED] wrote:
  Hi, All!
  
  I've convinced one girl to install linux instead of winXP. She is in
  Jerusalem and I am in Haifa, but we managed to install SuSe 9.2 and
  connect it to barak's cable internet through PPTP. Now I can ssh to
  the computer. Unfortunately, the connection is _very_ unstable and
  slow. Half of the packets are seems to be lost. I suppose, the problem
  can be somewhere in routing (this is my weak point, so the
  probability for the mistakes is high).
  I very want, that this girl will stay with linux, but internet
  connection seems to be the bottle neck. I'll appreciate any
  feedback, especially helpful one :)
  
  PPTP is set up by executing:
  linux:~# pppd call barak
  
  The settings (after PPTP is up) are:
  
  modem's address (in modem-eth0 network) seems to be 172.25.192.1
  
  linux:~ # ifconfig eth0; ifconfig ppp0
  eth0  Link encap:Ethernet  HWaddr 00:50:8D:62:7C:5B
inet addr:172.25.212.201  Bcast:255.255.255.255 
 Mask:255.255.2240
inet6 addr: fe80::250:8dff:fe62:7c5b/64 Scope:Link
UP BROADCAST NOTRAILERS RUNNING MULTICAST  MTU:1500  Metric:1
RX packets:253799 errors:0 dropped:0 overruns:0 frame:0
TX packets:14236 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:20082953 (19.1 Mb)  TX bytes:2413726 (2.3 Mb)
Interrupt:10 Base address:0xb000
  
  ppp0  Link encap:Point-to-Point Protocol
inet addr:85.64.160.99  P-t-P:172.26.255.17 
 Mask:255.255.255.255
UP POINTOPOINT RUNNING NOARP MULTICAST  MTU:1000  Metric:1
RX packets:203 errors:0 dropped:0 overruns:0 frame:0
TX packets:153 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:3
RX bytes:14557 (14.2 Kb)  TX bytes:13792 (13.4 Kb)
  
  linux:~ # route -n
  Kernel IP routing table
  Destination Gateway Genmask Flags Metric RefUse
 Iface
  172.26.255.17   172.25.192.1255.255.255.255 UGH   0  00
 eth0
  172.26.255.17   0.0.0.0 255.255.255.255 UH0  00
 ppp0
  172.25.192.00.0.0.0 255.255.224.0   U 0  00
 eth0
  169.254.0.0 0.0.0.0 255.255.0.0 U 0  00
 eth0
  127.0.0.0   0.0.0.0 255.0.0.0   U 0  00
 lo
  0.0.0.0 172.26.255.17   0.0.0.0 UG0  00
 ppp0
  
  Note: The first route (172.26.255.17/32 via 172.25.192.1) was manualy
  added before running pppd/pptp.
  
  linux:~ # cat /var/lib/dhcpcd/dhcpcd-eth0.info
  IPADDR=172.25.212.201
  NETMASK=255.255.224.0
  NETWORK=172.25.192.0
  BROADCAST=255.255.255.255
  GATEWAY=172.25.192.1
  HOSTNAME='linux'
  DNS=192.168.101.102,192.168.101.101
  DHCPSID=213.57.35.2
  DHCPGIADDR=172.25.160.1
  DHCPSIADDR=0.0.0.0
  DHCPCHADDR=00:50:8D:62:7C:5B
  DHCPSHADDR=00:05:00:E7:CD:A6
  DHCPSNAME=''
  LEASETIME=426054
  RENEWALTIME=213027
  REBINDTIME=372797
  INTERFACE='eth0'
  CLASSID='Linux 2.6.8-24-default i686'
  CLIENTID=00:50:8D:62:7C:5B
  
  linux:~ # rpm -q pptp
  pptp-1.5.0-2
  
  linux:~ # cat /etc/ppp/peers/barak
  linkname barak
  noauth
  user SOME_USERNAME
  pty /usr/sbin/pptp 172.26.255.17 --nolaunchpppd
  lock
  noipdefault
  usepeerdns
  nobsdcomp
  nodeflate
  lcp-echo-failure 10
  lcp-echo-interval 20
  defaultroute
  #mtu 1460
  mtu 1000
  mru 1000
  debug
  
  
  In /var/log/messages I see lots of pptp's messages like

Re: Connect to barak 013 through cable (PPTP)

[Ilia K.]

Hi All again!

Thanks to all guys, who have replied me!
However, nobody have told me yet, whether I've missed some point in
the configuration.
I want to emphasize the fact, that even under very low load (one ssh
connection only), _half_ of pings are lost when pinging the PPTP
server or some public servers.
Under no circumstances this can be treated as normal cable operation.

Please, someone who has working cable connection, send me output of:
ifconfig; route -n

This could help me decide, whether my configuration is OK or buggy
(and if it's OK, I'll apply to the cable company).

Thanks.

On 8/17/05, Ilia K. [EMAIL PROTECTED] wrote:
 Hi, All!
 
 I've convinced one girl to install linux instead of winXP. She is in
 Jerusalem and I am in Haifa, but we managed to install SuSe 9.2 and
 connect it to barak's cable internet through PPTP. Now I can ssh to
 the computer. Unfortunately, the connection is _very_ unstable and
 slow. Half of the packets are seems to be lost. I suppose, the problem
 can be somewhere in routing (this is my weak point, so the
 probability for the mistakes is high).
 I very want, that this girl will stay with linux, but internet
 connection seems to be the bottle neck. I'll appreciate any
 feedback, especially helpful one :)
 
 PPTP is set up by executing:
 linux:~# pppd call barak
 
 The settings (after PPTP is up) are:
 
 modem's address (in modem-eth0 network) seems to be 172.25.192.1
 
 linux:~ # ifconfig eth0; ifconfig ppp0
 eth0  Link encap:Ethernet  HWaddr 00:50:8D:62:7C:5B  
   inet addr:172.25.212.201  Bcast:255.255.255.255 Mask:255.255.224.0
   inet6 addr: fe80::250:8dff:fe62:7c5b/64 Scope:Link
   UP BROADCAST NOTRAILERS RUNNING MULTICAST  MTU:1500  Metric:1
   RX packets:253799 errors:0 dropped:0 overruns:0 frame:0
   TX packets:14236 errors:0 dropped:0 overruns:0 carrier:0
   collisions:0 txqueuelen:1000 
   RX bytes:20082953 (19.1 Mb)  TX bytes:2413726 (2.3 Mb)
   Interrupt:10 Base address:0xb000 
 
 ppp0  Link encap:Point-to-Point Protocol  
   inet addr:85.64.160.99  P-t-P:172.26.255.17  Mask:255.255.255.255
   UP POINTOPOINT RUNNING NOARP MULTICAST  MTU:1000  Metric:1
   RX packets:203 errors:0 dropped:0 overruns:0 frame:0
   TX packets:153 errors:0 dropped:0 overruns:0 carrier:0
   collisions:0 txqueuelen:3 
   RX bytes:14557 (14.2 Kb)  TX bytes:13792 (13.4 Kb)
 
 linux:~ # route -n
 Kernel IP routing table
 Destination Gateway Genmask Flags Metric RefUse Iface
 172.26.255.17   172.25.192.1255.255.255.255 UGH   0  00 eth0
 172.26.255.17   0.0.0.0 255.255.255.255 UH0  00 ppp0
 172.25.192.00.0.0.0 255.255.224.0   U 0  00 eth0
 169.254.0.0 0.0.0.0 255.255.0.0 U 0  00 eth0
 127.0.0.0   0.0.0.0 255.0.0.0   U 0  00 lo
 0.0.0.0 172.26.255.17   0.0.0.0 UG0  00 ppp0
 
 Note: The first route (172.26.255.17/32 via 172.25.192.1) was manualy
 added before running pppd/pptp.
 
 linux:~ # cat /var/lib/dhcpcd/dhcpcd-eth0.info
 IPADDR=172.25.212.201
 NETMASK=255.255.224.0
 NETWORK=172.25.192.0
 BROADCAST=255.255.255.255
 GATEWAY=172.25.192.1
 HOSTNAME='linux'
 DNS=192.168.101.102,192.168.101.101
 DHCPSID=213.57.35.2
 DHCPGIADDR=172.25.160.1
 DHCPSIADDR=0.0.0.0
 DHCPCHADDR=00:50:8D:62:7C:5B
 DHCPSHADDR=00:05:00:E7:CD:A6
 DHCPSNAME=''
 LEASETIME=426054
 RENEWALTIME=213027
 REBINDTIME=372797
 INTERFACE='eth0'
 CLASSID='Linux 2.6.8-24-default i686'
 CLIENTID=00:50:8D:62:7C:5B
 
 linux:~ # rpm -q pptp
 pptp-1.5.0-2
 
 linux:~ # cat /etc/ppp/peers/barak 
 linkname barak
 noauth
 user SOME_USERNAME
 pty /usr/sbin/pptp 172.26.255.17 --nolaunchpppd
 lock
 noipdefault
 usepeerdns
 nobsdcomp
 nodeflate
 lcp-echo-failure 10
 lcp-echo-interval 20
 defaultroute
 #mtu 1460
 mtu 1000
 mru 1000
 debug
 
 
 In /var/log/messages I see lots of pptp's messages like these:
 Aug 17 20:41:52 linux pptp[10933]: anon
 log[decaps_gre:pptp_gre.c:404]: buffering packet 4098 (expecting 4082,
 lost or reordered)
 Aug 17 20:41:52 linux pptp[10933]: anon
 log[decaps_gre:pptp_gre.c:404]: buffering packet 4099 (expecting 4082,
 lost or reordered)
 Aug 17 20:41:52 linux pptp[10933]: anon
 log[decaps_gre:pptp_gre.c:404]: buffering packet 4100 (expecting 4082,
 lost or reordered)
 Aug 17 20:41:52 linux pptp[10933]: anon
 log[decaps_gre:pptp_gre.c:404]: buffering packet 4101 (expecting 4082,
 lost or reordered)
 Aug 17 20:41:52 linux pptp[10933]: anon
 log[decaps_gre:pptp_gre.c:404]: buffering packet 4102 (expecting 4082,
 lost or reordered)
 Aug 17 20:42:27 linux pptp[10933]: anon
 log[decaps_gre:pptp_gre.c:395]: discarding duplicate or old packet
 4082 (expecting 4116)
 Aug 17 20:42:27 linux pptp[10933]: anon
 log[decaps_gre:pptp_gre.c:395]: discarding duplicate or old packet
 4083 (expecting 4116)
 Aug 17 20:42:27 linux

RE: Connect to barak 013 through cable (PPTP)

[Tzahi Fadida]

Ready yourself for a long process.
The first, second and maybe third guys are usually contract techies who
usually replace your modem or check wall connections and maybe change in
house
cables. Only after they can't solve your problem the expert guy will
come
to check your stairs and building connections. If that doesn't work they
will call
another expert guy with permission to check your building juncture.
For me it took about 2 month to get it right, however I also had
disconnection
to back me up. Yours might be more subtle and harder to make them fix.
In addition, you have to rule out any software or ISP problems. How you
are going
to do that I don't know. Perhaps you should keep a closer eye on the
cable modem
3rd led and see if sometimes it blinks (i.e. disconnection of the modem)
and if so,
its definitely a cable problem.
Also, note what time of day is worse and what is the weather outside.
broken lines
and problematic connectors are often susceptible to temperature changes.
You need to understand that even the most sophisticated equipment they
have sometimes
can't find your problem as was in my case where they found a cable
underground 
1 meter before the main juncture which was glued with duck tape! :) 
Hope, u'll solve your problem.

Regards,
tzahi.

 -Original Message-
 From: [EMAIL PROTECTED] 
 [mailto:[EMAIL PROTECTED] On Behalf Of Ilia K.
 Sent: Friday, August 19, 2005 1:45 AM
 To: Noam Meltzer
 Cc: Linux-IL
 Subject: Re: Connect to barak 013 through cable (PPTP)
 
 
 Hi!
 
 On 8/18/05, Noam Meltzer [EMAIL PROTECTED] wrote:
  Hi,
  One year ago in my previous apartment I had connection 
 problems to the 
  cables. After lot of misery I have found that the problem had 
  consisted of two
  elements:
  
  1. The cable connection is not very stable. In their underlying 
  protocol they lose many packets, resulting sometimes in 
 losing *your* 
  packets. I had cable technicians to my place at least 3 
 times, messing 
  around with the signals strength of the Rx and Tx until they got to 
  something which is similar to a stabler connection.
 
 What did you tell to cable company, so they come? Did it cost 
 you money? Is my case, half of the packets are lost even when 
 traffic is very low (only one ssh connection, sometimes also 
 ping test). How can I be sure, that the problem is not 
 inside the computer?
 
 
  2. The PPP configuration I had was too sensitive. I had to 
 change the 
  LCP configuration to do more LCP requests before declaring the 
  connection as dead (I think 5 lost packets in sequence) and to 
  enlarge the time interval between every LCP packet (to 30 sec. if i 
  recall correctly).
 The main problem is connection speed (packets loosing), 
 rather than unexpected termination. But I'll check lcp-echo-* 
 options too.
 
  
  After these two elements were fixed, I did not suffer from any 
  noticeable disruptions in my connection.
  
  Not sure that this is your problem, but maybe it would help...
 
 Probably, p.1 will help me!
 
  Noam
  
  On 8/17/05, Ilia K. [EMAIL PROTECTED] wrote:
   Hi, All!
   
   I've convinced one girl to install linux instead of 
 winXP. She is in 
   Jerusalem and I am in Haifa, but we managed to install 
 SuSe 9.2 and 
   connect it to barak's cable internet through PPTP. Now I 
 can ssh to 
   the computer. Unfortunately, the connection is _very_ 
 unstable and 
   slow. Half of the packets are seems to be lost. I suppose, the 
   problem can be somewhere in routing (this is my weak 
 point, so the 
   probability for the mistakes is high). I very want, that 
 this girl 
   will stay with linux, but internet connection seems to be the 
   bottle neck. I'll appreciate any feedback, especially 
 helpful one 
   :)
   
   PPTP is set up by executing:
   linux:~# pppd call barak
   
   The settings (after PPTP is up) are:
   
   modem's address (in modem-eth0 network) seems to be 172.25.192.1
   
   linux:~ # ifconfig eth0; ifconfig ppp0
   eth0  Link encap:Ethernet  HWaddr 00:50:8D:62:7C:5B
 inet addr:172.25.212.201  Bcast:255.255.255.255
  Mask:255.255.2240
 inet6 addr: fe80::250:8dff:fe62:7c5b/64 Scope:Link
 UP BROADCAST NOTRAILERS RUNNING MULTICAST  
 MTU:1500  Metric:1
 RX packets:253799 errors:0 dropped:0 overruns:0 frame:0
 TX packets:14236 errors:0 dropped:0 overruns:0 carrier:0
 collisions:0 txqueuelen:1000
 RX bytes:20082953 (19.1 Mb)  TX bytes:2413726 (2.3 Mb)
 Interrupt:10 Base address:0xb000
   
   ppp0  Link encap:Point-to-Point Protocol
 inet addr:85.64.160.99  P-t-P:172.26.255.17
  Mask:255.255.255.255
 UP POINTOPOINT RUNNING NOARP MULTICAST  
 MTU:1000  Metric:1
 RX packets:203 errors:0 dropped:0 overruns:0 frame:0
 TX packets:153 errors:0 dropped:0 overruns:0 carrier:0
 collisions:0 txqueuelen:3
 RX bytes:14557 (14.2 Kb)  TX bytes:13792 (13.4 Kb

Connect to barak 013 through cable (PPTP)

[Ilia K.]

Hi, All!

I've convinced one girl to install linux instead of winXP. She is in
Jerusalem and I am in Haifa, but we managed to install SuSe 9.2 and
connect it to barak's cable internet through PPTP. Now I can ssh to
the computer. Unfortunately, the connection is _very_ unstable and
slow. Half of the packets are seems to be lost. I suppose, the problem
can be somewhere in routing (this is my weak point, so the
probability for the mistakes is high).
I very want, that this girl will stay with linux, but internet
connection seems to be the bottle neck. I'll appreciate any
feedback, especially helpful one :)

PPTP is set up by executing:
linux:~# pppd call barak

The settings (after PPTP is up) are:

modem's address (in modem-eth0 network) seems to be 172.25.192.1

linux:~ # ifconfig eth0; ifconfig ppp0
eth0  Link encap:Ethernet  HWaddr 00:50:8D:62:7C:5B  
  inet addr:172.25.212.201  Bcast:255.255.255.255  Mask:255.255.2240
  inet6 addr: fe80::250:8dff:fe62:7c5b/64 Scope:Link
  UP BROADCAST NOTRAILERS RUNNING MULTICAST  MTU:1500  Metric:1
  RX packets:253799 errors:0 dropped:0 overruns:0 frame:0
  TX packets:14236 errors:0 dropped:0 overruns:0 carrier:0
  collisions:0 txqueuelen:1000 
  RX bytes:20082953 (19.1 Mb)  TX bytes:2413726 (2.3 Mb)
  Interrupt:10 Base address:0xb000 

ppp0  Link encap:Point-to-Point Protocol  
  inet addr:85.64.160.99  P-t-P:172.26.255.17  Mask:255.255.255.255
  UP POINTOPOINT RUNNING NOARP MULTICAST  MTU:1000  Metric:1
  RX packets:203 errors:0 dropped:0 overruns:0 frame:0
  TX packets:153 errors:0 dropped:0 overruns:0 carrier:0
  collisions:0 txqueuelen:3 
  RX bytes:14557 (14.2 Kb)  TX bytes:13792 (13.4 Kb)

linux:~ # route -n
Kernel IP routing table
Destination Gateway Genmask Flags Metric RefUse Iface
172.26.255.17   172.25.192.1255.255.255.255 UGH   0  00 eth0
172.26.255.17   0.0.0.0 255.255.255.255 UH0  00 ppp0
172.25.192.00.0.0.0 255.255.224.0   U 0  00 eth0
169.254.0.0 0.0.0.0 255.255.0.0 U 0  00 eth0
127.0.0.0   0.0.0.0 255.0.0.0   U 0  00 lo
0.0.0.0 172.26.255.17   0.0.0.0 UG0  00 ppp0

Note: The first route (172.26.255.17/32 via 172.25.192.1) was manualy
added before running pppd/pptp.

linux:~ # cat /var/lib/dhcpcd/dhcpcd-eth0.info
IPADDR=172.25.212.201
NETMASK=255.255.224.0
NETWORK=172.25.192.0
BROADCAST=255.255.255.255
GATEWAY=172.25.192.1
HOSTNAME='linux'
DNS=192.168.101.102,192.168.101.101
DHCPSID=213.57.35.2
DHCPGIADDR=172.25.160.1
DHCPSIADDR=0.0.0.0
DHCPCHADDR=00:50:8D:62:7C:5B
DHCPSHADDR=00:05:00:E7:CD:A6
DHCPSNAME=''
LEASETIME=426054
RENEWALTIME=213027
REBINDTIME=372797
INTERFACE='eth0'
CLASSID='Linux 2.6.8-24-default i686'
CLIENTID=00:50:8D:62:7C:5B

linux:~ # rpm -q pptp
pptp-1.5.0-2

linux:~ # cat /etc/ppp/peers/barak 
linkname barak
noauth
user SOME_USERNAME
pty /usr/sbin/pptp 172.26.255.17 --nolaunchpppd
lock
noipdefault
usepeerdns
nobsdcomp
nodeflate
lcp-echo-failure 10
lcp-echo-interval 20
defaultroute
#mtu 1460
mtu 1000
mru 1000
debug


In /var/log/messages I see lots of pptp's messages like these:
Aug 17 20:41:52 linux pptp[10933]: anon
log[decaps_gre:pptp_gre.c:404]: buffering packet 4098 (expecting 4082,
lost or reordered)
Aug 17 20:41:52 linux pptp[10933]: anon
log[decaps_gre:pptp_gre.c:404]: buffering packet 4099 (expecting 4082,
lost or reordered)
Aug 17 20:41:52 linux pptp[10933]: anon
log[decaps_gre:pptp_gre.c:404]: buffering packet 4100 (expecting 4082,
lost or reordered)
Aug 17 20:41:52 linux pptp[10933]: anon
log[decaps_gre:pptp_gre.c:404]: buffering packet 4101 (expecting 4082,
lost or reordered)
Aug 17 20:41:52 linux pptp[10933]: anon
log[decaps_gre:pptp_gre.c:404]: buffering packet 4102 (expecting 4082,
lost or reordered)
Aug 17 20:42:27 linux pptp[10933]: anon
log[decaps_gre:pptp_gre.c:395]: discarding duplicate or old packet
4082 (expecting 4116)
Aug 17 20:42:27 linux pptp[10933]: anon
log[decaps_gre:pptp_gre.c:395]: discarding duplicate or old packet
4083 (expecting 4116)
Aug 17 20:42:27 linux pptp[10933]: anon
log[decaps_gre:pptp_gre.c:395]: discarding duplicate or old packet
4084 (expecting 4116)
Aug 17 20:42:27 linux pptp[10933]: anon
log[decaps_gre:pptp_gre.c:395]: discarding duplicate or old packet
4085 (expecting 4116)

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: pptp over pppoe problem

[Shachar Shemesh]

Roberto S. Meyer wrote:


We setup ppp0 at 1492 (pppoe) and ppp1 at 1400 (pptp) and it
finally worked. Now I telnet to port 80 and receive the webpages
that didn't receive in the past.
 

Then the peer you do the pptp with is misconfigured. It was supposed to 
advertise the correct MTU.


On second thought, maybe most of the connections it gets are not through 
an ADSL modem, and thus do not need two size reduction. If that is the 
case, then manual configuration is, indeed, necessary here.


  Shachar

--
Shachar Shemesh
Lingnu Open Source Consulting ltd.
Have you backed up today's work? http://www.lingnu.com/backup.html


=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: pptp over pppoe problem

[Roberto S. Meyer]

Shachar Shemesh escribió/wrote/a écrit:

[snip]

 What do you mean by inner interface? ppp1 (pptp)?
 
 Which mtu/mru values do you suggest for eth0 (1500?), ppp0 
 (pppoe, 1492?) and ppp1 (pptp, 1300?)
 
 Thank you for your advice.
  
 
 If you're having trouble connecting from the actual machine doing the 
 dialing, then this is likely not the problem.

[snip]

Well... it's working! :-D

We setup ppp0 at 1492 (pppoe) and ppp1 at 1400 (pptp) and it
finally worked. Now I telnet to port 80 and receive the webpages
that didn't receive in the past.

I'm sure it wasn't our ISP who fixed the problem in the meantime,
because some days ago they called us to say they had given up with
it ;-)

Thanx a lot for your help.

-
Roberto

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: pptp over pppoe problem

[Shachar Shemesh]

Roberto S. Meyer wrote:


Muli Ben-Yehuda escribió/wrote/a écrit:

 


On Wed, Jun 01, 2005 at 12:42:34PM -0300, Roberto S. Meyer wrote:
   


Hi,

I'm having trouble that I think can be compared to this[1] FAQ,
but the link to the possible answer[2] is broken.
 


it's available at http://www.mulix.org/adsl-howto.txt. I'm amazed
anyone still reads it!

   


Could it be an mtu problem?
 


Certainly sounds that way. Just drop the MTU on the inner interface to
   



Great!


 


~1300 and see if that fixes it.
   



What do you mean by inner interface? ppp1 (pptp)?

Which mtu/mru values do you suggest for eth0 (1500?), ppp0 
(pppoe, 1492?) and ppp1 (pptp, 1300?)


Thank you for your advice.
 

If you're having trouble connecting from the actual machine doing the 
dialing, then this is likely not the problem.


Run ifconfig. What are the MTU values (when you perform no change) you 
get for all relevant interfaces (i.e. the ethernet, ppp0 and ppp1)?


 Shachar

--
Shachar Shemesh
Lingnu Open Source Consulting ltd.
Have you backed up today's work? http://www.lingnu.com/backup.html


=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

pptp over pppoe problem

[Roberto S. Meyer]

Hi,

I'm having trouble that I think can be compared to this[1] FAQ,
but the link to the possible answer[2] is broken.

I'll search for it, but also post it here, maybe someone had the
same problem or can post some thoughts about.

On a Debian Sarge computer, I setup ppp0 with pppoe. Once ppp0
was up I setup ppp1 with pptp through ppp0. Everything seemed 
fine until I had trouble getting some web pages.

It isn't a routing problem. I tried these webpages with lynx over
ppp0 without problems. When I setup ppp1 I didn't receive the
page I request from the server.

So, we tried telneting to port 80 and posting the HTTP commands
directly. Again the same problem. Over ppp0 everything worked
fine but over ppp1 (pptp) we connected, posted the commands and
didn't get the webpage.

Could it be an mtu problem?

TIA,

[1] http://iglu.org.il/faq/cache/158.html
[2] http://www.pointer.co.il/~mulix/adsl-howto.txt

-
Roberto
Mendoza - Argentina

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: pptp over pppoe problem

[Muli Ben-Yehuda]

On Wed, Jun 01, 2005 at 12:42:34PM -0300, Roberto S. Meyer wrote:
 Hi,
 
 I'm having trouble that I think can be compared to this[1] FAQ,
 but the link to the possible answer[2] is broken.

it's available at http://www.mulix.org/adsl-howto.txt. I'm amazed
anyone still reads it!

 Could it be an mtu problem?

Certainly sounds that way. Just drop the MTU on the inner interface to
~1300 and see if that fixes it.

Cheers,
Muli

-- 
Muli Ben-Yehuda
http://www.mulix.org | http://mulix.livejournal.com/


=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: pptp over pppoe problem

[Roberto S. Meyer]

Muli Ben-Yehuda escribió/wrote/a écrit:

 On Wed, Jun 01, 2005 at 12:42:34PM -0300, Roberto S. Meyer wrote:
  Hi,
  
  I'm having trouble that I think can be compared to this[1] FAQ,
  but the link to the possible answer[2] is broken.
 
 it's available at http://www.mulix.org/adsl-howto.txt. I'm amazed
 anyone still reads it!
 
  Could it be an mtu problem?
 
 Certainly sounds that way. Just drop the MTU on the inner interface to

Great!


 ~1300 and see if that fixes it.

What do you mean by inner interface? ppp1 (pptp)?

Which mtu/mru values do you suggest for eth0 (1500?), ppp0 
(pppoe, 1492?) and ppp1 (pptp, 1300?)

Thank you for your advice.

-
Roberto

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: 013 Barak PPTP on Linksys WRT54G wireless router doesnt work?

[solomon]

On Sunday 09 January 2005 08:45, Oron Peled wrote:
 On Sunday 09 January 2005 01:28, solomon wrote:
  Here's an improved version ...

 Few notes (I should have wrote them on your first post, but
 didn't have time then):

First of all, thanks for your post. I'm always willing to learn and will 
examine your version of the script. However, I would like to answer a few of 
the questions you raise about my version in self-defense ;-)


   - No need for multiple sed commands in your case. sed is
 capable for executing multiple commands on a single line
 using the '-e' flag (RTFM).
the -e flag is not necessary and in fact, if you look at my first sed line 
you'll see that I do have 3 commands on a single line

   - You seem to forget (or haven't learned) about regular
 expressions. Simply try to chop piece by piece which is tedious.
You're right. I do know about regular expressions, but I guess not enough.


   - There are many redirections to small files -- not good.
 A lot of garbage to clean or leave. Also raises some
Actually, most of the small files were for debugging (I wanted to see 
intermediate results) and I could have added the last 2 sed commands to the 3 
mentioned earlier. On the other hand, I don't see a problem with cleaning up 
since everything is written to /tmp.

 security concerns as these run as root (race conditions
 and symlink attacks come to mind).
True, but there's no sensitive information. Anyone who succeeds in reaching my 
machine already knows the IP address.

   - Why create a script and run it? Just run the command directly.
OK

   - What splitpea is doing that cannot be done with sed,grep,tr etc?
I played with sed and grep and didn't succeed. As said earlier, I'll study 
your version as a learning excercise


 So let's do it in a saner way (tested only the first significant line,
 you can test the rest for syntax errors etc.):
I'm not sure I understand what else has to be tested. I ran your script, as 
is, and it worked.



 #! /bin/sh

 # Get the other side IP
 other=`ifconfig | grep P-t-P: | \
sed -e 's/^.*P-t-P://' -e 's/ .*$//'`
 if [ $other =  ]; then
  echo 2 Something's wrong -- no P-t-P IP
  exit 1
 fi
 if ! route add -net default gw $other ppp0; then
  echo 2 Failed to add default route
  exit 1
 fi

 #  cut here ---

 For the shell challenged look no further than the 'abs'
 (Advanced Bash Scripting) guide at www.tldp.org.

-- 
Shlomo Solomon
http://come.to/shlomo.solomon
Sent by KMail 1.7.1 (KDE 3.2.3) on LINUX Mandrake 10.1



=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: 013 Barak PPTP on Linksys WRT54G wireless router doesnt work?

[solomon]

On Friday 07 January 2005 14:13, Oron Peled wrote:
 On Friday 07 January 2005 10:15, Danny Lieberman wrote:
  1. Altho I get a PPP address and teh DNS addresses of 013 - I CANT ping
  anything - get network unreachable

 Good. This error means you have problem in *your* routing
 table. You probably don't have a default route.

I had a similar problem with pptp on my new computer. Since no-one on the list 
was able to solve it, I wrote a script to re-set the default route. I still 
have no idea why the default route doesn't get set up automatically (as it 
did on my previous computer). In any case, maybe this script can solve Danny 
Lieberman's problem.

BTW - I'm sure there must be other or better ways to write the script, but it 
works, and I guess that's the important thing. All the commands used are 
standard except for splitpea which is a nice file spliiting utility you can 
get here: http://www.doxxx.net/splitpea.html


Here's the script::

#!/bin/bash
## create route command after running pptp

## get the IP addresses and remove extra text
ifconfig|grep  P-t-P|sed 's/inet addr://; s/Mask:255.255.255.255//; 
s/P-t-P://'  /tmp/tst-IP

## cut out the inet address so only the P-t-P address is left
/home/solomon/bin/splitpea -s --bytes=22 /tmp/tst-IP

## set an ENV variable
MYIP=`cat /tmp/tst-IP.002`

## create executable script
echo #!/bin/bash  /home/solomon/bin/myroute
echo route add -net default gw $MYIP ppp0  /home/solomon/bin/myroute
chmod +x /home/solomon/bin/myroute

## and run it
/home/solomon/bin/myroute



good luck


-- 
Shlomo Solomon
http://come.to/shlomo.solomon
Sent by KMail 1.7.1 (KDE 3.2.3) on LINUX Mandrake 10.1



=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: 013 Barak PPTP on Linksys WRT54G wireless router doesnt work?

[Ilya Konstantinov]

solomon wrote:
I still have no idea why the default route doesn't get set up automatically (as it 
did on my previous computer).

IIRC,  the 'defaultroute' option of pppd always came as a patch. Some 
distros integrate this patch while others prefer to implement it as a 
script in /etc/ppp/ip-up.d/ directory. Debian's approach uses the later, 
IMO.


=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: 013 Barak PPTP on Linksys WRT54G wireless router doesnt work?

[solomon]

On Sunday 09 January 2005 00:57, Ilya Konstantinov wrote:
 solomon wrote:
 I still have no idea why the default route doesn't get set up
  automatically (as it did on my previous computer).

 IIRC,  the 'defaultroute' option of pppd always came as a patch. Some
 distros integrate this patch while others prefer to implement it as a
 script in /etc/ppp/ip-up.d/ directory. Debian's approach uses the later,
 IMO.

You could be right, although:
1 - I use Mandrake and not Debian
2 - I did move from MDK10.0 to MDK10.1, but I doubt that this would have 
changed.
3 - I didn't find anything in /etc/ppp/ip-up.d/ directory (but, of course it 
could be somewhere else on Mandrake)
 

BTW - I noticed that in my previous post I sent an old version of my script to 
set the default route. Here's an improved version that solves the problem of 
the inet IP sometimes being a slightly different length, so I delete 
whitespace from the beginning and then pad extra whitespace before the P-t-P 
address, before splitting the file.



#!/bin/bash
## create route command after running pptp

## get the IP addresses and remove extra text
ifconfig | grep  P-t-P | sed 's/inet addr://; s/Mask:255.255.255.255//; 
s/P-t-P://'  /tmp/tst-IP

# delete leading whitespace (spaces, tabs) from front of the line
sed 's/^[ \t]*//' /tmp/tst-IP  /tmp/tst-IP1

#  add whitespace before IP address
sed 's/ 212./ 212./'  /tmp/tst-IP1  /tmp/tst-IP2

## cut out the inet address so only the P-t-P address is left
/home/solomon/bin/splitpea -s --bytes=20 /tmp/tst-IP2

## set an ENV variable
MYIP=`cat /tmp/tst-IP2.002`

## create executable script
echo #!/bin/bash  /home/solomon/bin/myroute
echo route add -net default gw $MYIP ppp0  /home/solomon/bin/myroute
chmod +x /home/solomon/bin/myroute

## and run it
/home/solomon/bin/myroute




-- 
Shlomo Solomon
http://come.to/shlomo.solomon
Sent by KMail 1.7.1 (KDE 3.2.3) on LINUX Mandrake 10.1



=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: 013 Barak PPTP on Linksys WRT54G wireless router doesnt work?

[Danny Lieberman]

Shaul Karl wrote:
On Thu, Jan 06, 2005 at 01:13:39PM +0200, Danny Lieberman wrote:
 

guys
I'm setting up a new account at home for 013 Barak over a cable modem 
and Linksys wrt54g

The 013 Barak PPTP doesnt work for my Linksys WRT54..(Works 
ok on a w2k box )  ;-( 

I setup the definitions with pns.barak.net.il and the box dials up and 
gets a PPP address and DNS addresses
but I get network unreachable - it feels like a network mask issue but 
I cant convince the Linksys to go the last mile

Ideas?
   


 Can you ping by IP address? What is in the logs?
Your message mentioned pns.barak.net.il. This is a typo, isn't it?
 

Shaul
1. Altho I get a PPP address and teh DNS addresses of 013 - I CANT ping 
anything - get network unreachable
2. pns.barak.net.il is the 013 cable modem PPTP server - in the Linksys 
you put in an IP addrss  172.26.255.17

danny

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: 013 Barak PPTP on Linksys WRT54G wireless router doesnt work?

[Oron Peled]

On Friday 07 January 2005 10:15, Danny Lieberman wrote:
 1. Altho I get a PPP address and teh DNS addresses of 013 - I CANT ping 
 anything - get network unreachable

Good. This error means you have problem in *your* routing
table. You probably don't have a default route.

-- 
Oron Peled Voice/Fax: +972-4-8228492
[EMAIL PROTECTED]  http://www.actcom.co.il/~oron
ICQ UIN: 16527398

Windows - How do you want to be exploited today.

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: 013 Barak PPTP on Linksys WRT54G wireless router doesnt work?

[David Harel]

Just for the sake of my understanding.
I also use cable technology and 013 Barak. I never had to do any dialing 
or any operation to have my communication started. Simple DHCP works 
just fine although I asked for what they call Permanent connection 
which is free of charge.
What is this dialing all about, what is pns.barak.net.il and what are 
the benefits of this PPTP technology?

Danny Lieberman wrote:
guys
I'm setting up a new account at home for 013 Barak over a cable modem 
and Linksys wrt54g

The 013 Barak PPTP doesnt work for my Linksys 
WRT54..(Works ok on a w2k box )  ;-(
I setup the definitions with pns.barak.net.il and the box dials up and 
gets a PPP address and DNS addresses
but I get network unreachable - it feels like a network mask issue but 
I cant convince the Linksys to go the last mile

Ideas?
thanks!
danny
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

--
Thanks.
David Harel,
==
Home office +972 4 6921986
Fax:+972 4 6921986
Cellular:   +972 54 4534502
Snail Mail: Amuka
   D.N Merom Hagalil
   13802
   Israel
Email:  [EMAIL PROTECTED]

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: 013 Barak PPTP on Linksys WRT54G wireless router doesnt work?

[Ilya Konstantinov]

David Harel wrote:
Just for the sake of my understanding.
I also use cable technology and 013 Barak. I never had to do any 
dialing or any operation to have my communication started. Simple DHCP 
works just fine although I asked for what they call Permanent 
connection which is free of charge.

What is this dialing all about, what is pns.barak.net.il and what are 
the benefits of this PPTP technology?

pns.barak.net.il is Barak's access point within the cable operator's 
private network. It's your gateway to the Internet by using Barak -- in 
case you're a Barak customer and can authenticate to this gateway.

--
By first accessing your cable operator's private network and dialing to 
the ISP from there, you get the benefit of being able to switch among 
ISPs dynamically; by choosing to which ISP to tunnel, you're sort-of 
dialing to it, only over IP.

Of course, dialing this way to an ISP requires an account at that ISP, 
and anyway I don't know anyone who'd wish to have a broadband account at 
more than one ISP (the speed is per the entire connection, so you'll 
gain no speed benefit).

Also, you can change your IP by disconnecting the PPTP connection and 
redialing.


=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

013 Barak PPTP on Linksys WRT54G wireless router doesnt work?

[Danny Lieberman]

guys
I'm setting up a new account at home for 013 Barak over a cable modem 
and Linksys wrt54g

The 013 Barak PPTP doesnt work for my Linksys WRT54..(Works 
ok on a w2k box )  ;-( 

I setup the definitions with pns.barak.net.il and the box dials up and 
gets a PPP address and DNS addresses
but I get network unreachable - it feels like a network mask issue but I 
cant convince the Linksys to go the last mile

Ideas?
thanks!
danny
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: 013 Barak PPTP on Linksys WRT54G wireless router doesnt work?

[Shaul Karl]

On Thu, Jan 06, 2005 at 01:13:39PM +0200, Danny Lieberman wrote:
 guys
 
 I'm setting up a new account at home for 013 Barak over a cable modem 
 and Linksys wrt54g
 
 The 013 Barak PPTP doesnt work for my Linksys WRT54..(Works 
 ok on a w2k box )  ;-( 
 
 I setup the definitions with pns.barak.net.il and the box dials up and 
 gets a PPP address and DNS addresses
 but I get network unreachable - it feels like a network mask issue but 
 I cant convince the Linksys to go the last mile
 
 Ideas?
 


  Can you ping by IP address? What is in the logs?
Your message mentioned pns.barak.net.il. This is a typo, isn't it?

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

pptp and Internet Zahav

[Ilya Dinkin]

Hi everybody,
I'm trying to get my Internet Zahav over cables internet connection to 
work...
I use pptp to do so and have encountered the following problem:
When I issue the command #pppd call cables/iz debug dump logfd 2 it stucks 
and does not write any useful debugging messages nowhere ( have checked 
/var/log/messages and /var/log/syslog ). So I can't figure why my connection 
does not work...

So my question is: How do I get this debug info? (pptp 1.5.0 + pppd 2.4.2 + 
Slack 10 + Kernel 2.4.26)

BTW. Have anybody successfuly got permission for connecting without dialers 
but directly with dhcp from Internet Zahav?

Thanks,
Ilya Dinkin
_
The new MSN 8: smart spam protection and 2 months FREE*  
http://join.msn.com/?page=features/junkmail

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: pptp discarding out-of-order

[Ilya Konstantinov]

On Tue, Jul 20, 2004 at 12:28:00PM +0200, Tzahi Fadida wrote:
 Jul 19 20:24:19 LinuxRules pptp[713]: log[decaps_gre:pptp_gre.c:262]:
 discarding out-of-order  seq is 9995 seqrecv is 9996

I have those as well, on NetVision cable:

Jul 18 06:04:38 furr pptp[12499]: anon log[decaps_gre:pptp_gre.c:397]:
buffering out-of-order packet 1091406 (expecting 1091405)
Jul 18 06:04:38 furr pptp[12499]: anon log[decaps_gre:pptp_gre.c:391]:
discarding duplicate or old packet 1091405 (expecting 1091407)

My guess: PPTP implements GRE itself (instead of using a network stack
provided by the kernel), and instead of reseting the connection over
an error (is there any way to request a re-send on GRE?), it swallows
the error.

No harm done, I guess. Those kind of errors probably happen daily on
your TCP connections but you don't notice them since TCP silently
recovers.

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

RE: pptp discarding out-of-order

[Tzahi Fadida]

probably. whats annoying is that probably you can't suppress these
messages without recompiling the source.
and we are talking, 1 every few minutes.
I removed the debug on the line but it only affects pppd
and not pptp.

Regards,
tzahi.

 -Original Message-
 From: Ilya Konstantinov [mailto:[EMAIL PROTECTED] On Behalf 
 Of Ilya Konstantinov
 Sent: Wednesday, July 21, 2004 6:31 PM
 To: Tzahi Fadida
 Cc: [EMAIL PROTECTED]
 Subject: Re: pptp discarding out-of-order
 
 
 On Tue, Jul 20, 2004 at 12:28:00PM +0200, Tzahi Fadida wrote:
  Jul 19 20:24:19 LinuxRules pptp[713]: 
 log[decaps_gre:pptp_gre.c:262]: 
  discarding out-of-order  seq is 9995 seqrecv is 9996
 
 I have those as well, on NetVision cable:
 
 Jul 18 06:04:38 furr pptp[12499]: anon 
 log[decaps_gre:pptp_gre.c:397]: buffering out-of-order packet 
 1091406 (expecting 1091405) Jul 18 06:04:38 furr pptp[12499]: 
 anon log[decaps_gre:pptp_gre.c:391]: discarding duplicate or 
 old packet 1091405 (expecting 1091407)
 
 My guess: PPTP implements GRE itself (instead of using a 
 network stack provided by the kernel), and instead of 
 reseting the connection over an error (is there any way to 
 request a re-send on GRE?), it swallows the error.
 
 No harm done, I guess. Those kind of errors probably happen 
 daily on your TCP connections but you don't notice them since 
 TCP silently recovers.
 
 



=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

pptp discarding out-of-order

[Tzahi Fadida]

Hi,
Does anyone knows what these mean? and how to minimize them.
there is one line for every few minutes.
background: I have a cables connection to netvision.
pptp 1.1.0
pppd 2.4.1b2
kernel 2.4.4
connection line:
/usr/sbin/pptp-linux cable.netvision.net.il debug user $USERNAME remotename
cable.netvision.net.il mtu 1460 mru 1460 defaultroute

Jul 19 20:24:19 LinuxRules pptp[713]: log[decaps_gre:pptp_gre.c:262]:
discarding out-of-order  seq is 9995 seqrecv is 9996
Jul 19 20:34:09 LinuxRules pptp[713]: log[decaps_gre:pptp_gre.c:262]:
discarding out-of-order  seq is 17895 seqrecv is 17896
Jul 19 20:37:09 LinuxRules pptp[713]: log[decaps_gre:pptp_gre.c:262]:
discarding out-of-order  seq is 19365 seqrecv is 19366
Jul 19 20:41:10 LinuxRules pptp[713]: log[decaps_gre:pptp_gre.c:262]:
discarding out-of-order  seq is 21550 seqrecv is 21551
Jul 19 20:53:00 LinuxRules pptp[713]: log[decaps_gre:pptp_gre.c:262]:
discarding out-of-order  seq is 32073 seqrecv is 32073
Jul 19 21:00:12 LinuxRules pptp[713]: log[decaps_gre:pptp_gre.c:262]:
discarding out-of-order  seq is 54684 seqrecv is 54685
Jul 19 21:07:11 LinuxRules pptp[713]: log[decaps_gre:pptp_gre.c:262]:
discarding out-of-order  seq is 67279 seqrecv is 67285

Regards,
tzahi.



=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: Hebrew fonts and PPTP

[Muli Ben-Yehuda]

On Mon, Jan 26, 2004 at 08:34:30AM +0200, Yaniv Almog wrote:

 2. Last June I installed a PPTP client (from pptpclient.sourceforge.net)
 to connect my computer through the cables to the Technion VPN server. It
 worked smoothly for two months but then, in August, because of
 Blasterworm, the Technion made some changes, one of them was to block
 the ping port. Since then I cannot connect to the internet. The PPTP
 clent dies off approximately one minuet after it starts (the login
 process works fine). Any ideas how to solve the problem? (I can still
 connect from my XP partition)

Some things to try, if no one knows the correct answer off the top of
her head: 

- the excellent pptp diagnosys howto:
http://pptpclient.sourceforge.net/howto-diagnosis.phtml 

- enabling debug
(http://pptpclient.sourceforge.net/howto-diagnosis.phtml#debug) so we
can know what is happening. 

Cheers, 
Muli 
-- 
Muli Ben-Yehuda
http://www.mulix.org | http://mulix.livejournal.com/

the nucleus of linux oscillates my world - [EMAIL PROTECTED]



signature.asc
Description: Digital signature

Re: Hebrew fonts and PPTP

[Geoffrey S. Mendelson]

Yaniv Almog wrote:
 2. Last June I installed a PPTP client (from pptpclient.sourceforge.net)
 to connect my computer through the cables to the Technion VPN server. It
 worked smoothly for two months but then, in August, because of
 Blasterworm, the Technion made some changes, one of them was to block
 the ping port. Since then I cannot connect to the internet. The PPTP
 clent dies off approximately one minuet after it starts (the login
 process works fine). Any ideas how to solve the problem? (I can still
 connect from my XP partition)

Is your XP set up to use PPTP or L2TP? 

Geoff.

-- 
Geoffrey S. Mendelson [EMAIL PROTECTED] 972-54-608-069
Icq/AIM Uin: 2661079 MSN IM: [EMAIL PROTECTED] (Not for email)



=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: Hebrew fonts and PPTP

[Tzafrir Cohen]

On Mon, Jan 26, 2004 at 08:34:30AM +0200, Yaniv Almog wrote:
 Shalom,
  
 1. I have recently installed Microsoft's TTF fonts on my Fedora core
 partition. I have installed them both with ttmkfdir and fc-cache.
  However, I am getting squares on my screen when I try to view them on
 my screen. I didn't have that problem with Redhat 9. any possible
 explanation and, of course, a solution?

There are two separate font mechanisms used for X programs: 

The older one, the core fonts: fonts with names such as
'-culmus-nachlieli-medium-r-normal--*-*-*-*-*-*-iso8859-8' (actually:
the '*' are wild cards). Most programs in your desktop don't use it.

There is also Xft2/fontconfig. Its main config file is
/etc/fonts/fonts.conf . However there is generally no need to edit it,
as it already includes /usr/share/fonts/ and everything under it.

However if every client would have needed to scan all the tree under
/usr/share/fonts upon startup this would have been a waste of time. This
is why Xft maintains a cache of fonts. The command fc-cache is for
maintaining it.

So to install a font you need to create a subdirectory under
/usr/share/fonts , put your fonts there, and then update the cache using
fc-cache.

ttmkfdir is unnecessary (except if you use the obsolete Xft1. And if you
don't use RH73 or Mandrake 8.2(?) you probably don't use it). It is
required if you want to make the fonts avialable as core fonts. But this
is for a different message (needed?)

  
 2. Last June I installed a PPTP client (from pptpclient.sourceforge.net)
 to connect my computer through the cables to the Technion VPN server. It
 worked smoothly for two months but then, in August, because of
 Blasterworm, the Technion made some changes, one of them was to block
 the ping port. Since then I cannot connect to the internet. The PPTP
 clent dies off approximately one minuet after it starts (the login
 process works fine). Any ideas how to solve the problem? (I can still
 connect from my XP partition)

One guess:

Sounds like a routing problem: your default route becomes the pptp
target, and then you try to send even the pptp tunnel packets through
that connection.

What is the output of 'route -n' immidetly after a disconnection?

-- 
Tzafrir Cohen   +---+
http://www.technion.ac.il/~tzafrir/ |vim is a mutt's best friend|
mailto:[EMAIL PROTECTED]   +---+

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Hebrew fonts and PPTP

[Yaniv Almog]

Shalom,



1. I have recently installed Microsofts TTF fonts on
my Fedora core partition. I have installed them both with ttmkfdir
and fc-cache.

However, I am
getting squares on my screen when I try to view them on my screen. I didnt
have that problem with Redhat 9. any
possible explanation and, of course, a solution?



2. Last June I installed a PPTP client (from
pptpclient.sourceforge.net) to connect my computer through the cables to the
Technion VPN server. It worked smoothly for two months but then, in August, because
of Blasterworm, the Technion made some changes, one
of them was to block the ping port. Since then I cannot connect to the internet.
The PPTP clent dies off approximately one minuet
after it starts (the login process works fine). Any ideas how to solve the
problem? (I can still connect from my XP partition)



Thanks,

Yaniv

Re: Barak Cables over PPTP

[Oded Arbel]

On Sunday 04 January 2004 07:02, Ittay Dror wrote:
 Hi everyone,

 I'm trying to setup a dialer to Barak over PPTP (and Ethernet). Does
 someone have a ready-made script (for Mandrake 9.1) so that I don't have
 to mess with it myself? Also, instructions for how to make a connection
 sharing (not a gw) so another computer can use its own dialer will be
 appriciated.

try DrakConnect. it does ok in 9.2, I have no reason to believe it won't be ok 
for 9.1. you might need to update your PPTP client though.

-- 
Oded

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: Barak Cables over PPTP

[Ez-Aton]

You could find something you can work with (it's written to RH9, but will work 
with Mandrake too, as far as I know) at 
http://www.iarc.org/~ezaton/cables
There are instructions, and a script, adjusted to Actcom, but you could adjust 
it easilly to any ISP you want.

Ez.

On Sunday 04 January 2004 07:02, Ittay Dror wrote:
 Hi everyone,

 I'm trying to setup a dialer to Barak over PPTP (and Ethernet). Does
 someone have a ready-made script (for Mandrake 9.1) so that I don't have
 to mess with it myself? Also, instructions for how to make a connection
 sharing (not a gw) so another computer can use its own dialer will be
 appriciated.

 Thanx,
 Ittay


=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Barak Cables over PPTP

[Ittay Dror]

Hi everyone,

I'm trying to setup a dialer to Barak over PPTP (and Ethernet). Does 
someone have a ready-made script (for Mandrake 9.1) so that I don't have 
to mess with it myself? Also, instructions for how to make a connection 
sharing (not a gw) so another computer can use its own dialer will be 
appriciated.

Thanx,
Ittay
--
===
Ittay Dror ([EMAIL PROTECTED])
User Space Team, RD
Qlusters Inc.
+972-3-6081956 Fax: +972-3-6081841
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Handy script for pptp on Cables (the 'default route' problem)

[Ilya Konstantinov]

Hi,

Those of you, who use Cable Internet, are probably aware of the fact
that you need to create a static route (with the 'route' or 'ip'
utility) to your ISP's PPTP server.

This Perl script wraps 'pptp' and creates/removes this static route for
you. Simply replace your existing 'pptp' command with 'pptp-routed' and
you no longer need to track whatever changes your ISP does to its
configuration, or write annoying startup scripts, especially if your
ISP has multiple round-robin PPTP servers (e.g. NetVision).

URL: http://www.iglu.org.il/~future/pptp-routed

Technically, it figures out the route you use to reach the PPTP server
when connecting, and creates a static route to the PPTP server for the
duration of the connection (when the default route would be overridden
by pppd). Microsoft's VPN dialer does this automatically and I'd love
to see this functionality merged into pptp itself.

--
Share and enjoy!

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

pptp forking wildly with kernel-2.6.0-test5

[Lior Kesos]

I tried to upgrade my home computer to 2.6.0-test5 and when I ran my 
cable connection script.
pptp started forking like crazy untill I ran out of proccess and got the 
infamous can't fork error.
anybody have a theory about what changes 2.6 introduces that could cause 
this.
BTW I use pppd calling pptp for a connection to 012 through Arutzey 
Zahav.

thanks for any info -
Lior Kesos.
--
Lior Kesos  -  [EMAIL PROTECTED]
Content Development Team Leader
==
Everything should be made as simple as possible -
but not simpler -- Albert Einstein 



=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

pptp client problem using user name with '#'

[YBEN]

Hi,

I'm trying to connect using user name that contain '#' sign (My ISP use it
for IP VPN) without success.
Connecting with user name without '#' work just fine.

I'm using:
1. kernel 2.4.20
2. pppd version 2.4.0b4
3. pptp-linux version 1.1.0
4. for the tests the pptp server install on Win2K server (the Win2K client
work fine...)
4. my ppp option file contains:
115200
lock
modem
crtscts
nodetach
asyncmap 0
nodetach
mtu 585
mru 585
5. the output I recive is:

==
172.17.2.39:/mnt/params/etc/ppp  /bin/pptp 172.17.1.40 user Yben#user
usepeerdn
s ipcp-accept-remote noipdefault noccp ipparam PPtP debug
using channel 23
Using interface ppp0
Connect: ppp0 -- /dev/pts/0
sent [LCP ConfReq id=0x1 mru 585 asyncmap 0x0 magic 0x2f9b1434 pcomp
ac
comp]
rcvd [LCP ConfReq id=0x0 auth chap 81 magic 0x1b4f35c0 pcomp accomp
cal
lback CBCP mrru 1614 endpoint
[local:03.53.6c.ad.7b.22.4d.f2.b7.24.5d.2a.bd.
c1.2d.3b.00.00.00.00]  17 04 01 3e]
sent [LCP ConfRej id=0x0 callback CBCP mrru 1614  17 04 01 3e]
rcvd [LCP ConfNak id=0x1 mru 1500]
sent [LCP ConfReq id=0x2 asyncmap 0x0 magic 0x2f9b1434 pcomp accomp]
rcvd [LCP ConfReq id=0x1 auth chap 81 magic 0x1b4f35c0 pcomp accomp
end
point [local:03.53.6c.ad.7b.22.4d.f2.b7.24.5d.2a.bd.c1.2d.3b.00.00.00.00]]
sent [LCP ConfNak id=0x1 auth pap]
rcvd [LCP ConfAck id=0x2 asyncmap 0x0 magic 0x2f9b1434 pcomp accomp]
rcvd [LCP ConfReq id=0x2 auth chap m$oft magic 0x1b4f35c0 pcomp
accomp 
endpoint
[local:03.53.6c.ad.7b.22.4d.f2.b7.24.5d.2a.bd.c1.2d.3b.00.00.00.00]]
sent [LCP ConfNak id=0x2 auth pap]
rcvd [LCP ConfReq id=0x3 auth chap m$oft magic 0x1b4f35c0 pcomp
accomp 
endpoint
[local:03.53.6c.ad.7b.22.4d.f2.b7.24.5d.2a.bd.c1.2d.3b.00.00.00.00]]
sent [LCP ConfNak id=0x3 auth pap]
rcvd [LCP ConfReq id=0x4 auth chap m$oft magic 0x1b4f35c0 pcomp
accomp 
endpoint
[local:03.53.6c.ad.7b.22.4d.f2.b7.24.5d.2a.bd.c1.2d.3b.00.00.00.00]]
sent [LCP ConfNak id=0x4 auth pap]
rcvd [LCP ConfReq id=0x5 auth chap m$oft magic 0x1b4f35c0 pcomp
accomp 
endpoint
[local:03.53.6c.ad.7b.22.4d.f2.b7.24.5d.2a.bd.c1.2d.3b.00.00.00.00]]
sent [LCP ConfNak id=0x5 auth pap]
rcvd [LCP ConfReq id=0x6 auth chap m$oft magic 0x1b4f35c0 pcomp
accomp 
endpoint
[local:03.53.6c.ad.7b.22.4d.f2.b7.24.5d.2a.bd.c1.2d.3b.00.00.00.00]]
sent [LCP ConfRej id=0x6 auth chap m$oft]
rcvd [LCP TermReq id=0x7 1b 4f 35 c0 00 3c cd 74 00 00 03 97]
sent [LCP TermAck id=0x7]
sent [LCP ConfReq id=0x2 asyncmap 0x0 magic 0x2f9b1434 pcomp accomp]
sent [LCP ConfReq id=0x2 asyncmap 0x0 magic 0x2f9b1434 pcomp accomp]
sent [LCP ConfReq id=0x2 asyncmap 0x0 magic 0x2f9b1434 pcomp accomp]
sent [LCP ConfReq id=0x2 asyncmap 0x0 magic 0x2f9b1434 pcomp accomp]
sent [LCP ConfReq id=0x2 asyncmap 0x0 magic 0x2f9b1434 pcomp accomp]
sent [LCP ConfReq id=0x2 asyncmap 0x0 magic 0x2f9b1434 pcomp accomp]
sent [LCP ConfReq id=0x2 asyncmap 0x0 magic 0x2f9b1434 pcomp accomp]
sent [LCP ConfReq id=0x2 asyncmap 0x0 magic 0x2f9b1434 pcomp accomp]
sent [LCP ConfReq id=0x2 asyncmap 0x0 magic 0x2f9b1434 pcomp accomp]
sent [LCP ConfReq id=0x2 asyncmap 0x0 magic 0x2f9b1434 pcomp accomp]
LCP: timeout sending Config-Requests
Connection terminated.

==
Any idea?

Thanks.

Best Regards

Yehoram Ben-Yaacov
Better On-line Solutions

[EMAIL PROTECTED]

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: pptp client problem using user name with '#'

[Shaul Karl]

On Mon, Sep 15, 2003 at 07:29:08PM +0300, [EMAIL PROTECTED] wrote:
 Hi,
 
 I'm trying to connect using user name that contain '#' sign (My ISP use it
 for IP VPN) without success.
 Connecting with user name without '#' work just fine.
 
   [ ... ]

 172.17.2.39:/mnt/params/etc/ppp  /bin/pptp 172.17.1.40 user Yben#user
 usepeerdn
 s ipcp-accept-remote noipdefault noccp ipparam PPtP debug


  This is the command you issued, isn't it? Have you tried replacing
Yben#user with Yben\#user? I do mean that you will use the backslash 
character.

  In addition, you might try comparing the output of the user that does
manage to connect to this user. 

-- 

Shaul Karl,shaulk @ actcom . net . il

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

pptp into rc file

[Yehuda Berlinger]

I would like to set up my working pptp command to automatically run when 
I start up by putting it into an rc file. I'm using redhat 9.0.

Has anyone done this already who can send me a file illustrating how to 
do it (minus my specific account info, of course)?

Yehuda

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: pptp into rc file

[Oleg Kobets]

Hi.

I would guess, just enter your command in the appropriate place in one of
the RC files.

Or do what I did, take the skeleton file from /etc/init.d, change it to
include your command and all the relevant info and make links so it would be
executed on startup. Under debian there is a command to do it automaticly,
update-rc.d

I hope that helped.

Alon.

- Original Message - 
From: Yehuda Berlinger [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Sent: Friday, August 29, 2003 11:04 AM
Subject: pptp into rc file


 I would like to set up my working pptp command to automatically run when
 I start up by putting it into an rc file. I'm using redhat 9.0.

 Has anyone done this already who can send me a file illustrating how to
 do it (minus my specific account info, of course)?

 Yehuda


 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]






=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: Strange pptp-linux problem

[Amit Margalit]

Hi Eliran,

Do you have 'debug' in your /etc/ppp/options? Also, can you post your
/etc/ppp/options ?

This file can cause all sorts of trouble with pppd if not configured properly.

Amit

Eliran wrote:

 Dani Arbel wrote:

 There isn't pptp-linux and/or pptp-adsl .
 There is only pptp .
 
 MDK9 CD1 has a package called pptp-adsl not pptp.
 The pptp MDK has is completly different than the one in Amir Tal's
 tar.gz package

 Try to understand the debug info and correct your setup. Anyway, since your
 station started the LCP stage of ppp , it means that pptp works fine, and it
 is ppp that fails.
 
 I installed ppp separatly, after the installation.

 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]


=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: Strange pptp-linux problem

[Dani Arbel]

Eliran,
There isn't pptp-linux and/or pptp-adsl .
There is only pptp .
Try to understand the debug info and correct your setup. Anyway, since your 
station started the LCP stage of ppp , it means that pptp works fine, and it 
is ppp that fails.
Dani

On Wednesday 05 February 2003 12:12, you wrote:
 Muli Ben-Yehuda wrote:
 Looks like the remote server does not want to do pap authentication
 with you. Try chap. No idea why it worked with the other distributions
 (assuming that you were precise when you said the configuration is the
 same).

 You mean, copy my authentication details into chaps-secrets instead of
 pap-secrets ?

 Such as?

 It shows me some source code + the errors.
 This isn't releavent now as I'm interested in getting the pptp-linux
 binary to work,
 rather than the 'pptp' one which (I assume) the pptp-adsl package
 installed.





 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]

-- 


Dr. Daniel Arbel[EMAIL PROTECTED]
Network Manager tel:   972-4-8294992
Technion Computer Centerfax:   972-4-8222872


=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: Strange pptp-linux problem

[Dani Arbel]

Hi!
From the logs it looks like the other side (servers) requests PAP 
authentication, and your machine Rej(ects) it. If you can't converge on an 
authentication method, ppp will fail at LCP level.
Check your dialer authentication setup.
Dani

On Tuesday 04 February 2003 19:04, Amir Tal wrote:
 On Tuesday 04 February 2003 18:28, you wrote:
  Hello !
 
  I recently installed MDK9 and after *successfuly* being able to get
  online under Debian and RedHat,
  found that things aren't the same in MDK.
 
  I use the same scripts and configuration as before but there seems to be
  an error with the
  pptp-linux binary. I first started it and noticed (pptp-linux notified
  me actually ) there is no
  'pppd' package so I couldn't run it So I launched my MDK CD and
  installed the appropriate RPM...
 
  Still, things are the same. There is no problem with the routing table
  neither eth cards.
  I checked the logs and it seems the modems HANGUP after a sequence of
  sent/received
  (or something like that), here is it:
 
  log
 
  Feb  4 16:12:51 rulix pptp[4062]:
  log[pptp_dispatch_ctrl_packet:pptp_ctrl.c:707]: Outgoing call
  established (call ID 0, peer's call ID 50633).
  Feb  4 16:12:51 rulix pppd[4064]: pppd 2.4.1 started by root, uid 0
  Feb  4 16:12:51 rulix pppd[4064]: using channel 9
  Feb  4 16:12:51 rulix pppd[4064]: Using interface ppp0
  Feb  4 16:12:51 rulix pppd[4064]: Connect: ppp0 -- /dev/pts/2
  Feb  4 16:12:51 rulix pppd[4064]: sent [LCP ConfReq id=0x1 mru 1460
  asyncmap 0x0 magic 0x5e3fc581 pcomp accomp]
  Feb  4 16:12:51 rulix /etc/hotplug/net.agent: assuming ppp0 is already up
 
  /* Holds for a few seconds ... And */
 
  Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x1 mru 1462
  auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent
  [LCP ConfRej id=0x1 auth pap]
  Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x2 mru 1462
  auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent
  [LCP ConfRej id=0x2 auth pap]
  Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x3 mru 1462
  auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent
  [LCP ConfRej id=0x3 auth pap]
  Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x4 mru 1462
  auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent
  [LCP ConfRej id=0x4 auth pap]
  Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x5 mru 1462
  auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent
  [LCP ConfRej id=0x5 auth pap]
  Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x6 mru 1462
  auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent
  [LCP ConfRej id=0x6 auth pap]
  Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x7 mru 1462
  auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent
  [LCP ConfRej id=0x7 auth pap]
  Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x8 mru 1462
  auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent
  [LCP ConfRej id=0x8 auth pap]
  Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x9 mru 1462
  auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent
  [LCP ConfRej id=0x9 auth pap]
  Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0xa mru 1462
  auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent
  [LCP ConfRej id=0xa auth pap]
  Feb  4 16:12:53 rulix pptp[4062]: log[call_callback:pptp_callmgr.c:88]:
  Closing connection
  Feb  4 16:12:53 rulix pppd[4064]: Hangup (SIGHUP)
  Feb  4 16:12:53 rulix pppd[4064]: Modem hangup
  Feb  4 16:12:53 rulix pppd[4064]: Connection terminated.
  Feb  4 16:12:53 rulix /etc/hotplug/net.agent: NET unregister event not
  supported
  Feb  4 16:12:54 rulix pppd[4064]: Exit.
 
  /log
 
  This line looks strange:
  Feb  4 16:12:51 rulix /etc/hotplug/net.agent: assuming ppp0 is already up
 
  If this is some sort of MDK feature don't shout on me, I just decided to
  give it a try ;-)

 i hope this is not a feature. god help mandrake if it is ;)
 you didn't say what type of connection is this. ADSL ? cables ?

  Then I went to rpmfind.net and d/l'ed the latest version 'pptp-client'
  for MDK which installed 'pptp'.
  This pptp binary is very strange and always show me bizarre errors.
 
  BTW Does any of you know why the Generic Disc of Debian-3.0 Woody does
  not come with a default of a 2.4 kernel ?
I have to download a separate CD for it. Otherwise I have to
  compile it manually after installing the 2.2 one.
Who would want to use 2.2 ??!!

 type bf24 at the prompt, and a 2.4.18 kernel will be loaded.
 tal.

  Thanks !!!
  Eliran
 
 
 
  =
  To unsubscribe, send mail to [EMAIL PROTECTED] with
  the word unsubscribe in the message body, e.g., run the command
  echo unsubscribe | mail [EMAIL PROTECTED]

 =
 To unsubscribe, send mail to [EMAIL PROTECTED

Re: Strange pptp-linux problem

[Eliran]

Muli Ben-Yehuda wrote:


Looks like the remote server does not want to do pap authentication
with you. Try chap. No idea why it worked with the other distributions
(assuming that you were precise when you said the configuration is the
same). 

You mean, copy my authentication details into chaps-secrets instead of
pap-secrets ?


Such as? 

It shows me some source code + the errors.
This isn't releavent now as I'm interested in getting the pptp-linux
binary to work,
rather than the 'pptp' one which (I assume) the pptp-adsl package installed.





=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: Strange pptp-linux problem

[Eliran]

Dani Arbel wrote:


There isn't pptp-linux and/or pptp-adsl .
There is only pptp .


MDK9 CD1 has a package called pptp-adsl not pptp.
The pptp MDK has is completly different than the one in Amir Tal's 
tar.gz package

Try to understand the debug info and correct your setup. Anyway, since your 
station started the LCP stage of ppp , it means that pptp works fine, and it 
is ppp that fails.

I installed ppp separatly, after the installation.




=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Strange pptp-linux problem

[Eliran]

Hello !

I recently installed MDK9 and after *successfuly* being able to get 
online under Debian and RedHat,
found that things aren't the same in MDK.

I use the same scripts and configuration as before but there seems to be 
an error with the
pptp-linux binary. I first started it and noticed (pptp-linux notified 
me actually ) there is no
'pppd' package so I couldn't run it So I launched my MDK CD and 
installed the appropriate RPM...

Still, things are the same. There is no problem with the routing table 
neither eth cards.
I checked the logs and it seems the modems HANGUP after a sequence of 
sent/received
(or something like that), here is it:

log

Feb  4 16:12:51 rulix pptp[4062]: 
log[pptp_dispatch_ctrl_packet:pptp_ctrl.c:707]: Outgoing call 
established (call ID 0, peer's call ID 50633).
Feb  4 16:12:51 rulix pppd[4064]: pppd 2.4.1 started by root, uid 0
Feb  4 16:12:51 rulix pppd[4064]: using channel 9
Feb  4 16:12:51 rulix pppd[4064]: Using interface ppp0
Feb  4 16:12:51 rulix pppd[4064]: Connect: ppp0 -- /dev/pts/2
Feb  4 16:12:51 rulix pppd[4064]: sent [LCP ConfReq id=0x1 mru 1460 
asyncmap 0x0 magic 0x5e3fc581 pcomp accomp]
Feb  4 16:12:51 rulix /etc/hotplug/net.agent: assuming ppp0 is already up

/* Holds for a few seconds ... And */

Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x1 mru 1462 
auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent 
[LCP ConfRej id=0x1 auth pap]
Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x2 mru 1462 
auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent 
[LCP ConfRej id=0x2 auth pap]
Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x3 mru 1462 
auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent 
[LCP ConfRej id=0x3 auth pap]
Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x4 mru 1462 
auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent 
[LCP ConfRej id=0x4 auth pap]
Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x5 mru 1462 
auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent 
[LCP ConfRej id=0x5 auth pap]
Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x6 mru 1462 
auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent 
[LCP ConfRej id=0x6 auth pap]
Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x7 mru 1462 
auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent 
[LCP ConfRej id=0x7 auth pap]
Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x8 mru 1462 
auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent 
[LCP ConfRej id=0x8 auth pap]
Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x9 mru 1462 
auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent 
[LCP ConfRej id=0x9 auth pap]
Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0xa mru 1462 
auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent 
[LCP ConfRej id=0xa auth pap]
Feb  4 16:12:53 rulix pptp[4062]: log[call_callback:pptp_callmgr.c:88]: 
Closing connection
Feb  4 16:12:53 rulix pppd[4064]: Hangup (SIGHUP)
Feb  4 16:12:53 rulix pppd[4064]: Modem hangup
Feb  4 16:12:53 rulix pppd[4064]: Connection terminated.
Feb  4 16:12:53 rulix /etc/hotplug/net.agent: NET unregister event not 
supported
Feb  4 16:12:54 rulix pppd[4064]: Exit.

/log

This line looks strange:
Feb  4 16:12:51 rulix /etc/hotplug/net.agent: assuming ppp0 is already up

If this is some sort of MDK feature don't shout on me, I just decided to 
give it a try ;-)

Then I went to rpmfind.net and d/l'ed the latest version 'pptp-client' 
for MDK which installed 'pptp'.
This pptp binary is very strange and always show me bizarre errors.

BTW Does any of you know why the Generic Disc of Debian-3.0 Woody does 
not come with a default of a 2.4 kernel ?
 I have to download a separate CD for it. Otherwise I have to 
compile it manually after installing the 2.2 one.
 Who would want to use 2.2 ??!!

Thanks !!!
Eliran



=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: Strange pptp-linux problem

[Amir Tal]

On Tuesday 04 February 2003 18:28, you wrote:
 Hello !

 I recently installed MDK9 and after *successfuly* being able to get
 online under Debian and RedHat,
 found that things aren't the same in MDK.

 I use the same scripts and configuration as before but there seems to be
 an error with the
 pptp-linux binary. I first started it and noticed (pptp-linux notified
 me actually ) there is no
 'pppd' package so I couldn't run it So I launched my MDK CD and
 installed the appropriate RPM...

 Still, things are the same. There is no problem with the routing table
 neither eth cards.
 I checked the logs and it seems the modems HANGUP after a sequence of
 sent/received
 (or something like that), here is it:

 log

 Feb  4 16:12:51 rulix pptp[4062]:
 log[pptp_dispatch_ctrl_packet:pptp_ctrl.c:707]: Outgoing call
 established (call ID 0, peer's call ID 50633).
 Feb  4 16:12:51 rulix pppd[4064]: pppd 2.4.1 started by root, uid 0
 Feb  4 16:12:51 rulix pppd[4064]: using channel 9
 Feb  4 16:12:51 rulix pppd[4064]: Using interface ppp0
 Feb  4 16:12:51 rulix pppd[4064]: Connect: ppp0 -- /dev/pts/2
 Feb  4 16:12:51 rulix pppd[4064]: sent [LCP ConfReq id=0x1 mru 1460
 asyncmap 0x0 magic 0x5e3fc581 pcomp accomp]
 Feb  4 16:12:51 rulix /etc/hotplug/net.agent: assuming ppp0 is already up

 /* Holds for a few seconds ... And */

 Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x1 mru 1462
 auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent
 [LCP ConfRej id=0x1 auth pap]
 Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x2 mru 1462
 auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent
 [LCP ConfRej id=0x2 auth pap]
 Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x3 mru 1462
 auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent
 [LCP ConfRej id=0x3 auth pap]
 Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x4 mru 1462
 auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent
 [LCP ConfRej id=0x4 auth pap]
 Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x5 mru 1462
 auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent
 [LCP ConfRej id=0x5 auth pap]
 Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x6 mru 1462
 auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent
 [LCP ConfRej id=0x6 auth pap]
 Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x7 mru 1462
 auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent
 [LCP ConfRej id=0x7 auth pap]
 Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x8 mru 1462
 auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent
 [LCP ConfRej id=0x8 auth pap]
 Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x9 mru 1462
 auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent
 [LCP ConfRej id=0x9 auth pap]
 Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0xa mru 1462
 auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent
 [LCP ConfRej id=0xa auth pap]
 Feb  4 16:12:53 rulix pptp[4062]: log[call_callback:pptp_callmgr.c:88]:
 Closing connection
 Feb  4 16:12:53 rulix pppd[4064]: Hangup (SIGHUP)
 Feb  4 16:12:53 rulix pppd[4064]: Modem hangup
 Feb  4 16:12:53 rulix pppd[4064]: Connection terminated.
 Feb  4 16:12:53 rulix /etc/hotplug/net.agent: NET unregister event not
 supported
 Feb  4 16:12:54 rulix pppd[4064]: Exit.

 /log

 This line looks strange:
 Feb  4 16:12:51 rulix /etc/hotplug/net.agent: assuming ppp0 is already up

 If this is some sort of MDK feature don't shout on me, I just decided to
 give it a try ;-)


i hope this is not a feature. god help mandrake if it is ;)
you didn't say what type of connection is this. ADSL ? cables ?

 Then I went to rpmfind.net and d/l'ed the latest version 'pptp-client'
 for MDK which installed 'pptp'.
 This pptp binary is very strange and always show me bizarre errors.

 BTW Does any of you know why the Generic Disc of Debian-3.0 Woody does
 not come with a default of a 2.4 kernel ?
   I have to download a separate CD for it. Otherwise I have to
 compile it manually after installing the 2.2 one.
   Who would want to use 2.2 ??!!

type bf24 at the prompt, and a 2.4.18 kernel will be loaded.
tal.



 Thanks !!!
 Eliran



 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]


=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: Strange pptp-linux problem

[Muli Ben-Yehuda]

On Tue, Feb 04, 2003 at 06:28:34PM +0200, Eliran wrote:

 Feb  4 16:12:53 rulix pppd[4064]: rcvd [LCP ConfReq id=0x1 mru 1462 
 auth pap magic 0x1335ceb0]Feb  4 16:12:53 rulix pppd[4064]: sent 
 [LCP ConfRej id=0x1 auth pap]

Looks like the remote server does not want to do pap authentication
with you. Try chap. No idea why it worked with the other distributions
(assuming that you were precise when you said the configuration is the
same). 

 Then I went to rpmfind.net and d/l'ed the latest version 'pptp-client' 
 for MDK which installed 'pptp'.
 This pptp binary is very strange and always show me bizarre errors.

Such as? 


=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Updates for CABLES-DHCP-PPTP-HOWTO

[Skliarouk Arie]

Hello Tal,

Today I connected an Linux firewall to Kavey Zahav (www.012.net)
and wanted to share information I gathered:

The VPN is established against Kavey Zahav's vpn servers.

The hostname is aztv.012.net.il and is resolved to SEVERAL ip numbers.

Thus you need manually to pick up one of the IP numbers and use that
number in the cablestart script and /etc/ppp/pap-secrets file.

I used pptp-linux package supplied with unstable debian branch (instead of
recommended pptp-linux downloadable from netvision) and it worked well.

PS. The computer is connected to ADSL router (not modem) as well and
connection equalization (using policy routing) works very nice.

---
Bye,  | Fax: (972)-2-6796453
Arieh | Phone: (972)-6795364



=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: Updates for CABLES-DHCP-PPTP-HOWTO

[Amir Tal]

On Tuesday 26 November 2002 15:46, Skliarouk Arie wrote:
 Hello Tal,

 Today I connected an Linux firewall to Kavey Zahav (www.012.net)
 and wanted to share information I gathered:

 The VPN is established against Kavey Zahav's vpn servers.

 The hostname is aztv.012.net.il and is resolved to SEVERAL ip numbers.

can you post your configuration here ? users on linux-il and learn from it, 
and i will also update the cables-howto with it.
please post your ifconfig, resolv.conf and your kernel's routing table 
(netstat -r -n) .



 Thus you need manually to pick up one of the IP numbers and use that
 number in the cablestart script and /etc/ppp/pap-secrets file.

 I used pptp-linux package supplied with unstable debian branch (instead of
 recommended pptp-linux downloadable from netvision) and it worked well.

actualy, i think its the same binary ;)

tal.



 PS. The computer is connected to ADSL router (not modem) as well and
 connection equalization (using policy routing) works very nice.

 ---
 Bye,  | Fax: (972)-2-6796453
 Arieh | Phone: (972)-6795364


=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: PPPoE instead of PPTP on Bezeq ADSL - howto

[Oleg Kobets]

It sounds right in theory, has anybody tried it, besides Doron ???

As you know, there is 2 common versions of Alctatel modems, version 2.x and
version 3.x

Now, the config is slightly different and so is the functionality. What I
need to know, is there someone who succedded in this setup over the 3.x
version ?

---
Oleg Kobets
Network Administrator
Breakthrough LTD.
054-747132
03-6349922 Ext 26

Black hole is God divided by zero

- Original Message -
From: Doron Shikmoni [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Sent: Sunday, November 24, 2002 1:31 AM
Subject: PPPoE instead of PPTP on Bezeq ADSL - howto


 Hi all,

 Bezeq's installation instructions for ADSL user require setting up a
 PPTP VPN (tunnel).
 It seems as if many people believe that this is a firm requirement, and
 that it would not
 be possible to use different schemes - case in point, PPPoE - to connect
 to ADSL,
 because Bezeq does not support PPPoE.  A bunch of low-cost home/SOHO
 routers
 come equipped with a PPPoE client, but not with a PPTP client. By many,
 they are
 thought to be unusable in Israel.
 (PPPoE is common in the US and in many other parts of the world, as the
 protocol
 of choice to connect to ADSL).

 Well, you can do it now. If you want to do away with PPTP, and use PPPoE
 to connect
 to Bezeq's ADSL, you are invited to read over the instructions at:
 http://www.isoc.org.il/~doron/PPPoE.html .
 (at this point, the document deals with the Alcatel ADSL modem. I hope
 to be able
 to update it for other types as well).

 Comments welcome.

 Enjoy!
 Doron




 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]



=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

PPPoE instead of PPTP on Bezeq ADSL - howto

[Doron Shikmoni]

Hi all,

Bezeq's installation instructions for ADSL user require setting up a 
PPTP VPN (tunnel).
It seems as if many people believe that this is a firm requirement, and 
that it would not
be possible to use different schemes - case in point, PPPoE - to connect 
to ADSL,
because Bezeq does not support PPPoE.  A bunch of low-cost home/SOHO 
routers
come equipped with a PPPoE client, but not with a PPTP client. By many, 
they are
thought to be unusable in Israel.
(PPPoE is common in the US and in many other parts of the world, as the 
protocol
of choice to connect to ADSL).

Well, you can do it now. If you want to do away with PPTP, and use PPPoE 
to connect
to Bezeq's ADSL, you are invited to read over the instructions at:
http://www.isoc.org.il/~doron/PPPoE.html .
(at this point, the document deals with the Alcatel ADSL modem. I hope 
to be able
to update it for other types as well).

Comments welcome.

Enjoy!
Doron




=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: PPPoE instead of PPTP on Bezeq ADSL - howto

[Mark Veltzer]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Sunday 24 November 2002 01:31, you wrote:
 Comments welcome.

Just one comment: Great stuff. Since most distros come with PPPoE and not 
PPTP it makes installing Bezeq ADSL a little easier for newbie Linux users. 
Thanks for writing the document. Any chance of someone adding it to the main 
documents at iglu (at the FAQ section which deals with networking and ADSL) ?

Cheers,
Mark.


 Enjoy!
 Doron

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQE94B1oxlxDIcceXTgRAvmBAKCFa1Na38yLbd8N+K4UsfEW9CtJAgCeNKkk
6wJSxhGchu5jHsW8IM5BSsY=
=ZWX7
-END PGP SIGNATURE-

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

PPtP problem

[Michael Sternberg]

I have problem using Win2K VPN server from Linux.

I success to connect to it, to receive an IP address in remote network,
but when I try to ping something inside intranet I fail. I even can not
ping other side of PPtP link. When I replace my default gateway with
second side of link - PPtP connection stops after a few seconds.

The whole process of connection goes well, I have an MPPE patched 2.4.16
kernel and no error messages are seen when I start pptp with debug
option.

Where I can look to understand the problem ?
I feel that this is something related to routing tables - but what ?

Thanks, Michael.

-- 



=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: PPtP problem

[Baruch Even]

First thing to check is if you added a routing entry for the internal
network.

PPTP by default only adds a route to the internal ip of the pptp
gateway, that is if the pptp server gives you a 192.168.1.2 and it has a
192.168.1.1, there will be only a route to 192.168.1.1, you need to add
a route of 192.168.1.0/24 yourself.

If you remove the default gateway route and don't place some other route
on which the tcp/gre packets can travel the link will die when the ppp
process will send an LCP echo request and receive an ICMP unreachable
error packet.

After you connect with pptp check what routes you have, you need to have
one route on the ppp+ link, you can then try to ping the internal ip of
the pptp server, this should work (ping 192.168.1.1, with the numbers
above).

If it doesn't work, there is probably a firewall on that pptp server
that prevents it from answering your icmp packets.

If it did work just add the route for the internal network over the ppp+
link and you should be able to access any non-firewalled computer on the
internal network.

Baruch

* Michael Sternberg [EMAIL PROTECTED] [021121 19:21]:
 
 I have problem using Win2K VPN server from Linux.
 
 I success to connect to it, to receive an IP address in remote network,
 but when I try to ping something inside intranet I fail. I even can not
 ping other side of PPtP link. When I replace my default gateway with
 second side of link - PPtP connection stops after a few seconds.
 
 The whole process of connection goes well, I have an MPPE patched 2.4.16
 kernel and no error messages are seen when I start pptp with debug
 option.
 
 Where I can look to understand the problem ?
 I feel that this is something related to routing tables - but what ?
 
 Thanks, Michael.
 
 -- 
 
 
 
 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]
 

-- 
Baruch Even
http://baruch.ev-en.org/

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Problems with pptp client and PopTop VPN server

[Michael Sternberg]

Hello

I'm trying to establish PPtP VPN connection between two Linux
boxes using pptp client and pptpd (poptop) server.

The most annoying thing is that I'm able to connect to MS VPN
server using pptp client and I'm able to connect to Linux pptpd 
server using MS VPN connection.

Problem is from Linux to Linux.

I started both client and server in debug mode.
Client starts to send LCP ConfReq in loop.
Server prints messages saying:
pppd : Connect: ppp0 -- /dev/ttyp1
pptpd: GRE: read from network failed: status=-1 error=Protocol not available
pptpd: CTRL: GRE read or PTY write failed (gre,pty)=(5,4)
pptpd: CTRL: Client control connection finished
pptpd: CTRL: Exiting now
pptpd: MGR: Reaped child
pppd : Modem hangup
pppd : Connection terminated
and quits.

I tried first without ip_gre modules loaded - it still worked with
MS apps. Loading ip_gre module did not changed the situation on 
Linux vs. Linux connections.

Please help.
Michael.

-- 

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: Problems with pptp client and PopTop VPN server

[Oleg Kobets]

maybe it has something to do with MPPE ? As far as I understood, you need
mppe enabled pptp and some kernel module ?

please correct me if I wrong.

Oleg.

- Original Message -
From: Michael Sternberg [EMAIL PROTECTED]
To: LinuxIL [EMAIL PROTECTED]
Sent: Thursday, October 24, 2002 10:26 AM
Subject: Problems with pptp client and PopTop VPN server



 Hello

 I'm trying to establish PPtP VPN connection between two Linux
 boxes using pptp client and pptpd (poptop) server.

 The most annoying thing is that I'm able to connect to MS VPN
 server using pptp client and I'm able to connect to Linux pptpd
 server using MS VPN connection.

 Problem is from Linux to Linux.

 I started both client and server in debug mode.
 Client starts to send LCP ConfReq in loop.
 Server prints messages saying:
 pppd : Connect: ppp0 -- /dev/ttyp1
 pptpd: GRE: read from network failed: status=-1 error=Protocol not
available
 pptpd: CTRL: GRE read or PTY write failed (gre,pty)=(5,4)
 pptpd: CTRL: Client control connection finished
 pptpd: CTRL: Exiting now
 pptpd: MGR: Reaped child
 pppd : Modem hangup
 pppd : Connection terminated
 and quits.

 I tried first without ip_gre modules loaded - it still worked with
 MS apps. Loading ip_gre module did not changed the situation on
 Linux vs. Linux connections.

 Please help.
 Michael.

 --

 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]



=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: Problems with pptp client and PopTop VPN server

[Baruch Even]

In order to see the problem I suggest adding the 'debug' option to
ppp/pptp, you can do this on the client by adding debug to the pptp
command. You can also add the debug option on the pptpd server in the
pptpd.conf file.

You need to have the ip_gre module in the kernel, the pptp communication
of the LCP packets is also in the GRE channel. Make sure both sides have
the module loaded.

Make sure you don't have a firewall blocking the gre packets from
arriving to either side, a fascist firewall may need an extra rule for
this.

If it still doesn't work send the log of the pptp and ppp (both of them
with debug).

Baruch

* Michael Sternberg [EMAIL PROTECTED] [021024 11:54]:
 
 Hello
 
 I'm trying to establish PPtP VPN connection between two Linux
 boxes using pptp client and pptpd (poptop) server.
 
 The most annoying thing is that I'm able to connect to MS VPN
 server using pptp client and I'm able to connect to Linux pptpd 
 server using MS VPN connection.
 
 Problem is from Linux to Linux.
 
 I started both client and server in debug mode.
 Client starts to send LCP ConfReq in loop.
 Server prints messages saying:
 pppd : Connect: ppp0 -- /dev/ttyp1
 pptpd: GRE: read from network failed: status=-1 error=Protocol not available
 pptpd: CTRL: GRE read or PTY write failed (gre,pty)=(5,4)
 pptpd: CTRL: Client control connection finished
 pptpd: CTRL: Exiting now
 pptpd: MGR: Reaped child
 pppd : Modem hangup
 pppd : Connection terminated
 and quits.
 
 I tried first without ip_gre modules loaded - it still worked with
 MS apps. Loading ip_gre module did not changed the situation on 
 Linux vs. Linux connections.
 
   Please help.
   Michael.
 
 -- 
 
 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Baruch Even
http://baruch.ev-en.org/

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

VPN design (Was: Problems with pptp client and PopTop VPN server)

[Michael Sternberg]

On Thu, 24 Oct 2002 11:17:59 +0200
Oleg Kobets [EMAIL PROTECTED] wrote:

 maybe it has something to do with MPPE ? As far as I understood,you need
 mppe enabled pptp and some kernel module ?

No, the problem solved by loading ip_gre module on both (!) ends of
connection. Strange that I don't need it to connect to/from Windows.

Now for much more interesting question:

Let's say I have four computers (A,B,C,D) in four different networks and
want to organize a VPN between them. I can make VPN from one computer 
(A) to three others (B,C,D) and I can even use the same IP for all
three connections on originating computer (A). I can even use it as
router between all four. But - if this computer fails - all my VPN
network fail because I have a single point of failure here.

Solution that seems to be right - make additional VPN connections between
three remaining computers (maybe using same IPs as in the first three
connection) and use those connections as backup. But now I have a routing
problem. In first setup VPN IP of A was router for C,B,D to get to VPN
network. Now I have more than one path to get from D to B for example.
Perfect solution would allow switch to another route if first one failed.

What solution to the routing problem would be right ? Some kind of
dynamic routing over the VPN connection ? Something (hopefully)
simpler ?

Thanks.
Michael.

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: VPN design (Was: Problems with pptp client and PopTop VPN server)

[Baruch Even]

* Michael Sternberg [EMAIL PROTECTED] [021024 19:16]:
 Let's say I have four computers (A,B,C,D) in four different networks and
 want to organize a VPN between them. I can make VPN from one computer 
 (A) to three others (B,C,D) and I can even use the same IP for all
 three connections on originating computer (A). I can even use it as
 router between all four. But - if this computer fails - all my VPN
 network fail because I have a single point of failure here.
 
 Solution that seems to be right - make additional VPN connections between
 three remaining computers (maybe using same IPs as in the first three
 connection) and use those connections as backup. But now I have a routing
 problem. In first setup VPN IP of A was router for C,B,D to get to VPN
 network. Now I have more than one path to get from D to B for example.
 Perfect solution would allow switch to another route if first one failed.
 
 What solution to the routing problem would be right ? Some kind of
 dynamic routing over the VPN connection ? Something (hopefully)
 simpler ?

Assign a subnet for the VPN, say 192.168.69.0/24, each host has an IP in
the network .1, .2, .3 and .4

Connect all of them fully (or partially as you like), whenever a link
goes up setup a specific route to the host it is in the network and a
default route, be sure that when connecting two machines only one of
them sets a default route!

This will give you a short link when possible (one vpn hop), and some
other link if the direct one is not up.

Never tried to setup such a thing so it's mostly a guess.

Baruch

-- 
Baruch Even
http://baruch.ev-en.org/

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: [long] asking for help properly [Was: pptp kernel freeze]

[Ira Abramov]

Quoting Guy Cohen, from the post of Sat, 19 Oct:
 [blah blah blah so much crap i can cry]
 
 And if i tell you what kernel version im using would you know
 how to solve this problem? let me think about it Nope you wouldn't.

I have one too, but I'll save you the long read of the annoying exact
details... 

there may be a bug in a program that causes some version of kernel on
some machines to freeze. carefull! it might happen to you too!

(Guy, are you familliar with the concept of FUD? I'm reading this thread
and I can't figure out why would anyone try to be so unhelpful on
purpose)

-- 
Your guardian angel
Ira Abramov

http://ira.abramov.org/email/ This post is encrypted twice with ROT-13.
Documenting or attempting to crack this encryption is illegal.



msg22684/pgp0.pgp
Description: PGP signature

Re: [long] asking for help properly [Was: pptp kernel freeze]

[Nadav Har'El]

On Sat, Oct 19, 2002, Guy Cohen wrote about Re: [long] asking for help properly [Was: 
pptp kernel freeze]:
 [blah blah blah so much crap i can cry]

Do you really think that with such an attitude anyone would want to help
you now?

 And if i tell you what kernel version im using would you know
 how to solve this problem? let me think about it Nope you wouldn't.

This is a self-fulfilling prophecy...


Think about this:

ISP hotline: Acme ISP customer support, how may I help you?
customer: Hi, I can't connect to the internet - it's my Windows fault.
  What do I need to do to fix it?
hotline:  How do you know it's your windows fault? What Windows is this?
  What kind of message do you see when you try to log on?
customer: Don't you trust me? I'm telling you it's my Windows fault.
  Believe me. It always happens when I try to log in using the
  only OS I use, Windows.
hotline:  But how can I can help you fix this problem if you don't tell
  me what Windows version you have, and what error message you see
  when you try to log on? So far I have no clue if this is a known
  problem, or how to fix it, because you avoided giving me any
  details! I'm not even at a situation now where I know if it's
  a problem I can solve myself or if I need to find a more knowlegable
  support person to help you. Please tell me which version of Windows
  you are running and exactly what happens when you press that dial
  icon.
customer: blah blah blah so much crap i can cry.
  And if i tell you what Windows version im using would you know
  how to solve this problem? let me think about it Nope you
  wouldn't.
hotline:  hang up

-- 
Nadav Har'El|   Saturday, Oct 19 2002, 13 Heshvan 5763
[EMAIL PROTECTED] |-
Phone: +972-53-245868, ICQ 13349191 |Windows detected you moved your mouse.
http://nadav.harel.org.il   |Reboot for this change to take effect.

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: [long] asking for help properly [Was: pptp kernel freeze]

[Gilad Ben-Yossef]

On Sat, 2002-10-19 at 06:13, Guy Cohen wrote:
 On Sat, Oct 19, 2002 at 02:32:13AM +0200, Oleg Goldshmidt wrote:
   Latest kernel/pptp/pppd.
   no error messages, but pppd terminated before freeze. it happens to me
   on a regular besis.
  
  Flames notwithstanding this list is more often helpful than not,
  *provided* a question is asked to which an answer can be given. The
  above is an example of how *not* to ask for help on linux-il.
  
 
 Dear Oleg G.
 
 [blah blah blah so much crap i can cry]
 
Guy, I'm going to assume that you are simply ignorant as opposed to
being a regular member of the orifice set and answer you seriously. I
don't know why I am being so nice, must be the morning drugs kicking
in...

 And if i tell you what kernel version im using would you know
 how to solve this problem? let me think about it Nope you wouldn't.

Yes, he would. Or at least - there is a very good chance he would. Your
assumption that because *you* have no idea what to do means that it's
same for Oleg is false. If you give the *exact* version Oleg could have
installed the same version on a box of his own and attempt to duplicate
the problem. If he succeeded he could have install a kernel debugger and
trace what causes the malfunction and perhaps fix it. 

I've seen Oleg do this kind of job at least once and the question of
whether or not he would have done it this time is directly related to
your attitude and the respect you show for his effort and precious time.
Right now I don't think he will bother giving you the time of day and so
shall I.

 And I did not, In any place in my email said I have a problem, please 
 help me or anything similar. I simply mentioned there's a bug that 
 freeze the machine.
 
 So spear me your too long too boring Let there be guideline! flame.
 

We we're trying to be helpfull and simply requested from you the details
that we *needed in order to help you*. With this kind of attitude I
think we will 'spear' you all right - spare you from your misery and
send your posts hereafter to the forgotten limbo planes realms of our
kill files...

Gilad.

-- 
Gilad Ben-Yossef [EMAIL PROTECTED]
http://benyossef.com

 Geeks rock bands cool name #8192: RAID against the machine


=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: [long] asking for help properly [Was: pptp kernel freeze]

[Guy Cohen]

On Sat, Oct 19, 2002 at 09:21:31AM +0200, Ira Abramov wrote:
 and I can't figure out why would anyone try to be so unhelpful on
 purpose)
I'm not. I'm simply resent him being so arrogant.

Anyhow to the matter if you really want to know. It's an old p-133
running with kernel 2.4.19. pppd version 2.4.1, pptp version 1.1.0-1.
The only reason i suspect that is a bug with pppd is because 2 times
I was lucky enough to be on the logging terminal when it happened
and the last thing showed before the freeze was pppd: terminated by peer.


-- 
http://www.uadm.com |   Local and Remote Unix/Linux
[EMAIL PROTECTED]   |   Administration. Security,
Phone: +972 3 6201373   |   Installations, Support  Upgrades.

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: [long] asking for help properly [Was: pptp kernel freeze]

[Guy Cohen]

so then again Gilad, spear me the time for your long flame.
Will you? Not likely, but i have to try.

On Sat, Oct 19, 2002 at 11:30:50AM +0200, Gilad Ben-Yossef wrote:
 On Sat, 2002-10-19 at 06:13, Guy Cohen wrote:
  On Sat, Oct 19, 2002 at 02:32:13AM +0200, Oleg Goldshmidt wrote:
Latest kernel/pptp/pppd.
no error messages, but pppd terminated before freeze. it happens to me
on a regular besis.
   
   Flames notwithstanding this list is more often helpful than not,
   *provided* a question is asked to which an answer can be given. The
   above is an example of how *not* to ask for help on linux-il.
   
  
  Dear Oleg G.
  
  [blah blah blah so much crap i can cry]
  
 Guy, I'm going to assume that you are simply ignorant as opposed to
 being a regular member of the orifice set and answer you seriously. I
 don't know why I am being so nice, must be the morning drugs kicking
 in...
 
  And if i tell you what kernel version im using would you know
  how to solve this problem? let me think about it Nope you wouldn't.
 
 Yes, he would. Or at least - there is a very good chance he would. Your
 assumption that because *you* have no idea what to do means that it's
 same for Oleg is false. If you give the *exact* version Oleg could have
 installed the same version on a box of his own and attempt to duplicate
 the problem. If he succeeded he could have install a kernel debugger and
 trace what causes the malfunction and perhaps fix it. 
 
 I've seen Oleg do this kind of job at least once and the question of
 whether or not he would have done it this time is directly related to
 your attitude and the respect you show for his effort and precious time.
 Right now I don't think he will bother giving you the time of day and so
 shall I.
 
  And I did not, In any place in my email said I have a problem, please 
  help me or anything similar. I simply mentioned there's a bug that 
  freeze the machine.
  
  So spear me your too long too boring Let there be guideline! flame.
  
 
 We we're trying to be helpfull and simply requested from you the details
 that we *needed in order to help you*. With this kind of attitude I
 think we will 'spear' you all right - spare you from your misery and
 send your posts hereafter to the forgotten limbo planes realms of our
 kill files...
 
 Gilad.
 
 -- 
 Gilad Ben-Yossef [EMAIL PROTECTED]
 http://benyossef.com
 
  Geeks rock bands cool name #8192: RAID against the machine

-- 
http://www.uadm.com |   Local and Remote Unix/Linux
[EMAIL PROTECTED]   |   Administration. Security,
Phone: +972 3 6201373   |   Installations, Support  Upgrades.

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: [long] asking for help properly [Was: pptp kernel freeze]

[Ira Abramov]

Quoting Gilad Ben-Yossef, from the post of Sat, 19 Oct:
 Right now I don't think he will bother giving you the time of day and
 so shall I.

much ado about a dead issue. I think I summed it up 3 messages ago. I
now regret forwarding Guy's CV to our recruiter, since he's obviously a
lone wolf and not a team player. As he said, he just wanted to post a
me too and not actually ask for a solution, and we're the big sucksers
here that thought that if someone posts to linux-il they need our help.
Guy must have missed the FAQs of linux-il and the basic idea of what a
community in general is, and the OSS community in particular.

now can we move on to more interesting subjects?  please? :)

for instance, how come whenever I dump a CD to my machine (with dd into
an iso file) it always dies 1-2 kilobytes befor the end of the image
saying read error?
happens with 3 different drives (SCSI and IDE), and the burnt image size
is anywhere from 50-700 meg. both off CD-roms and CD-Rs.

Kernel 2.4.18, 19, and for all I remember a few more versions back too.

(feel free to change the subject header if you have an answer)


-- 
Naked Chef
Ira Abramov

http://ira.abramov.org/email/ This post is encrypted twice with ROT-13.
Documenting or attempting to crack this encryption is illegal.



msg22689/pgp0.pgp
Description: PGP signature

Re: [long] asking for help properly [Was: pptp kernel freeze]

[Oleg Kobets]

 I've seen Oleg do this kind of job at least once and the question of
 whether or not he would have done it this time is directly related to
 your attitude and the respect you show for his effort and precious time.
 Right now I don't think he will bother giving you the time of day and so
 shall I.

Gilad, actually I posted this problem to the list and not Guy, so please
don't ignore me in the process of flaming Guy. You may flame, or you may
help, as I don't see any wrong about what Guy wrote, he just agreed with me
that the problem occurs as well and then I posted my versions (kernel, pptp,
debian) and never got any constructive answer, just flames for Guy.

People, what's wrong with you ? Are you so stupid and ignorant that you are
willing to flame instead of help ?!?!? I noticed this on several occurences
now, there is always some member that get up from the wrong foot and starts
expressing his displeasure about some noncence that nobody cares about.

If anybody (still) knows how to fix this problem of pptp kernel freeze,
please write to me directly as not to allow this list the opportunity to
flame anyone else.

Oleg Kobets.



=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

pptp kernel freeze v2

[Oleg Kobets]

I already supplied the relevant details per Nadav request.

Kernel 2.4.18 vanilla (ie unpatched)
Debian 3.0 testing
pptp 1.1.0
pppd 2.4.1

Kernel compiled with gcc 2.95 (not 3.2), everything is NOT modules but
compiled in the kernel. It is worth mentioning that it is Debian Kernel,
that means that as I have not applied patches to it, Debian folks did. The
problem occurs only when I d/l large files at top speed, but perhaps it's
only coincidence, as Ira said that pppd craches before the kernel freeze.
Also Ira said that he experienced the problem with 2.4.19 kernel as well.

Unfortunatelly I cannot supply the kernel oops info as I already rebooted
the machine and can't see it. Logs shows nothing. Looking in the pptp
mailing list and site proved unhelpful and I want this list opinion before I
write to both pptp and kernel mailing lists.

- Original Message -
From: Tzafrir Cohen [EMAIL PROTECTED]
To: Oleg Kobets [EMAIL PROTECTED]
Sent: Saturday, October 19, 2002 12:39 PM
Subject: Re: [long] asking for help properly [Was: pptp kernel freeze]


 On Sat, 19 Oct 2002, Oleg Kobets wrote:

  If anybody (still) knows howto fix this problem of pptp kernel freeze,
  please write to me directly as not to allow this list the opportunity to
  flame anyone else.

 I think people are still waiting for the details of your problems.

 Please read what Oleg Goldshmidt wrote, and supply those details (or
 similar relevant details). There are potentially many things that can
 cause a kernel problem. Maybe it is an interaction between some two
 modules? MAybe it is some hardware-related problem?

 The oops text is espcially useful.
 --
 Tzafrir Cohen
 mailto:tzafrir;technion.ac.il
 http://www.technion.ac.il/~tzafrir






=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: pptp kernel freeze v2

[Gilad Ben-Yossef]

On Sat, 2002-10-19 at 13:31, Oleg Kobets wrote:
 Kernel compiled with gcc 2.95 (not 3.2), everything is NOT modules but
 compiled in the kernel. 


From the pptp web site for the latest version:

Ensure that the configuration builds PPP support as modules. We've
found that static compilation results in pppd being unable to load MPPE.
Set CONFIG_PPP=m in .config.

As of August 2002, kernel-patch-mppe is in testing and unstable, not in
stable. It was dropped from stable due to installation bugs.

Whether this does or does not relveant is unknown to me.

Hope this helps,
Gilad.





-- 
Gilad Ben-Yossef [EMAIL PROTECTED]
http://benyossef.com

 Geeks rock bands cool name #8192: RAID against the machine


=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: [long] asking for help properly [Was: pptp kernel freeze]

[Amir Tal]

On Saturday 19 October 2002 11:52, Guy Cohen wrote:
 On Sat, Oct 19, 2002 at 09:21:31AM +0200, Ira Abramov wrote:
  and I can't figure out why would anyone try to be so unhelpful on
  purpose)

 I'm not. I'm simply resent him being so arrogant.

 Anyhow to the matter if you really want to know.

after the way you were talking to people here, no one *really* wants to know 
anything about your problem anymore.
either start behaving, or shut the hell up !
no one here works for you.

tal.


 It's an old p-133
 running with kernel 2.4.19. pppd version 2.4.1, pptp version 1.1.0-1.
 The only reason i suspect that is a bug with pppd is because 2 times
 I was lucky enough to be on the logging terminal when it happened
 and the last thing showed before the freeze was pppd: terminated by peer.


=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: [long] asking for help properly [Was: pptp kernel freeze]

[Guy Cohen]

What the heck are you blubing about again? You want to be part of 
the team, the flaming team? well, you are. Good for you.
What a dream team.

now jump out the window please.

And btw, I wasn't talking to poeple. I was talking to one person.


On Sat, Oct 19, 2002 at 05:22:51PM +0200, Amir Tal wrote:
 On Saturday 19 October 2002 11:52, Guy Cohen wrote:
  On Sat, Oct 19, 2002 at 09:21:31AM +0200, Ira Abramov wrote:
   and I can't figure out why would anyone try to be so unhelpful on
   purpose)
 
  I'm not. I'm simply resent him being so arrogant.
 
  Anyhow to the matter if you really want to know.
 
 after the way you were talking to people here, no one *really* wants to know 
 anything about your problem anymore.
 either start behaving, or shut the hell up !
 no one here works for you.
 
 tal.
 
 
  It's an old p-133
  running with kernel 2.4.19. pppd version 2.4.1, pptp version 1.1.0-1.
  The only reason i suspect that is a bug with pppd is because 2 times
  I was lucky enough to be on the logging terminal when it happened
  and the last thing showed before the freeze was pppd: terminated by peer.

-- 
http://www.uadm.com |   Local and Remote Unix/Linux
[EMAIL PROTECTED]   |   Administration. Security,
Phone: +972 3 6201373   |   Installations, Support  Upgrades.

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: [long] asking for help properly [Was: pptp kernel freeze]

[Amir Tal]

On Saturday 30 March 2002 23:13, you wrote:
 What the heck are you blubing about again? You want to be part of
 the team, the flaming team? well, you are. Good for you.
 What a dream team.

 now jump out the window please.

after you.


 And btw, I wasn't talking to poeple. I was talking to one person.


you did it in public, you did it on this list, so anyone subscribed to it has 
a right to say something about it.
you wanna get pisseed at someone ?? do it offlist.


 On Sat, Oct 19, 2002 at 05:22:51PM +0200, Amir Tal wrote:
  On Saturday 19 October 2002 11:52, Guy Cohen wrote:
   On Sat, Oct 19, 2002 at 09:21:31AM +0200, Ira Abramov wrote:
and I can't figure out why would anyone try to be so unhelpful on
purpose)
  
   I'm not. I'm simply resent him being so arrogant.
  
   Anyhow to the matter if you really want to know.
 
  after the way you were talking to people here, no one *really* wants to
  know anything about your problem anymore.
  either start behaving, or shut the hell up !
  no one here works for you.
 
  tal.
 
 
   It's an old p-133
 
   running with kernel 2.4.19. pppd version 2.4.1, pptp version 1.1.0-1.
   The only reason i suspect that is a bug with pppd is because 2 times
   I was lucky enough to be on the logging terminal when it happened
   and the last thing showed before the freeze was pppd: terminated by
   peer.


=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: pptp kernel freeze v2

[Robert Wallner]

Yo list,

I don't know if you ppl don't already know this, but the so called alcatel 
modems are in fact routers with some features disabled. I was fiddling with 
pptp since bezeq started the adsl service, and ate a lot of crap from it, 
till I found this thing. Since imho the majority of modems are alcatel, they 
can be switched to pro mode by changing 2 bytes in the modem. Details can 
be found at http://modem.co.il/alcatel/alcatel.html (cp1255).
I'm _not_ using pptp for more than a year now, and I'm happy.

Regards,
Robert Wallner

To unsubscribe, send 
mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: pptp kernel freeze v2

[Ira Abramov]

Quoting Robert Wallner, from the post of Sat, 19 Oct:
 I don't know if you ppl don't already know this, but the so called
 alcatel modems are in fact routers with some features disabled. I

no news, we had the howto published on the list a few months back. the
one thing you forgot to mention is that the Pro mode makes it much
more vulnerable to carcking from the outside, since it gets a real IP
all of a sudden.

another option (if you are rich) is to get an S-Box (www.s-box.com), or
if you are poor, a 4 port 10/100 hub and pptp router from linksys:
http://linksys.com/Products/product.asp?grid=23prid=20
and other such neat router/packet filter/switch combos that are $45-$80
on Amazon and other places (search DSL router). some of them even have
IPSEC and can connect you to the Office CP FW-1 based VPN server.

in any case, Alcatel, is NOT a very good idea, security-wise.

-- 
Don Quixote and Abraham Lincoln rolled into one
Ira Abramov

http://ira.abramov.org/email/ This post is encrypted twice with ROT-13.
Documenting or attempting to crack this encryption is illegal.



msg22703/pgp0.pgp
Description: PGP signature

Re: [long] asking for help properly [Was: pptp kernel freeze]

[Omer Zak]

On 19 Oct 2002, Oleg Goldshmidt wrote:

 http://www.iglu.org.il/mailing-lists/linux-il.html
 
 If none exist, I am willing to write (and maintain) one in my copious
 spare time [NB: the usual meaning of this is never, cf. the Jargon
 File; here I really mean I am willing to do it if it is needed, but be
 patient].  If nothing else, it will give everyone a warm feeling of
 righteousness when flogging an offender with a differential SCSI
 cable.
 
 If you didn't read all the way through to here, then obviously you
 think we don't need detailed posting guidelines. If you did, let me
 know if you think we do. I won't attempt to do anything unless I am
 convinced there is an itch. A seven-year itch, considering the age 
 of linux-il. ;-)

I remember such a FAQ.
My suggestion is to check how the original poster missed the gnubies-il
mailing list (this is the mailing list, which was created for and
tolerates questions like Please help me!  All LEDs in my PC turned off
and the screen went black!! What to do now???) 

 --- Omer
WARNING TO SPAMMERS:  see at http://www.zak.co.il/spamwarning.html


=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

RE: pptp kernel freeze v2

[Tzahi Fadida]

I must disagree with the degree of certainty of your conclusion.
Alcatel security has little to do with the security of the machine
running behind it, since when alcatel don't get an ip, the machine gets
the ip.
i.e. the machine, which is where the data is, is vulnerable to internet
hazards. setting up a firewall, and IDS if u r running a public service
is the only way to enhance the security of your server.
so what if ur modem gets hacked. it can't bypass the firewall because
the Linux firewall, when you choose a gateway, should in default define
the gateway to be separated from the computers behind it.
a. ok now lets hypothesize that your modem gets hacked on the pro mode.
using the above assumption as the reality of the Linux configuration,
the modem will in effect be just another computer on the network.
b. lets now return to the pptp session configuration. if somehow the
hacker hacked the modem, he could potentially find a way to piggy bag on
the pptp session on which case the firewall have no protection against.
by piggy bag
i mean that a bug in the modem would allow to initiate a telnet session
as comming from inside the pppX of your machine. i.e. by not changing
anything in the code or adding to it.

conclusion: you are in effect have a lesser degree of security with the
pptp session. also please note that pptp is an encapsulation protocol,
and tcp over ppp is not, that is, if you reject all traffic originating
from the modem you are more safe then accepting pptp traffic from the
modem.

That said, it could be that there is a way to install some sniffer+tcp
injection code+packet router inside the modem, but I have my doubts.

nb: i use pptp and have alcatel. why?, lazy + don't care + i pitty the
guy/girl (lets be politicaly correct) who would want to hack my p90
machine who can't be sure what time of day is it. i would put it in a
senior home, but i don't have the heart.

* - * - *
Tzahi Fadida
[EMAIL PROTECTED]
Technion Email: [EMAIL PROTECTED]
My Cool Site: HTTP://WWW.My2Nis.Com
* - * - * - * - * - * - * - * - * - *

WARNING TO SPAMMERS:  see at
http://members.lycos.co.uk/my2nis/spamwarning.html

 -Original Message-
 From: [EMAIL PROTECTED]
 [mailto:linux-il-bounce;cs.huji.ac.il]On Behalf Of Ira Abramov
 Sent: Sunday, October 20, 2002 12:19 AM
 To: Linux-IL
 Subject: Re: pptp kernel freeze v2


 Quoting Robert Wallner, from the post of Sat, 19 Oct:
  I don't know if you ppl don't already know this, but the so called
  alcatel modems are in fact routers with some features disabled. I

 no news, we had the howto published on the list a few
 months back. the
 one thing you forgot to mention is that the Pro mode makes it much
 more vulnerable to carcking from the outside, since it gets a real IP
 all of a sudden

 another option (if you are rich) is to get an S-Box
(www.s-box.com), or
if you are poor, a 4 port 10/100 hub and pptp router from linksys:
http://linksys.com/Products/product.asp?grid=23prid=20
and other such neat router/packet filter/switch combos that are $45-$80
on Amazon and other places (search DSL router). some of them even have
IPSEC and can connect you to the Office CP FW-1 based VPN server

in any case, Alcatel, is NOT a very good idea, security-wise

--
Don Quixote and Abraham Lincoln rolled into one
Ira Abramov

http://ira.abramov.org/email/ This post is encrypted twice with ROT-13
Documenting or attempting to crack this encryption is illegal



=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

pptp kernel freeze

[Oleg Kobets]

Hi!

I had a really weird occurence yesterday. My server 
froze with kernel freeze and the message stated that what froze it was pptp. It 
was not the first time it happened, so I am a little worried here.

Does anyone has such experience or knows how to fix 
it ?
---Oleg KobetsNetwork 
Administratorwww.clean-mail.net

Re: pptp kernel freeze

[Guy Cohen]

Yup, had that too. My only guess is that is either a bug with pptp, pppd, 
or the kernel ;)

On Fri, Oct 18, 2002 at 09:32:24PM +0200, Oleg Kobets wrote:
 Hi!
 
 I had a really weird occurence yesterday. My server froze with kernel freeze and the 
message stated that what froze it was pptp. It was not the first time it happened, so 
I am a little worried here.
 
 Does anyone has such experience or knows how to fix it ?
 
 ---
 Oleg Kobets
 Network Administrator
 www.clean-mail.net

-- 
http://www.uadm.com |   Local and Remote Unix/Linux
[EMAIL PROTECTED]   |   Administration. Security,
Phone: +972 3 6201373   |   Installations, Support  Upgrades.

=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: pptp kernel freeze

[Oleg Kobets]

kernel 2.4.18 vanilla
debian testing
pptp 1.1.0 (or debian package name 1.1.0-2)

You know it's pptp's fault because your computer freezes with kernel debug
info and what module caused the freeze. Trust me, it's pptp.

Also I noticed a pattern, it freezes when there is a high load on network
(in both times I was d/l 700Mb files with max speed). Perhaps my machine
can't take the load ? It's celeron 800 with 390 ram.

- Original Message -
From: Nadav Har'El [EMAIL PROTECTED]
To: Oleg Kobets [EMAIL PROTECTED]
Cc: Linux-IL [EMAIL PROTECTED]
Sent: Friday, October 18, 2002 10:13 PM
Subject: Re: pptp kernel freeze


 On Fri, Oct 18, 2002, Oleg Kobets wrote about pptp kernel freeze:
  I had a really weird occurence yesterday. My server froze with kernel
freeze and the message stated that what froze it was pptp. It was not the
first time it happened, so I am a little worried here.
 
  Does anyone has such experience or knows how to fix it ?

 Sorry, I cannot help you. But in situations like these, mentioning the
 version of the kernel you are running pptp, and/or your distribution
 can be helpful. So is writing here the exact message that led you to
belive
 it was a kernel freeze and that it was pptp's fault.

 --
 Nadav Har'El| Friday, Oct 18 2002, 13 Heshvan
5763
 [EMAIL PROTECTED]
|-
 Phone: +972-53-245868, ICQ 13349191 |If you lost your left arm, your right
arm
 http://nadav.harel.org.il   |would be left.





=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Re: pptp kernel freeze

[Gilad Ben-Yossef]

On Sat, 2002-10-19 at 00:40, Oleg Kobets wrote:
 kernel 2.4.18 vanilla

Have you tried it with a more recent kernel? Perhaps it the problem is
already solved?

 debian testing
 pptp 1.1.0 (or debian package name 1.1.0-2)
 
 You know it's pptp's fault because your computer freezes with kernel debug
 info and what module caused the freeze. Trust me, it's pptp.

We trust you but we can't help you unless you publish the dump. If what
your descirption is refering to the message commonly known as a 'kernel
oopss' message, I would advise to put it though the ksymsopps utility.

 
 Also I noticed a pattern, it freezes when there is a high load on network
 (in both times I was d/l 700Mb files with max speed). Perhaps my machine
 can't take the load ? It's celeron 800 with 390 ram.

Freezing the machine is not an acceptable behaviour at any event.


 
 - Original Message -
 From: Nadav Har'El [EMAIL PROTECTED]
 To: Oleg Kobets [EMAIL PROTECTED]
 Cc: Linux-IL [EMAIL PROTECTED]
 Sent: Friday, October 18, 2002 10:13 PM
 Subject: Re: pptp kernel freeze
 
 
  On Fri, Oct 18, 2002, Oleg Kobets wrote about pptp kernel freeze:
   I had a really weird occurence yesterday. My server froze with kernel
 freeze and the message stated that what froze it was pptp. It was not the
 first time it happened, so I am a little worried here.
  
   Does anyone has such experience or knows how to fix it ?
 
  Sorry, I cannot help you. But in situations like these, mentioning the
  version of the kernel you are running pptp, and/or your distribution
  can be helpful. So is writing here the exact message that led you to
 belive
  it was a kernel freeze and that it was pptp's fault.
 
  --
  Nadav Har'El| Friday, Oct 18 2002, 13 Heshvan
 5763
  [EMAIL PROTECTED]
 |-
  Phone: +972-53-245868, ICQ 13349191 |If you lost your left arm, your right
 arm
  http://nadav.harel.org.il   |would be left.
 
 
 
 
 
 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]
 
-- 
Gilad Ben-Yossef [EMAIL PROTECTED]
http://benyossef.com

 Geeks rock bands cool name #8192: RAID against the machine


=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

[long] asking for help properly [Was: pptp kernel freeze]

[Oleg Goldshmidt]

Guy Cohen [EMAIL PROTECTED] writes:

  Sorry, I cannot help you. But in situations like these, mentioning the
  version of the kernel you are running pptp, and/or your distribution
  can be helpful. So is writing here the exact message that led you to belive
  it was a kernel freeze and that it was pptp's fault.
 
 Latest kernel/pptp/pppd.
 no error messages, but pppd terminated before freeze. it happens to me
 on a regular besis.

Flames notwithstanding this list is more often helpful than not,
*provided* a question is asked to which an answer can be given. The
above is an example of how *not* to ask for help on linux-il.

Saying latest kernel does not help. Saying No error messages does
not help either. Latest kernel may be the latest Marcello's stable
2.4, or the latest cutting edge 2.5, or the latest Red Hat, or one of
a baker's dozen other latest versions. Same for pptp/pppd - latest
official / patched (e.g. with Bezeq-specific stuff) / rpm? Latest Red
Hat rpm and latest PLD rpm are often different, are built from
different versions of the software, etc.

Please tell us your distribution, the details of the kernel (whether
it comes from the distribution out of the box, or was custom
configured and/or compiled - provide the configuration details in the
latter case - etc), what modules were loaded, what hardware you are
using, all the relevant details.

As an example, reporting a pptp/pppd problem I would at least report
that I run fully up2dated RH7.2 with the latest RH kernel,

bash-2.05$ rpm -q kernel ppp pptp-linux
kernel-2.4.18-17.7.x
ppp-2.4.1-3
pptp-linux-1.1.0-1

- this gives the full info of what version of the software you are
running. Better yet include the output of uname -r - this will tell
us what kernel you are actually running as opposed to which kernel you
think you are running (I was bitten by this myself once or twice -
trust me). That the computer is a 800 MHz Intel PIII Coppermine with
256MB of RAM may or may not be relevant - if you don't know what the
problem is, do include the info.

I would also include the output of /sbin/lsmod - *most* of the things
there are directly relevant in my case, such as the various modules
related to ppp, iptables, NIC drivers, etc. I'll spare you the actual
output, but I count 16 relevant lines out of 23 in the output of lsmod
on my machine - and some of the irrelevant 7 are due to the weird fact
that I hooked a printer to the computer that is my gateway to the
world, for physical location reasons in my apartment, so 80% or more
of lsmod can be relevant. Note that this info will provide the makes of
my NICs and the drivers I am using, which may be important.

Please include the relevant portion of the syslog output - there must
be something there if pptp/pppd/kernel barf in any way - better yet
(if the problem is reproducible) after the problem happens when
pptp/pppd run with their debugging options on. Do specify how
*exactly* pptp/pppd were invoked. A lot of useful info may exist in
/proc, but if you are not an expert then likely you will be guided
through that if needed after you post.

Include the exit codes - they may give us clues even if there is no
explicit error message. 

Tell us what you have tried yourself to determine what the problem was
and/or to fix it.

In general, please, when you ask for help, do provide the information
needed. Saying I have a problem, please help me is nowhere close to
sufficient. You either need to provide enough info for someone really
knowledgeable to be able to identify the problem and maybe suggest a
solution *quickly*, or pay enough money to a consultant to find all
the details out himself/herself on site.

Finally, if you just try to compile a decent problem description, it
may suddenly become clear to you, and maybe you won't need to write to
the list at all, and if you still do your chances of getting help will
be much higher. I am not saying that you didn't try to solve the
problem, I am only saying that a) I didn't see any evidence you tried,
b) I cannot offer any help at all given the lack of details, c) I am
not at all sure I will be able to help given all the details, but
maybe Nadav or someone else will - right now no one can.

There used to be a FAQ or guidelines that explained both how to post
proper questions and when a poster deserved flogging with a
differential SCSI cable and when applying a parallel port replicator 
to the poster's temple was in order. It was all spelled out in clear
Anglo-Saxon. Is it gone? Did I see it in the Mirror of Erised? 

In any case, I think such guidelines should be prominent on the IGLU
site and the linux-il link,

http://www.iglu.org.il/mailing-lists/linux-il.html

If none exist, I am willing to write (and maintain) one in my copious
spare time [NB: the usual meaning of this is never, cf. the Jargon
File; here I really mean I am willing to do it if it is needed, but be
patient].  If nothing else, it will give everyone a warm feeling of
righteousness when