Linux-Networking Digest #831, Volume #10 Mon, 12 Apr 99 12:13:40 EDT
Contents:
2.2 kernel, diskless clients, chicken/egg problem? (Nils Ulltveit-Moe)
Re: multiple ethernet cards ("Curt")
Re: Linux Networking Performance Question? (Marco BANO)
pppd demand dialing - ioctl(SIOCSIFDSTADDR): Cannot assign requested address(99)
(Nick Lucent)
Re: OSPF on Linux ("M.C. van den Bovenkamp")
Samba Baltic letters problem (Mart Pirita)
Re: setting up apache (Kwagga)
Re: Need help with PPP... (Andras Gefferth)
ppp/pap for planet-interkom account ("Christian Kaiser")
pppd quits with no output to console or modem (Matthew King)
Re: Machine name themes - what do you use? (J.D. Baldwin)
Re: yppasswd not working? ([EMAIL PROTECTED])
Help. Unable to browse HTTPS from IP Masq'ed workstations ("Mike Ellis")
automount/mount does not react (Joachim Gloger)
----------------------------------------------------------------------------
Crossposted-To: comp.os.linux.development.system
Subject: 2.2 kernel, diskless clients, chicken/egg problem?
From: Nils Ulltveit-Moe <[EMAIL PROTECTED]>
Date: 12 Apr 1999 12:30:10 +0200
Have anyone managed to make NFS-root work on diskless clients running
a 2.2 kernel?
NFS root works OK up to kernel 2.0.36, but I have not yet managed to get
it working on 2.2.X kernels.
The problem now is that RPC seems not to work... After searching
newsgroups it seems like the 2.2 kernel may require to have the local
portmapper running for RPC services to work. (I cannot imagine for what
reason.. It should suffice to contact the remote portmapper.)
If this is the case, the 2.2 kernel has added a
"chicken/egg" problem since the root nfs code may require the local
portmapper running before mounting the root file system, but that is
impossible because the portmapper should be loaded from the root file
system via NFS..
At first I tried to patch ipconfig.c (thanks to Bodo Bellut
[EMAIL PROTECTED]) to make the kernel being able to mount the root
FS via NFS:
--- linux/net/ipv4/ipconfig.c.OLD Sun Feb 28 23:36:50 1999
+++ linux/net/ipv4/ipconfig.c Mon Mar 1 00:39:32 1999
@@ -812,6 +812,9 @@
return 0;
DBG(("IP-Config: Entered.\n"));
+
+ if (root_server_addr == INADDR_NONE)
+ root_server_addr = ic_servaddr;
/* Setup all network devices */
if (ic_open_devs() < 0)
With this patch the client attempts to mount the root FS via NFS:
Root-NFS: Mounting /tftpboot/rootfs/skaalen.linux on server
193.180.211.70 as root
Root-NFS: rsize=4096, wsize=4096, timeout=7, retrans=3
Root-NFS: acreg(min,max)=(3.60), acdir(min,max)=(30,60)
Root-NFS: nfsd port=-1, mountd port=0, flags=00000200
Looking up port of RPC 100003/2 on 193.180.211.70
RPC: sendmsg returned error 101 (ENETUNREACH)
RPC: sendmsg returned error 101 (ENETUNREACH)
RPC: sendmsg returned error 101 (ENETUNREACH)
RPC: sendmsg returned error 101 (ENETUNREACH)
RPC: sendmsg returned error 101 (ENETUNREACH)
Portmap: Server 193.180.211.70 not responding, timed out.
The portmapper IS running and has rpc.mountd/rpc.nfsd running:
eldhus:/usr/home/etonumo# ps -ax | egrep -v grep | egrep "rpc"
25794 ? S 0:00 rpc.mountd
25796 ? S 2:31 rpc.nfsd
104 ? S 0:01 /usr/sbin/rpc.portmap
Programs are registered at portmapper:
eldhus:/usr/home/etonumo# rpcinfo -p eldhus
program vers proto port
100000 2 tcp 111 portmapper
100000 2 udp 111 portmapper
100005 1 udp 954 mountd
100005 1 tcp 956 mountd
100003 2 udp 2049 nfs
100003 2 tcp 2049 nfs
Mount works fine from another machine:
numopc:/usr/src/linux/fs/nfs# mount eldhus:/tftpboot/rootfs/skaalen.linux /mnt
numopc:/usr/src/linux/fs/nfs# ls /mnt
System.map dev lib proc server var
bin etc mnt root tmp
bootImage home opt sbin usr
Please tell me if you have any clue of how to fix the problem.
(e.g. if there are patches floating around for it..)
Regards,
Nils Ulltveit-Moe
------------------------------
From: "Curt" <[EMAIL PROTECTED]>
Subject: Re: multiple ethernet cards
Date: Mon, 12 Apr 1999 05:36:41 -0500
Do you have a route to each subnet defined?
for example:
route add -net 192.168.0.0 netmask 255.255.255.0 eth0
route add -net 192.168.1.0 netmask 255.255.255.0 eth1
run 'route' or 'netstat -rn' to see if the route is there to both subnets
_ wrote in message <[EMAIL PROTECTED]>...
>HELP! Can anyone offer some suggestions on how to get 2 ethernet cards
>working on 2 different subnets? I followed the multiple-Ethernet how-to
and
>added the proper line to lilo.conf. Both cards are seen by the kernel on
>boot up
>
>kernel: wd.c:v1.10 9/23/94 Donald Becker ([EMAIL PROTECTED])
>kernel: eth0: WD80x3 at 0x300, 00 00 C0 85 48 57 WD8013, IRQ 10, shared
>memory at 0xcc000-0xcffff.
>kernel: eth1: WD80x3 at 0x240, 00 00 C0 B0 5B 2D WD8013, IRQ 7, shared
>memory at 0xd0000-0xd3fff.
>
>This seems to work correctly, and i don't have any other conflicts. This
>computer also has ip masq enabled (this does work). eth0 can connect to
>other computers but eth1 seems to do nothing. I can't ping any computer
>connected to it. if i change eth1 to the default device by using "route
add
>default eth1" I can't connect to anything, and my ppp device stops working.
>Am i missing something really simple?
>
>Oh, i'm using slackware 3.6 with 2.0.36
>
>please reply to [EMAIL PROTECTED], i'll repost a solution if this ever gets
worked
>out ;)
>
>output from ifconfig looks like
>
>lo Link encap:Local Loopback
> inet addr:127.0.0.1 Bcast:127.255.255.255 Mask:255.0.0.0
> UP BROADCAST LOOPBACK RUNNING MTU:3584 Metric:1
> RX packets:4999 errors:0 dropped:0 overruns:0 frame:0
> TX packets:4999 errors:0 dropped:0 overruns:0 carrier:0
> Collisions:0
>
>eth0 Link encap:Ethernet HWaddr 00:00:C0:85:48:57
> inet addr:192.168.1.1 Bcast:192.168.1.255 Mask:255.255.255.0
> UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
> RX packets:159385 errors:0 dropped:24 overruns:0 frame:6
> TX packets:162811 errors:0 dropped:0 overruns:0 carrier:0
> Collisions:178
> Interrupt:10 Base address:0x310 Memory:cc000-d0000
>
>eth1 Link encap:Ethernet HWaddr 00:00:C0:B0:5B:2D
> inet addr:131.151.139.1 Bcast:131.151.139.255 Mask:255.255.0.0
> UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
> RX packets:0 errors:0 dropped:0 overruns:0 frame:0
> TX packets:40 errors:0 dropped:0 overruns:0 carrier:0
> Collisions:0
> Interrupt:7 Base address:0x250 Memory:50000-54000
>
>ppp0 Link encap:Point-to-Point Protocol
> inet addr:131.151.64.202 P-t-P:131.151.64.154 Mask:255.255.0.0
> UP POINTOPOINT RUNNING MTU:1500 Metric:1
> RX packets:16179 errors:1 dropped:1 overruns:0 frame:0
> TX packets:16113 errors:0 dropped:0 overruns:0 carrier:0
> Collisions:0
> Memory:bfc038-bfcc04
>
>
>
>
>
------------------------------
From: Marco BANO <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware,comp.periphs.scsi
Subject: Re: Linux Networking Performance Question?
Date: Mon, 12 Apr 1999 12:00:20 +0200
Reply-To: [EMAIL PROTECTED]
Chris Pitzel wrote:
> > It would cost us over �10,000 pounds to upgrade the HP9000 hard disks and
> > controller to Ultra2 SCSI (80Mbs) and have considered storing the data on
>
> One thing to consider is the fact that your existing 20mb/sec
> controllers on the HP9000 probably aren't even being utilized to their
> fullest capacity. Modern hard drives only put out around 4-5mb/sec each
> at the most, and it's unlikely that all of the drives in your HP 9000
> are active at any given time.
>
> There are other considerations, of course, with the HP9000, such as the
> fact that you likely will have to buy 'HP-brand' drives (which usually
> are just Seagate Barracudas anyways..), and of course, you will need to
> consider your backup needs as well.
>
> > linux boxes over the 100BaseT network.
> > Our primary data is UniData database files. A linux box with Ultra2 drives
> > and controller can be purchased for under �2000 with increased storage
> > space.
>
> Yes, or you could just add some Ultra2 drives to the HP 9000 machine.
> Since SCSI is forwards and backwards compatible, you should be able to
> do this without any problems whatsoever. The fact that the HP
> controller and existing drives only support 20mb/sec isn't really an
> issue, considering that you're dealing with database files which are not
> terribly taxing on the bandwidth of your SCSI controller.
>
> > 1) What would the performance increase be like if I were to go for this
> > method, considering our current local HP drives are rated at 20Mbs.
>
> You wouldn't see a performance increase. Your HP machine has plenty of
> capacity for additional drives.
>
> > 2) Where would the primary bottleneck be.
>
> Unless the HP9000 is multihomed (ie: fitted with multiple network
> cards), the network card would be busy doing the NFS traffic from the
> Linux box, and would be impaired in it's capability to serve the network
> it's intended to serve.
>
> With 100mb/sec ethernet, you only get around 10mb/sec of effective
> capacity, if even. If you're doing some heavy work on Linux mounted
> SCSI drives that you propose, that could nearly eat up all the
> bandwidth, leaving none for your real clients to use.
>
> > 3) As the network is switched, would there be a degredation in network
> > performance.
>
> Definitely, due to saturation of the network interface on the HP machine
> itself, not to mention the increased load on the switch, and
> corresponding decrease in reliability as well (because when you add an
> extra machine, you've just established another point of failure..).
>
> > Any tips or information would be most appreciated.
>
> I would look into fitting Ultra2 drives onto your existing 20mb/sec
> controller. You won't see any performance degradation.
Sorry for the intrusion...
But Are you sure that you could add Ultra2 drives on a 20mb/sec controller ? it
depends on which type of controller you have . I have in a D350 a FWD, so I need
differential (high voltage) disks . isn't it ?
I wish I could add "normal Ultra2" drives (for investiment.. save).
Please drop me a line ... to help me.
--
Marco BANO
Network administrator Consultant
EUMETSAT
Am Kavalleriesand 31
64295 DARMSTADT
Germany
Office : ++49 6151 807536
------------------------------
From: [EMAIL PROTECTED] (Nick Lucent)
Subject: pppd demand dialing - ioctl(SIOCSIFDSTADDR): Cannot assign requested
address(99)
Date: 12 Apr 1999 10:49:16 GMT
Reply-To: [EMAIL PROTECTED]
Im trying to get pppd (2.3.7, kernel 2.2.5) to dial out on demand. I added
"demand :168.121.1.1 idle 600 holdoff 20 ipcp-accept-local" to my dialout,
and now when i run pon I get this error in my logs, and pppd dies.
ioctl(SIOCSIFDSTADDR): Cannot assign requested address(99)
I get this error no matter what IP I assign to the remote server. i searched
on dejanews, google, and astalavista looking for a solution, but all i found
was to add the ipcp-accept-local line, and echo 1 >
/proc/sys/net/ipv4/ip_dynaddr.
If anyone has seen this problem before or possibly has a solution I would
love to here it.
Nick
[EMAIL PROTECTED] (reverse)
------------------------------
From: "M.C. van den Bovenkamp" <[EMAIL PROTECTED]>
Subject: Re: OSPF on Linux
Date: Mon, 12 Apr 1999 13:03:00 +0200
Tim Leung wrote:
> Have anybody got OSPF package on Linux?
>
> Where to get it?
Gated will do OSPF. It's at:
http://metalab.unc.edu/pub/Linux/system/network/daemons/gatedb1_bin_1.3.x.tgz
Regards,
--
Marco van den Bovenkamp.
CIO EMEA Network Design Engineer,
Lucent Technologies Nederland.
Room: HVS BZK 32
Tel.: (+31-35-687)2724
Mail: [EMAIL PROTECTED]
------------------------------
Date: Mon, 12 Apr 1999 14:18:14 +0300
From: Mart Pirita <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Samba Baltic letters problem
Hello.
I live in Baltic States, in Estonia and we use at work both, Macintosh
and PC computers. Lately we installed new (previous OS whose NT)
software (RH 5.2) to our local fileserver, started Samba 2.0.3 and
installed Netatalk 1.4, and before install, we did backup off all data
in Macintosh format.
Later we copyed all data from Mac back to the server, but now we have
problem with filenames, all old files are for Mac users OK, but PC users
have big problems :
Mac writes PC reads
� :8a
� :9a
� :9b
� :9f
� :80
� :85
� :cd
� :86
And if PC users write filenames, Mac users have big problems:
PC writes Mac reads
� �
� �
� �
� �
� �
� �
� �
� �
Please help me, can we convert filenames somehow, should we use other
character set (Baltic - isolatin7), how Samba and Netatalk can work
together.
P.S. Sorry about my bad english.
--
Mart
------------------------------
From: Kwagga <[EMAIL PROTECTED]>
Subject: Re: setting up apache
Date: Mon, 12 Apr 1999 11:12:31 GMT
I had the same problem. I thought that it was a problem with my apache
configuration, but it turns out that not being able to ping the localhost was
the problem.
I found that lo was inactive (using netcfg) to solve this problem I typed
ifconfig lo 127.0.0.1
route add -host 127.0.0.1 lo
Localhost can now be pinged and apache works.
I hope that this solves your problem.
Note: the best solution I have come with so far is to add those two lines to
/etc/rc.d/rc.local, otherwise they have to be typed in each time after a
reboot or shutdown.
-Kwagga
=============================================================================
-- Linux - the choice of a GNU generation
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: Andras Gefferth <[EMAIL PROTECTED]>
Subject: Re: Need help with PPP...
Date: Mon, 12 Apr 1999 11:14:26 +0000
Jerry Wen wrote:
> I brought SuSE 6.0 about 1 1/2 weeks ago. It came with kernel version
> 2.0.36. I setup the PPP configurations, and was able to connect to my local
> ISP. From there on I downloaded the latest kernel, and went through the
> usual upgrading process. The new kernel booted up fine, and I spent some
> time reconfiguring the kernel to work with my SCSI and NT partitions.
> Afterwards I tried to log on the net with Linux 2.2.5, and it wouldn't work.
> I can dial out and connect with my ISP, but the PPP wouldn't route correctly
> when I try to connect outside servers. I got PPP working before, so I went
> thru the how-to again. The problem area is described in section 14 and 18.4
> of the PPP how-to. I can ping my own computer, but not the ISP DNS. The
> default route is not set, when I check the "route -n" command.
>
> Like I said earlier, I got PPP working before. Can someone help me out on
> fixing this thing? I don't know what went wrong... =(
Hi!
No, I can not help you unfortunatelly, but I had a similar problem.
>From home I use two different ISP-s, both provide ppp connection.
PPP connection works fine with one of them.
But not with the other.
I have a very similar script, to set up the two connections, and just as in
your case
the ppp connection sets up correctly, I can check it with ifconfig, and I can
also telnet to the remote ppp port.
But when it comes to modifying the routing table it breaks.
If I type 'route' it shows the header of the routing table and then is stops
doing
anything, I have to kill it with Ctrl-C or something.
It is the same whether I do it by hand, or let pppd do it for me.
I have kernel 2.2.1 running and the 'ISP' runs linux kernel 2.2.0.
IP forwarding is turned on at the ISP.
(Well, in fact I am the root at the ISP linux box.)
I can list /proc/net/route by hand an it sais:
Iface Destination Gateway Flags RefCnt Use Metric Mask
MTU Window IRTT
ppp0 ADF64298 00000000 0005 0 0 0
FFFFFFFF0 0 0
lo 0000007F 00000000 0001 0 0 0
000000FF0 0 0
ppp0 00000000 00000000 0001 0 0 0
000000000 0 0
ppp0 00000000 ADF64298 0003 0 0 0
000000000 0 0
Can anyone help?
Andras Gefferth
------------------------------
From: "Christian Kaiser" <[EMAIL PROTECTED]>
Subject: ppp/pap for planet-interkom account
Date: Mon, 12 Apr 1999 13:20:06 +0200
i have great problems to configure my ppp/pap options
to a planet-intrekom account
i use redhat 5.2 with kernel 2.0.36, the modem dials up, and after a short
time it terminate the connection. netscape doesn't find the connection.
thank you for help
mfg
christian kaiser
------------------------------
From: Matthew King <[EMAIL PROTECTED]>
Subject: pppd quits with no output to console or modem
Date: Mon, 12 Apr 1999 14:12:04 +0100
Please cc to [EMAIL PROTECTED] as my news reader does not alert me.
Using the default scripts with RedHat 4.2, pppd will start a connection
(I check with ps) but then quits after a few seconds (no. indeterminate)
Nothing is output to any consoles or serial ports
I am newish to Linux, but not to DOS. Unfortunately Win95 (which I was
forced to use) has surrounded my brain with pink (black?) fluffy clouds
and numbed it so I cannot work it out myself
Many Thanks,
Matthew King.
------------------------------
Crossposted-To:
vmsnet.networks.misc,microsoft.public.windowsnt.domain,comp.unix.solaris,comp.os.os2.networking.server,comp.os.ms-windows.nt.admin.networking,comp.infosystems.www.servers.unix,comp.protocols.tcp-ip.domains
From: [EMAIL PROTECTED] (J.D. Baldwin)
Subject: Re: Machine name themes - what do you use?
Date: Mon, 12 Apr 1999 14:51:21 GMT
In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED]
<[EMAIL PROTECTED]> wrote:
> i name mine after american dictators
>
> eg reagan, johnson, kissenger, nixon etc
What, no 'haig'?
--
From the catapult of J.D. Baldwin |+| "If anyone disagrees with anything I
_,_ Finger [EMAIL PROTECTED] |+| say, I am quite prepared not only to
_|70|___:::)=}- for PGP public |+| retract it, but also to deny under
\ / key information. |+| oath that I ever said it." --T. Lehrer
***~~~~-----------------------------------------------------------------------
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: yppasswd not working?
Date: Mon, 12 Apr 1999 14:52:21 GMT
In article <7es71e$d14$[EMAIL PROTECTED]>,
Thorsten Kukuk <[EMAIL PROTECTED]> wrote:
> Look in the file where rpc.yppasswdd is startet, read the manual page,
> and set the correct arguments. So easy.
Well, I had read the manual page, but anyway -- I got it to work by mv the
passwd file to /etc/yp/ and then making a symlink from /etc/passwd to
/etc/yp/passwd. Seems to work, but is that proper way to do it?
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: "Mike Ellis" <[EMAIL PROTECTED]>
Subject: Help. Unable to browse HTTPS from IP Masq'ed workstations
Date: Mon, 12 Apr 1999 09:43:32 -0700
SYSTEM:
Small network of Linux (RH 5.2) and Windows boxes.
Internet connection through ppp0 on one of the
Linux boxes. This box's hostname is austin.
Domain is frogwing.com
ppp0 has static IP from ISP
eth0 = 192.168.1.1
Other boxes obtain 192.168.1.X IP's from austin's
dhcpd. Austin is running ip_masq as described in
the how-to. Austin is also the domain nameserver.
PROBLEM:
Other boxes are unable to access certain websites,
especially HTTPS, but also a few HTTP sites
(www.compaq.com is one).
Problem is common to both Netscape on Linux and
IE4 on Win95. Problem does not occur with Netscape
run on directly austin.
MORE INFORMATION:
Here are the ipfwadm rules from rc.local
ipfwadm -F f
ipfwadm -F -p deny
ipfwadm -F -a m -S 192.168.1.0/24 -D 0.0.0.0/0
As a diagnostic effort I tried changing the default
forwarding policy from deny to accept, but it
had no impact.
All other clients are working correctly from
all boxes, telnet, ftp, mail, dns, icmp ...
Before installing RH 5.2, Austin was running
Slackware (2.0.22) with a similar setup, except
no DHCP. All clients browsed with zero problems.
I'm stumped. Any ideas?
Thanks
Mike Ellis
------------------------------
From: Joachim Gloger <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,de.comp.os.unix.linux.misc
Subject: automount/mount does not react
Date: Mon, 12 Apr 1999 17:53:25 +0200
Hi,
we are having a problem relating to the automount/mount utility. When
using the automount program we try to mount some other computers of our
local net to the /Net directory. Some of the other computers are
sometimes not available. If this happens the whole automount process
hangs.
Below you find the entries presented by the "ps aux" command. Both
commands do not return, since the computer named "strawson" do not
respond.
root 332 0.0 0.3 1080 648 ? S 17:02 0:00 automount -t
300 /private/automount/Net file /etc/auto.net
root 333 0.0 0.2 944 552 ? S 17:02 0:00 /bin/mount -t
nfs -s -o rw,bg,soft,timeo=5 strawson:/ /private/automount/Net/strawson
What options must be set that those never responding machines are
skipped? We tried to diminish the timeout values of the automount as
well as of the mount command, but with no success. Also we play around
with the possible nfs options of the mount command, but again, no
positive reaction.
Next problem: What can we do when the computer gives a response to the
ping command, but do no export a valid nfs file system. A Win95/98
client for example.
Thanks in advance for any help. Please send also a Cc: of your answer to
my email address
===========================================================
J. M. Gloger, DaimlerChrysler AG, Research Center Ulm
P.O. Box 2360, 89013 Ulm, Germany
Phone: +49 731 505 2353
Fax: +49 731 505 4113
Email: [EMAIL PROTECTED]
walk the talk
===========================================================
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.networking) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Networking Digest
******************************
