Linux-Networking Digest #446, Volume #12 Thu, 2 Sep 99 03:13:56 EDT
Contents:
Re: RH 6.0, PPP, and my ISP (Josh Thompson)
Communication needed between linux server/client to windows client/server ("Long Ta")
I've gotten on, but how do I get off???? ("Scott")
printer scripts with lpd ("Kevin Williams")
Re: Setting up Masquerading under RH6.0 (Stephen Torri)
RedHat 6.0 and tty problem (Victor Kwok)
Re: PCMCIA/DHCP help ! ([EMAIL PROTECTED])
CHAP-Trouble (Bodo Meseke)
Networking. (Nails)
Problems with NFS client (Bernt Christopher Rosland)
Re: NFS and permission denied ("William B. Cattell")
Re: Routing in Ring (Cluster) ([EMAIL PROTECTED])
Need help with telnet (Mike)
Re: Installing Via Crossover Cable Connection (Semi-Newbie) ([EMAIL PROTECTED])
Re: Linux friendly ISPs ([EMAIL PROTECTED])
2 PCMCIA Ethernet Cards (Ben Stragnell)
Re: NFS and permission denied ("William B. Cattell")
Re: PCMCIA/DHCP help ! (David Hinds)
eth0: Too much work at interrupt (Tero Niemi)
Re: NFS and GNU Linker producing corrupted executables (RHL (Jennie Haywood)
IPLOG - problems compiling/installing (Vlar Schreidlocke)
Re: Setting up Masquerading under RH6.0 (Stephen Torri)
----------------------------------------------------------------------------
From: Josh Thompson <[EMAIL PROTECTED]>
Subject: Re: RH 6.0, PPP, and my ISP
Date: Thu, 02 Sep 1999 02:56:38 +0000
I've tried RedHat Network Configurator (netcfg), and linuxconf. Neither
seems to work. Tried changing
NAME=<my PAP name>
to
export NAME=<my PAP name>
that didn't work either.
I agree with David that it seems my end is rejecting the use of PAP. I
have a pap-secrets file in /etc/ppp that has one line in it of the
following format:
<my PAP name> * <my passwd>
the fields are separated by one tab character. I have tried setting the
name option in the options, the user option, and tried using these on
the command line when I run pppd.
Is there an option I can use to force pppd to use PAP?
Thanks a bunch.
Josh
------------------------------
From: "Long Ta" <[EMAIL PROTECTED]>
Subject: Communication needed between linux server/client to windows client/server
Date: Thu, 02 Sep 1999 00:59:01 GMT
Dear Linux Pros,
I want my linux system see the window 98 file system, but don't know how.
Is there any windows Linux service program or equivalent program out there
that windows can serve files to a Linux machine.
My window cannot log into my Linux neither. I was able to run telnet and
connect to the linux system using the IP address. The prompt comes up
nicely, but the login is refused saying password is incorrect as I try to
log in as root.
Thanks much.
Long
------------------------------
From: "Scott" <[EMAIL PROTECTED]>
Subject: I've gotten on, but how do I get off????
Date: Wed, 1 Sep 1999 23:42:02 -0400
I'm using minicom and pppd to access my provider, but after I'm done I can't
disconnect. I've tried hanging up and quitting and resetting on minicom.
Only way I can disconnect is by rebooting!!!!
There must be a better way.
Scott
------------------------------
From: "Kevin Williams" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc
Subject: printer scripts with lpd
Date: Thu, 2 Sep 1999 16:29:28 +1200
Hi All
Do any of you Linux gurus out there know how to pass parameters to a print
job and then process them in a script.
The if in the printcap file is too limited with the parameters it sends.
I am using RH 5.2 linux to print remote print jobs from a Unix host.
I am after something like the interface scripts (/etc/lp/interfaces/printer
name) that are used with lp in Unix.
Many thanks
Kevin.
------------------------------
From: Stephen Torri <[EMAIL PROTECTED]>
Crossposted-To: comp.security.firewalls
Subject: Re: Setting up Masquerading under RH6.0
Date: Wed, 01 Sep 1999 23:54:55 -0400
Thanks. I took care of the default policy and the wrong variable. Is
this enough to do masquerading? Do I need to set a route from my
internal network to the ppp0? Will this route packets?
Stephen
Gilles 2 wrote:
>
> First, just delete first default policy. One is sufficient.
>
> After,
> #------------------------------------
> # Local Traffic Rules
> #------------------------------------
> /sbin/ipchains -A input -j ACCEPT -i $INTERFACE_DEV -s $NETWORK/24
> -d $INTERFACE_IP
> # Accept any packet from local network to server on input of eth0
> /sbin/ipchains -A input -j ACCEPT -i $LOOPBACK_IP -s $NETWORK/24 -d
> $INTERFACE_IP
> # Accept any packet from local network to server on input of lo
> /sbin/ipchains -A output -j ACCEPT -i $INTERFACE_DEV -s $INTERFACE_IP
> -d $NETWORK/24
> # Accept any packet from server to local network on output of eth0
> /sbin/ipchains -A output -j ACCEPT -i $LOOPBACK_IP -s $INTERFACE_IP -d
> $NETWORK/24
> # Accept any packet from server to local network on output of lo
>
> and you mistake UNIVERSE and INTERNET ...
>
> and the mask is not 255.255.255.0 but 24 ...
>
> --
> Gilles C.
> Sysop at France Multimedia
------------------------------
From: Victor Kwok <[EMAIL PROTECTED]>
Subject: RedHat 6.0 and tty problem
Date: Thu, 02 Sep 1999 12:25:33 +0800
Hi,
I used to install RH5.2 for software development, which will
detect the tty of the session. In RH 5.2, it uses ttyp* for telnet
section.
I recently pick up RH6.0 and install it. However, I find that the
tty has been changed to pts/*. Is there anywhere I can find
information about it? Also, is there anyway I can change the tty type
back to the old standard?
Thank you very much!
Victor
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.portable
Subject: Re: PCMCIA/DHCP help !
Date: Thu, 02 Sep 1999 05:19:08 GMT
In article <7pm8f4$s8s$[EMAIL PROTECTED]>,
"Ruben" <[EMAIL PROTECTED]> wrote:
> Hi there fellas,
>
> I got my PCMCIA card running, it even do the DHCP
routine.
> It gets its IP
> address and everything. Now the problem is that none of
the
> communication
> programs (ping,ftp,telnet,lynx,netscape and so forth)
seems to
> realize that it
> already has a internet connection, therefore giving me
errors
> such as:
> connection refused , unknown remote host , and several
other.
> I(since using
> BugHat 6 )have got the linuxconf and network tools, but
these
> are really
> prodigal. They're just a waste of time. I even tried to
> configure'em manually
> /etc/hosts
> /etc/host.conf
>
> This effort's been futile. Since for some reason RH just
> overwrite'em each
> time I reboot. I went to LDP web site and found some
> information about DHCP
> and PCMCIA but I'm not able to connect still.
>
> Help is appreciated,
>
> Ruben
> "Is it hot in here or it's just me?" -An Overclocked CPU
>
>
Hi,
Do you set the file /etc/pcmcia/network.opts?
It need you set the parameters such as gateway,dns sever...
BTW:
How do you get your PCMCIA run using dhcp?
I can use it only when I specify an IP address.
There is an parameter in network.opts that control the dhcp.
I set it, but no result. No dhcpcd is runnig after reboot. When run
dhcpcd manually in the shell,I just get the error message like "ioctl
SIOCSIFBRDADDR (ifConfig): Cannot assign requested address".
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: Bodo Meseke <[EMAIL PROTECTED]>
Date: Thu, 02 Sep 1999 05:27:35 GMT
Subject: CHAP-Trouble
Hello,
I've some trouble while setting up a dial-in-server. I configured a=20
linux-box (SuSE 6.2) as internet-router for a little network at home.=20=
The same pc should act as a dial-in-server and use chap=20
authentication. The machine's name is =84router=93. I'm using an ISDN-ca=
rd=20
and bound ippp0 to Channel 0 (Dial-Out) and ippp1 to Channel 1=20
(Dial-In).
The problem is, that you now can dial into my linux-box, using the=20
username/Password combination my ISP is using. (It's german internet=20=
by call, so anyone knows this combination).
My /etc/ppp/options.ippp1 contains
name =84router=93
+chap
My /etc/ppp/chap-secrets
arcor * internet *
dialin router secret *
Can anyone gimme a hint, where I made a mistake ?
Thanks in advance
--=20
Viele Gr=FC=DFe,
Bodo Meseke
Member of Team StarOffice - http://www.teamstaroffice.org
Visit my homepage at http://www.star-page.de
------------------------------
From: [EMAIL PROTECTED] (Nails)
Subject: Networking.
Date: Thu, 02 Sep 1999 05:19:47 GMT
I know nothing about networking but would like to know a little more!
Can it be done without an ethernet adapter? (i have a dial up
connection)
Can anyone point me in the direction of some basic learning
resources??
All help appreciated!!
Paul
I really want to know what is the purpose of networking and what can
it be used for!!
------------------------------
From: Bernt Christopher Rosland <[EMAIL PROTECTED]>
Subject: Problems with NFS client
Date: Thu, 02 Sep 1999 07:41:02 +0200
I'm running RH 6.0 and am trying to set up a disk-less client using NFS,
but I'm having problems booting the client. When it gets to starting
local in rc it stopps after this comes up:
INIT: Id "1" respawning too fast: disabled for 5 minutes
" 2 "
" 3 "
" 4 "
" 5 "
" 6 "
INIT: no more processes left in this runlevel
Can anyone tell me what this is?
-Bernt
------------------------------
From: "William B. Cattell" <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: NFS and permission denied
Date: Thu, 02 Sep 1999 05:26:22 GMT
Allen Wong wrote:
>
> Bill,
>
> What do you get when you issue this command?
>
> /usr/bin/showmount -e <NFS servername>
>
> Could you also post your exports file?
>
> Allen
> --
> Linux: If you're not careful, you might actually learn something.
Allen - I have two computers on different subnets -
c659784-b is 24.7.149.136 and is the NFS server. c659784-a
is 24.5.52.97 and is the NFS client. I can ping and telnet
between them so connectivity is not an issue.
Which tells me that the server is not advertising the
exports. I'm also puzzeled by the results of
/usr/sbin/rpcinfo -p which shows udp and tcp connections for
all services except nfs. According to the FAQ there
*should* be a udp and tcp connection for nfs...
[root@c659784-b /etc]# /usr/sbin/rpcinfo -p
program vers proto port
100000 2 tcp 111 rpcbind
100000 2 udp 111 rpcbind
100024 1 udp 606 status
100024 1 tcp 608 status
100011 1 udp 617 rquotad
100011 2 udp 617 rquotad
100005 1 udp 627 mountd
100005 1 tcp 629 mountd
100005 2 udp 632 mountd
100005 2 tcp 634 mountd
100005 3 udp 637 mountd
100005 3 tcp 639 mountd
100003 2 udp 2049 nfs
100021 1 udp 1027 nlockmgr
100021 3 udp 1027 nlockmgr
100021 1 tcp 1050 nlockmgr
100021 3 tcp 1050 nlockmgr
[root@c659784-b /etc]# /sbin/lsmod
Module Size Used by
nfsd 150936 8 (autoclean)
lockd 30856 1 (autoclean) [nfsd]
sunrpc 52356 1 (autoclean) [nfsd lockd]
tulip 25620 1 (autoclean)
[root@c659784-b /etc]# /usr/sbin/exportfs
/home/httpd c659784-a.flrmnd1.tx.home.com
/ c659784-a.flrmnd1.tx.home.com
TIA,
Bill
--
==============================================================
http://members.home.com/wcattell
==============================================================
Park not thy Harley in the darkness of thine garage, that it
may collect dust for want of being oft ridden. Ride thy
Harley
with thy brethren, and rejoice in the spirit of the road.
==============================================================
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Routing in Ring (Cluster)
Date: Thu, 02 Sep 1999 05:25:31 GMT
[EMAIL PROTECTED] wrote:
>: I have a cluster with 32 computers in a ring.
>: The problem I have is, that, I can not access
>: a computer, that is routed more than over
>: 15 computers.
>: So, if I try to access 1->17, than the packets
>: are lost (killed in computer 18).
>:
>: can anyone help me out here?
Sure. Remove one NIC from each machine. Take them (and the same
number of cables, minus five) to a local used-computer-parts store and
see if they will buy the things from you. Use the money to buy six
8-port hubs at the same store. (The odds are they don't have anything
larger than 8-port - if they do, compare total price and get what's
cheaper.) From *one* hub, connect ports to the uplink ports (switched
to uplink mode) on the other hubs. Do not connect any computers to
that hub, but connect them as you like to the other hubs.
Now no machine is more than one logical hop, or four physical hops,
from any other machine; you have three free ports for additional
computers and three more for additional concentrators. In addition,
your network performance is improved and you have an easier time
sorting out addresses and configuring the machines.
------------------------------
From: Mike <[EMAIL PROTECTED]>
Subject: Need help with telnet
Date: Thu, 02 Sep 1999 05:31:21 GMT
When trying to telnet into my linux machine from a aix box, I get a
message that says 'No terrminfo entry for "xterm-color"'. What
settings do I change where, my linux machine or my aix machine?
Thanks
Mike
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.setup
Subject: Re: Installing Via Crossover Cable Connection (Semi-Newbie)
Date: Thu, 02 Sep 1999 05:26:08 GMT
On Wed, 1 Sep 1999 08:55:20 -0500, "Kevin G." <[EMAIL PROTECTED]>
wrote:
>I'm using RH 5.2 Kernel 2.0.36
>
>I have a Nic on both machines but only one cd-rom. I dual-boot Win98 and
>Linux on main system and want to install linux on my second system via
>crossover cable (486/66, 16MB, 540MB-HD, NE2000 Compatible NIC). I don't
>have any type of networking set up yet. What type installation or how do I
>install linux on my second machine.
I have done this, with approximately the same version and hardware and
a tiny bit more memory. However, I already had Windows file and
printer sharing set up and working. That is the first step. Anyone
who doesn't need to know how to do that, skip this message.
Alternatively, if you have NFS set up under Linux on your bigger box
(don't ask me how to do that) you can just do an NFS install. Since I
don't know how to do that, having noted the option I will now ignore
it.
You have the network card installed in Windows95? Then you have the
Network Neighborhood icon on your desktop. If not, go through the
"Add New Hardware" process to install it. (If it isn't in the machine
- you WILL need two network cards, one in each machine.)
Right-click on the Network Neighborhood icon and select Properties.
It comes up with three tabs.
The first tab is "Configuration". It needs to have at least the
following entries; if any are missing, click Add and add them. In no
particular order (actually, in the order they occur on my machine)
(1) Client for Microsoft Networks (in the Add list, it's a Client by
Microsoft)
(2) Network card adapter
(3) TCP/IP
(4) File and Printer Sharing for Microsoft networks
If you have two network adapters, including any virtual network
adapters such as those used by Dial-Up Networking and AOL, then you
need to make sure that each of these are connected (through
intermediaries in some cases) to the correct adapter: click the
adapter, click Properties, go to Bindings, and make sure TCP/IP is
selected; say OK; click TCP/IP -> your adapter, click Properties, go
to Bindings, and make sure the other two items are selected; say OK.
Set the Primary Network Login to "Windows Login".
Click the "Identification" tab. If either "Computer Name" or
"Workgroup" is blank, make up a value (don't make them the same).
Note them both, you'll need them later.
While we're wandering, click the "Access Control" tab and make sure
that "Share-level access control" is selected.
Now go back to the Configuration tab. Click the File and Printer
Sharing button and check both check boxes; click OK to get rid of
that. Select "Client for Microsoft Networks" (directed to TCP/IP on
your adapter, if you have a choice) and click Properties.
Click "Log on to Windows domain" so that it is checked, then key in
the domain name you noted a moment ago in the box below. Now UNcheck
that box. (Yes, this is apparently necessary.) We don't care about
the bottom half; click OK.
Select "TCP/IP" (directed to your adapter, if you have a choice) and
click Properties. Click the "IP Address" tab. Click "Specify an IP
address". Give it the address 192.168.0.1 and the subnet mask
255.255.0.0. Leave everything else alone for now (this is a quick and
dirty setup - if you want to do anything extensive, like set up a
router, you need to understand networking enough that you don't need
instructions this detailed, so I'm leaving a lot out). Click OK.
You should be back to the "Network" window. Click OK. It'll go
through a lot of motions and then ask if you want to reboot. The
answer is most emphatically yes.
When it comes back up, stick the Linux CD in the drive and
double-click the "My Computer" icon. Right-click the CD drive and
select Sharing. Click Shared As:, name it CD, select Read Only, skip
the password part, click OK. I don't think it'll ask you to reboot
now, but if it does, reboot. While you're here, go to the Readme file
on the CD and find out how to make a boot floppy AND a supplemental
floppy - you need both.
Now, go to the to-be-Linux box and boot it off the boot floppy. At
the appropriate point, tell it you want to install via Samba. It will
ask for various things, respond appropriately. When it asks for the
location to install from, it's /your-machine-name/CD
It should go from there.
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Linux friendly ISPs
Date: Thu, 02 Sep 1999 05:36:03 GMT
On Wed, 1 Sep 1999 21:44:29 -0500, "Todd K. Tuttle"
<[EMAIL PROTECTED]> wrote:
>I Dial in,
>Get "Connect 2X000/ARQ/V34/LAPM/V42BIS",
>I hit <Enter>,
>I get "Username:",
>I enter my username and hit <enter>,
>I get "Password:",
>I enter my password and hit <enter>,
>I get a ">" prompt,
>I enter the command "ppp",
>I get PPP characters,
>
>Now this doesn't seem complicated. But whatever I try to put in, it hangs at
>the "Connect" part and just sits. And I just sits...I've tried going through
>every chat options, from the chat man page. All the suggestions in the How
>To PPP guide. The how to hook up PPP by W.G Unruh. Absolutely nothing works.
>
>So I'm asking, begging actually, if you where using a Linux PPP connection
>using pppd to connect to the above connection, what would your script look
>like?
Well, here's the script I copied off my ISP's web page, suitable
modified:
ABORT "NO DIALTONE"
ABORT "ERROR"
ABORT "NO ANSWER"
ABORT "BUSY"
ABORT "Username/Password Incorrect"
"" "at"
OK "atm0"
OK "atdt{phone number}"
ogin: "{userid}-ppp"
word: "{password}"
For yours, I suggest this: first copy the above down to and including
the line that starts OK "atdt
change the phone number and continue with
CONNECT ""
name: "{userid}"
word: "{password}"
> "ppp"
Warning, this is at least half speculation. If it doesn't work, the
*first* thing I would do is read the man page for chat and see if
there's a way to insert about a 2-second delay between receiving the
CONNECT and sending the empty command that follows. After that, I'd be
completely guessing.
------------------------------
From: Ben Stragnell <[EMAIL PROTECTED]>
Crossposted-To: redhat.networking.general
Subject: 2 PCMCIA Ethernet Cards
Date: Thu, 02 Sep 1999 04:31:05 GMT
Hi,
I realise this is a fairly unusual situation, but I'd like to configure an
old laptop as a router. It has 2 PCMCIA slots, and I have 2 PCMCIA ethernet
cards that appear to be NE compatible.
Linux seems to recognise both cards on powerup, and brings up eth0 and
eth1. However, the moment I try pinging anything on eth1, I start getting
messages like "pcnet_8390_reset did not complete". A glance at ifconfig
suggests that the eth1 card is not receiving any interrupts. I'm using
Redhat 5.2.
Any help would be gratefully received, preferably via email.
Cheers,
Ben
================== Posted via CNET Linux Help ==================
http://www.searchlinux.com
------------------------------
From: "William B. Cattell" <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: NFS and permission denied
Date: Thu, 02 Sep 1999 05:38:16 GMT
jsrockford wrote:
>
> I'm real new to this NFS thing so take my advice w/ a grain of
> salt...been doing something similar to what you are trying (between RH6
> & Mandrake 6 boxen)...your /etc/hosts files should be similar between
> the two machines (mine looks something like:
>
> 192.168.0.1 main.mydomain.net main
> 192.168.0.3 linux.mydomain.net linux
>
> Then your /etc/exports file on the server (in this case 'linux') should
> have something like:
>
> /path/to/directory/to/export main(rw)
>
> I use 'main' as the name of my machine I'm allowing export of my
> filesystem to.
>
> Now I had similar problems to you with the 'permission denied' error
> until I changed my /etc/hosts.allowed file to include:
>
> PORTMAPPER: 192.168.0.0/255.255.255.0 # NOTE: insert your network IP
> address for my 192.168.0.0
>
> I had the 'ALL: LOCAL' in there but didn't seem to work...don't know
> why. You may need to restart inet: /etc/rc.d/init.d/inet restart (I
> think; doing this from memory, but should be something like that....else
> reboot).
>
> I've restarted NFS at this point (don't know if you have to when running
> 'exportfs' though)...using command: /etc/rc.d/init.d/nfs stop
> then: /etc/rc.d/init.d/nfs start
>
> Then I usually run 'exportfs' but this may be redundant..works for me.
>
> Okay now on the client machine try:
>
> mount -o rsize=1024,wsize=1024 linux:/path/to/directory/to/export
> /mnt/point
>
> This has always worked for me...you will want to make sure permissions
> are set on the server side for the directory so you can write to it from
> the client machine. If this is clear as mud...check out the HOWTO for
> NFS at http://metalab.unc.edu/mdw/HOWTO/NFS-HOWTO.html
>
> Good luck!
>
> In article <[EMAIL PROTECTED]>,
> [EMAIL PROTECTED] wrote:
> > I'm starting to work on creating an NFS link from one Red
> > Hat Linux 6.0 system to another. I had to back rev to
> > kernel 2.2.5 (from 2.2.11) to resolve an error when trying
> > to load rpc.nfsd. Now that NFS is loading and gets started
> > ok (rpcinfo -p shows that it's listening) I'm trying to
> > mount the exported directories.
> >
> > I've gone so far as to change my hosts.allow and hosts.deny
> > to let *everything* in (not permanently, just when I'm
> > working on it) but still get a 'permission denied' when
> > trying the mount. I try to mount as a normal user I get
> > 'root has to do that' even though I set it to be user
> > mountable. If I try to mount it as root then I get the
> > 'permission denied'. I figure I'm missing something easy
> > but between the HOWTO and the O'Reilly Assoc. "Managing NFS
> > and NIS" I've managed to numb my brain. Any hints, tips
> > suggestions (or flames) would be appreciated. TIA.
It's working... I had to stop the NFS services AND unload
the NFS modules. I did a /etc/rc.d/init.d/nfs start and now
I can mount. Thanks Allen and jsrockford for their insight
and assistance...
Bill
--
==============================================================
http://members.home.com/wcattell
==============================================================
Park not thy Harley in the darkness of thine garage, that it
may collect dust for want of being oft ridden. Ride thy
Harley
with thy brethren, and rejoice in the spirit of the road.
==============================================================
------------------------------
From: [EMAIL PROTECTED] (David Hinds)
Crossposted-To: comp.os.linux.portable
Subject: Re: PCMCIA/DHCP help !
Date: 2 Sep 1999 05:38:32 GMT
[EMAIL PROTECTED] wrote:
: There is an parameter in network.opts that control the dhcp.
: I set it, but no result. No dhcpcd is runnig after reboot. When run
: dhcpcd manually in the shell,I just get the error message like "ioctl
: SIOCSIFBRDADDR (ifConfig): Cannot assign requested address".
I suspect that this means your dhcpcd is out of date.
-- Dave Hinds
------------------------------
From: Tero Niemi <[EMAIL PROTECTED]>
Subject: eth0: Too much work at interrupt
Date: Thu, 02 Sep 1999 09:02:15 +0300
I got the following error message:
" eth0: Too much work at interrupt, status 0x01"
status changes once in a while to 0x40
I got apache 1.2.6 running and eth0 has one IP-alias. Kernel is 2.0.36
Could someone please tell me what this error means and from which server
prints it comes from?
------------------------------
From: Jennie Haywood <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,comp.unix.aix
Subject: Re: NFS and GNU Linker producing corrupted executables (RHL
Date: Thu, 02 Sep 1999 00:57:03 -0500
Reply-To: [EMAIL PROTECTED]
Niklas Edmundsson wrote:
> > Do you have more references? It is always easier to get a fix from
> > IBM if you can tell them that it is a known problem. :-)
>
> From an old posting of mine:
>
> The SunSolve bug id is 4071076 data over length in nfs header was
> written to disk.
>
I'm working on getting the code from Sun. Sun has already released ONCPLUS 2.2
and we are somewhere BETWEEN 2.1 and 2.2..... It's a mess to maintain.
Sometimes we get to it (fixing a bug) before they do ......Sometimes we have to
fix it differently.
>
> Linus Torvald's comment on the Solaris-bug was:
> "Actually, it appears fine on the wire, this particular problem seems to
> be due to Solaris getting a "merge adjacent packets" case wrong when the
> merge happens to cross a 8kB boundary and the data payload of the first
> packet is not divisible by four.."
>
> > : Strangely enough, Linux seems to be the only OS that triggers it.
>
> o> 4.2.1 NFS server (we didn't have the problem with 3.2.5 servers). I have
> > reported this problem to IBM at the end of 1997, beginning of 1998 (don't
> > remember exactly), but the fix they sent me was for bos.net.nfs.client,
> > not for bos.net.nfs.server. We still have the problem, and I 'solved'
The NFS server is delivered in bos.net.nfs.client. In fact, the client and
server
are both in bos.net.nfs.client AND they are ONE extension
(/etc/drivers/nfs.ext), atleast after 4.2.1.
The only REAL way to determine what release of the server you have, is to do a
what on the nfs.ext and look at the levels of nfs3_srv and nfs_srv. That won't
tell the folks in Level 2 anything, but it tells me ALOT..
I am working on this ......when I have an APAR available, I'll post it.
>
> > it by writing a wrapper for the compilers on the HP box, that creates .o
> > files in the local /tmp and moving them to the NFS server afterwards.
> > Same for the linker. I guess IBM didn't do anything about it because
> > the HP-UX client is still running HP-UX 9.0x, and IBM claims it is an
> > HP problem.....
>
> Oooh. The old "it's not my problem it's yours" situation ;)
>
> > : My standard-test is compiling glib: ./configure --prefix=/tmp/foo ; make
>
> > : It usually fails with messages like:
> > : BFD: gstring.o: invalid string offset 13824 >= 77 for section `'
> > : hash-test.o: file not recognized: File truncated
>
> > : And so on...
>
> > Hmm, the HP linker doesn't complain, but the executables won't run.
> > I get either a core dump or the program starts writing strange
> > characters to stderr and sits there forever.
>
> Cool... :) Seriously, the linker shouldn't any kind of garbage as an
> object-file. But it's HP-UX so I'm not that surprised :)
>
------------------------------
From: [EMAIL PROTECTED] (Vlar Schreidlocke)
Subject: IPLOG - problems compiling/installing
Date: Thu, 02 Sep 1999 03:44:05 GMT
Someone reccomended IPLOG from http://www.numb.org/~odin/ as a good
port/firewall logging program. I am relative new to Linux (Red Hat
6.0) and I am trying to compile/install IPLOG. I had trouble with the
./configure script in that it apparently was missing the LIBPCAP
library. I found this at www.freshmeat.net and downloaded and intalled
it. I still can't get IPLOG to compile properly. Could someone give me
some advice on how to do this? I have a P166 running Red Hat 6.0.
------------------------------
From: Stephen Torri <[EMAIL PROTECTED]>
Subject: Re: Setting up Masquerading under RH6.0
Date: Thu, 02 Sep 1999 00:23:53 -0400
I made the changes that you recommended. I removed the three lines
accepting packets from the internet for protocols (icmp, tcp, and udp).
I then reran the rc.firewall. I could ping froma client (10.0.0.2) to
the dial out server at (10.0.0.6). I could not ping from the client to
the $PPP_IP assigned to me by the ISP. Is there something I must do for
routing?
Stephen
------------------------------
** 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
******************************
