Network dies under Kernal 2.2.5-15 & 22 -- Kernal BUG??

Network dies under Kernal 2.2.5-15 & 22 -- Kernal BUG??

Post by Jason Boerne » Thu, 24 Jun 1999 04:00:00



Maybe this issue is RedHat specific but I'm seeing similare symptoms of
my problems listed fairly frequrently these days.  The RedHat customers
that I'm talking too are generally just moving back to    5.2 but I
really don't like that answer.  Can someone help me shed some light on
this issue.

My box is a Dell Latitude CPi D300XT.
The PCMCIA net card works fine under RH5.2 and other OS's:
# lsmod
Module                  Size  Used by
nfsd                  150936   8  (autoclean)
lockd                  30856   1  (autoclean) [nfsd]
sunrpc                 52356   1  (autoclean) [nfsd lockd]
serial_cs               3696   0  (unused)
3c574_cs                8648   1
ds                      5740   2  [serial_cs 3c574_cs]
i82365                 21956   2
pcmcia_core            39720   0  [serial_cs 3c574_cs ds i82365]
opl3                   11208   0  (unused)
sound                  57208   0  [opl3]
soundlow                 300   0  [sound]
soundcore               2372   3  [sound]

#  /sbin/ifconfig -a
eth0      Link encap:Ethernet  HWaddr 00:00:86:1F:4D:DC
          inet addr:192.168.100.2  Bcast:192.168.100.255
Mask:255.255.255.0
          UP BROADCAST RUNNING  MTU:1500  Metric:1
          RX packets:511 errors:0 dropped:0 overruns:0 frame:0
          TX packets:269 errors:47 dropped:0 overruns:0 carrier:4
          collisions:6 txqueuelen:100
          Interrupt:3 Base address:0x300

lo        Link encap:Local Loopback
          inet addr:127.0.0.1  Mask:255.0.0.0
          UP LOOPBACK RUNNING  MTU:3924  Metric:1
          RX packets:49 errors:0 dropped:0 overruns:0 frame:0
          TX packets:49 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0

#  netstat -r
Kernel IP routing table
Destination     Gateway         Genmask         Flags   MSS Window  irtt
Iface
192.168.100.0   *               255.255.255.0   U         0 0          0
eth0
127.0.0.0       *               255.0.0.0       U         0 0          0
lo
default         192.168.100.1   0.0.0.0         UG        0 0          0
eth0

Here are my previous two postings seperated with a ******:
******
Subject:
           Re: RH6 upgrade and lost network
       Date:
           Wed, 23 Jun 1999 11:49:05 -0700
       From:

   Reply-To:

 Organization:
           Cox Communications
 Newsgroups:
           redhat.networking.general
  References:
           1

I posted a similar problem a few weeks ago where it looks
like some kind of
kernal bug.  I dropped it while my time was taken elsewhere
but now I'm
forced to confront this issue.

During an FTP operation for example, I keep getting the
message:
eth0: Transmit timed out, Tx_status 00 status 2000 Tx FIFO
rooom 4092

My arp tables look fine and the network works fine until this
mysterious
time-out issue shows up.  I am not running any routing
protocols.

This is running 2.2.5-22.  Things were FAR worse under
2.2.5-15.

If someone can push me along to finding some better debug
tools I would
appreciate it.

Thanks


> After I upgraded to RedHat 6.0, my Linux machine stopped
communicating
> with the rest of my network. I'm no longer able to ping the
gateway or
> any other machines on the same network and I can't ping the
Linux
> machine from any of the other machines. (Ehem yes, the
cabling is okay -
> it works when I plug it into a different machine <g>).
However, the
> Linux machine can ping itself from the console, using both
the loopback
> and the ethernet interface (127.0.0.1 and 192.168.154.12).

> netstat -r shows a normal routing table, but netstat -i
looks kind of
> unusual:

> IFace  mtu Met  rx-ok rx-err rx-drp rx-ovr  tx-ok tx-err
tx-drp tx-ovr
> Flg
> eth0  1500   0      0      0      0      0   1107      0
0   1101
> BRU
> lo    3924   0    480      0      0      0    480      0
0      0
> LRU

> arp -a shows an empty arp table.

> Has anybody seen this before?

> Klaus Sonnenleiter
> The Media Machine, LLC

*******
Subject:
           Kernal Hacks Only: Network stops responding when stressed
           under RedHat6
       Date:
           Fri, 04 Jun 1999 21:27:34 -0700
       From:

   Reply-To:

 Organization:
           Cox Communications
        CC:

 Newsgroups:
           redhat.networking.general

Network stops responding when stressed under RedHat 6

I am using redhat 6 kernel 2.2.5-15 and -22

If I start to do something network intensive (A CVS "cvs  -z3
<EVERYTHING>" operation toward mozilla to be specific), the
network
stops responding.

If I do an /etc/rc.../network stop; sleep 5;
/etc/rc.../network start
I can slip some more traffic through for a brief period but
then I need
to restart the network again.

At kernel level -22 I get slightly different symptoms.
I no longer need to restart the network.  However my response
times
become VERY large (35000 ms range; up from 40 ms range).  I
will get
these response times for a few minutes and then some resource
would
appear to be briefly available again (my response times drop
back to the
40 ms range).

This cycle persists until my large operation operation ends.

I can recreate this every time.

Any idea?

I have a gut feeling that this is not a network issue but I'm
out of
ideas to try.
Can someone give me hints as to getting  snapshots of what
resources are
in use at any given time so that I can start to look for
starvation
issues?

Thanks

 
 
 

Network dies under Kernal 2.2.5-15 & 22 -- Kernal BUG??

Post by Marc Mut » Thu, 24 Jun 1999 04:00:00


kernel 2.2.5 is said to contain a networking bug, altough I do not know
more. You should probably upgrade to 2.2.10, as this version also fixes
a vulnerability for DoS-Attacks (see www.kernelnotes.org).

Marc

PS: Don't post so big an article. No-one will read them completely...

 
 
 

1. Network Slows and all but dies on RedHat 6.0 w/2.2-15 Kernel

I have two different kinds of systems. Both have different hardware and
ethernet cards. Both will, after an unusually large amount of network
traffic, have the NIC cards slow down to a crawl and all but die. (It
takes about 10 minutes to ssh in sometimes.) And I can fix it by logging
on and issuing

nohup /etc/rc.d/init.d/network restart

That fixes it but I can't be online 24-7 watching these boxes and
running that command. This has never happened with any of the many other
versions of RedHat I've run with similar hardware configurations. So I
suspect that it's not be causing this.

Is there a known issue with RedHat 6.0 and network speeds/death? I've
heard a few people suggest that it's a kernel issue and so I'm in the
process of upgrading the kernels but I want to be sure I'm getting this
fixed. Its been a month since the last time before today that I had to
restart the network... so waiting and assuming that it's working isn't
such a great idea. Has anyone encountered this and fixed this sort of
problem?

Please respond to the below e-mail address as well as the group. Thanks
in advance.

--

Ted M Harapat

UNIX Network Administrator
McLeodUSA - InternetWorks Group
http://www.mcleodusa.com

2. keyboard mapping for terminal emulation

3. Newest kernal ? latest kernal for redhat install

4. How does one install an 'Imakefile'

5. ...FreeBSD kernal vs Linux ( say slackware... ) kernal.

6. FAQ wanted

7. BOOT KERNAL WORKS, INSTALLED KERNAL FAILS :(

8. Line drawing characters in text environments

9. HELP : Kernal Panic when rebooting with new Kernal

10. HELP NEEDED to upgrade from kernal 1.2.1 to kernal 2.0.0

11. Network slows to a crawl, possible chipset/driver/kernal bug!

12. SANE 2000 - May 22-25, 2000 (2nd Int'l System Administration & Networking Conference)