Bug#498141: general network corruption running linux-image-2.6.26-1-686

09-07-2008, 02:15 PM

Package: linux-image-2.6.26-1-686
Version: 2.6.26-3
Severity: normal

I installed linux-image-2.6.26-1-686 on my etch system to try and help vet
it a bit. I immediately had two problems that look network related. The
first was that xterms I opened on another local machine were garbled so
badly that I could not use them. The fonts were all garbled, the prompt
was messed up, etc.

The second issue was that my IMAP connection to a local machine was also
having serious corruption issues. It would show the headers, but when I
went to look at the messages, everything would go blank. (This was in
kmail.) Reverting back to 2.6.22-6~bpo40+2 made everything happy again.
So I realize I'm assuming two things - 1) that the kernel was at fault - it
was the only thing that changed, and 2) that it was the networking that was
wonky, and not something else.

Here is some info about the hardware:
AMD Athlon(tm) 64 X2 Dual Core Processor 3600+
~> lspci
00:00.0 Host bridge: nVidia Corporation nForce3 250Gb Host Bridge (rev a1)
00:01.0 ISA bridge: nVidia Corporation nForce3 250Gb LPC Bridge (rev a2)
00:01.1 SMBus: nVidia Corporation nForce 250Gb PCI System Management (rev
a1)
00:02.0 USB Controller: nVidia Corporation CK8S USB Controller (rev a1)
00:02.1 USB Controller: nVidia Corporation CK8S USB Controller (rev a1)
00:02.2 USB Controller: nVidia Corporation nForce3 EHCI USB 2.0 Controller
(rev a2)
00:06.0 Multimedia audio controller: nVidia Corporation nForce3 250Gb AC'97
Audio Controller (rev a1)
00:08.0 IDE interface: nVidia Corporation CK8S Parallel ATA Controller
(v2.5) (rev a2)
00:0a.0 IDE interface: nVidia Corporation CK8S Serial ATA Controller (v2.5)
(rev a2)
00:0b.0 PCI bridge: nVidia Corporation nForce3 250Gb AGP Host to PCI Bridge
(rev a2)
00:0e.0 PCI bridge: nVidia Corporation nForce3 250Gb PCI-to-PCI Bridge (rev
a2)
00:18.0 Host bridge: Advanced Micro Devices [AMD] K8 [Athlon64/Opteron]
HyperTransport Technology Configuration
00:18.1 Host bridge: Advanced Micro Devices [AMD] K8 [Athlon64/Opteron]
Address Map
00:18.2 Host bridge: Advanced Micro Devices [AMD] K8 [Athlon64/Opteron]
DRAM Controller
00:18.3 Host bridge: Advanced Micro Devices [AMD] K8 [Athlon64/Opteron]
Miscellaneous Control
01:00.0 VGA compatible controller: ATI Technologies Inc RV350 AP [Radeon
9600]
01:00.1 Display controller: ATI Technologies Inc RV350 AP [Radeon 9600]
(Secondary)
02:07.0 FireWire (IEEE 1394): VIA Technologies, Inc. IEEE 1394 Host
Controller (rev 80)
02:08.0 Ethernet controller: Marvell Technology Group Ltd. 88E8001 Gigabit
Ethernet Controller (rev 13)

~> lsmod
Module Size Used by
radeon 112480 1
drm 76884 2 radeon
binfmt_misc 11592 1
cpufreq_ondemand 8780 1
cpufreq_powersave 2240 0
rfcomm 37272 0
l2cap 23488 5 rfcomm
bluetooth 50404 4 rfcomm,l2cap
ppdev 9156 0
parport_pc 34468 0
lp 11428 0
parport 34568 3 ppdev,parport_pc,lp
autofs4 20996 2
button 8400 0
ac 5636 0
battery 10436 0
ipv6 239908 18
nfs 225644 2
lockd 61448 2 nfs
nfs_acl 3904 1 nfs
sunrpc 159292 4 nfs,lockd,nfs_acl
ext3 122952 1
jbd 56040 1 ext3
mbcache 8772 1 ext3
dm_snapshot 17380 0
dm_mirror 21568 0
dm_mod 52992 2 dm_snapshot,dm_mirror
cpufreq_stats 5568 0
cpufreq_userspace 4640 0
powernow_k8 15104 1
freq_table 4896 3 cpufreq_ondemand,cpufreq_stats,powernow_k8
firewire_sbp2 12676 0
psmouse 36624 0
ide_generic 1600 0 [permanent]
ide_disk 17088 0
snd_intel8x0 32604 2
snd_ac97_codec 93412 1 snd_intel8x0
ac97_bus 2752 1 snd_ac97_codec
snd_pcm_oss 40096 0
snd_mixer_oss 16000 1 snd_pcm_oss
snd_pcm 73092 3 snd_intel8x0,snd_ac97_codec,snd_pcm_oss
snd_seq_dummy 4164 0
snd_seq_oss 29952 0
snd_seq_midi 8608 0
snd_rawmidi 23520 1 snd_seq_midi
snd_seq_midi_event 7488 2 snd_seq_oss,snd_seq_midi
snd_seq 47120 6
snd_seq_dummy,snd_seq_oss,snd_seq_midi,snd_seq_mid i_event
snd_timer 21636 2 snd_pcm,snd_seq
snd_seq_device 8204 5
snd_seq_dummy,snd_seq_oss,snd_seq_midi,snd_rawmidi ,snd_seq
snd 48996 14
snd_intel8x0,snd_ac97_codec,snd_pcm_oss,snd_mixer_ oss,snd_pcm,snd_seq_oss,snd_rawmidi,snd_seq,snd_ti mer,snd_seq_device
k8temp 6016 0
soundcore 7968 1 snd
i2c_nforce2 6272 0
amd64_agp 12612 1
agpgart 32456 2 drm,amd64_agp
shpchp 31636 0
pci_hotplug 29824 1 shpchp
evdev 9792 3
tsdev 8512 0
snd_page_alloc 10568 2 snd_intel8x0,snd_pcm
pcspkr 3584 0
i2c_core 24256 1 i2c_nforce2
rtc 13336 0
xfs 506840 2
ide_cd 37088 0
cdrom 33376 1 ide_cd
generic 5252 0 [permanent]
usbhid 26400 0
hid 25984 1 usbhid
amd74xx 13916 0 [permanent]
ide_core 114756 5 ide_generic,ide_disk,ide_cd,generic,amd74xx
ata_generic 8004 0
skge 39184 0
firewire_ohci 17152 0
firewire_core 39872 2 firewire_sbp2,firewire_ohci
crc_itu_t 2560 1 firewire_core
ehci_hcd 31500 0
ohci_hcd 20228 0
usbcore 126856 4 usbhid,ehci_hcd,ohci_hcd
sd_mod 27776 5
thermal 13896 1
processor 31816 2 powernow_k8,thermal
fan 5252 1
sata_nv 19204 4
libata 117104 2 ata_generic,sata_nv
scsi_mod 137676 3 firewire_sbp2,sd_mod,libata

-- System Information:
Debian Release: 4.0
APT prefers testing
APT policy: (500, 'testing'), (500, 'stable')
Architecture: i386 (i686)
Shell: /bin/sh linked to /bin/bash
Kernel: Linux 2.6.22-4-k7
Locale: LANG=en_US, LC_CTYPE=en_US (charmap=ISO-8859-1)

Versions of packages linux-image-2.6.26-1-686 depends on:
ii debconf [debconf-2.0] 1.5.11etch2 Debian configuration management sy
ii initramfs-tools [linux-initr 0.85i tools for generating an initramfs
ii module-init-tools 3.3-pre4-2 tools for managing Linux kernel mo

Versions of packages linux-image-2.6.26-1-686 recommends:
ii libc6-i686 2.3.6.ds1-13etch7 GNU C Library: Shared libraries [i

--
To UNSUBSCRIBE, email to debian-kernel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

09-07-2008, 08:26 PM

tags 498141 moreinfo
thanks

On Sun, Sep 07, 2008 at 10:15:21AM -0400, Dale E. Martin wrote:
> I installed linux-image-2.6.26-1-686 on my etch system to try and help vet
> it a bit. I immediately had two problems that look network related. The
> first was that xterms I opened on another local machine were garbled so
> badly that I could not use them. The fonts were all garbled, the prompt
> was messed up, etc.

Please use a protocol which uses cryptographic checksums like ssh and
see if something pops up. As the packets go through the TCP checksum,
which is computed over the complete tcp header and text, is _not_
broken, which means that either a higher layer introduces the
corruption, which would show up on much more machines, or that the
devices uses checksum offloading which is broken.

> The second issue was that my IMAP connection to a local machine was also
> having serious corruption issues. It would show the headers, but when I
> went to look at the messages, everything would go blank. (This was in
> kmail.) Reverting back to 2.6.22-6~bpo40+2 made everything happy again.
> So I realize I'm assuming two things - 1) that the kernel was at fault - it
> was the only thing that changed, and 2) that it was the networking that was
> wonky, and not something else.

Can you please use wireshark and get a dump from both sides of such a
corrupted connection?

Bastian

--
Punishment becomes ineffective after a certain point. Men become insensitive.
-- Eneg, "Patterns of Force", stardate 2534.7

--
To UNSUBSCRIBE, email to debian-kernel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

09-07-2008, 09:31 PM

The xterm was running over an "ssh -X" connection:
ssh -X -P -C -f $HOST "xterm -ls -n $HOST"

The NIC is hooked to a gigabit switch but the machine I was connected to
(for both the xterm and imap) is on a 100Mbit switch.

The client machine is a Shuttle SN95G5. (V2.0 I think.)

I'll do the wireshark trick next time I see this.

Thanks,
Dale
--
Dale E. Martin - dale@the-martins.org
http://the-martins.org/~dmartin

--
To UNSUBSCRIBE, email to debian-kernel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

09-07-2008, 10:04 PM

On Sun, Sep 07, 2008 at 05:31:35PM -0400, Dale E. Martin wrote:
> The xterm was running over an "ssh -X" connection:
> ssh -X -P -C -f $HOST "xterm -ls -n $HOST"

Then I doubt that it is a kernel problem at all. The ssh connection
would not survive corruption in deeper layers.

Please use memtest86 to test if you don't have broken memory.

Bastian

--
Those who hate and fight must stop themselves -- otherwise it is not stopped.
-- Spock, "Day of the Dove", stardate unknown

--
To UNSUBSCRIBE, email to debian-kernel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

09-07-2008, 11:05 PM

> Then I doubt that it is a kernel problem at all. The ssh connection
> would not survive corruption in deeper layers.
>
> Please use memtest86 to test if you don't have broken memory.

I'm not going to say it's not possible - I've been using Linux a long time,
and know about compiling floppy.c as a torture test. On the other hand,
this machine has been stable for 3 years, with a variety of kernels, 64
bit, 32 bit, also in windows, is used with high cpu load frequently, etc.
It has been rock solid.

I installed a new kernel, immediately experienced problems, boot back to
the old kernel, and problems go away. Occam points at the new kernel.
(And yes, I get the new kernel could point to an underlying issue, could be
heating the CPU differently, or something - but it doesn't feel like the
most likely scenario to me.)

It could also be that it's not in the network stack but some other kernel
issue.

Thanks,
Dale
--
Dale E. Martin - dale@the-martins.org
http://the-martins.org/~dmartin

--
To UNSUBSCRIBE, email to debian-kernel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org