FC5 - recently introduced system hangs, multiple systems

Hi,

First off sorry for the long post, but I want to try and get everything
down systematically.

1) I built a desktop system for use as a file server on my home network.
It was built with the FC 5 cd's, I put Samba, Gnome and Kde and some fun
stuff on the machine, and ran yum update on the system. This all happened
about 3 months ago. As far as I can tell from the logs, the kernel in use
after the yum update was 2.6.17-1.2157. The system ran for a couple of
months, entirely fault free.

2) A couple of weeks ago I needed to re-build my work laptop. Having had
such a good experience with the desktop, I decided to put FC 5 onto the
laptop. The installation was completed again with Gnome and Kde and some
other stuff. then I ran yum update and updated ~200 packages on the
system. I think the kernel loaded by yum update was 2.6.18-1.2200.

3) Right from the start the laptop has exhibited hangs typically about
1 per day - the system will lock up entirely, no mouse movement, no
ability to restart X, no ability to ctrl-alt-del, no network response,
and no messages in any system log files. One minute I'm using the
desktop as usual, the next it locks up. I tried runing with the previous
kernel (in this case the FC 5 install kernel which I think is
2.6.15-1.2054) but the hangs still occured. As this machine is my work
system I was 'working' - so my desktop was pretty heavily used with a
typical mixture of xterms, probably eclipse, and most certainly firefox.
Over the weeks I've tried substituting konqeror for firefox, but the
problem still exists.

4) Stupidly ... the same day that I built the laptop, I reliased that
I'd not run yum update on my home desktop system for a bit (probably
2 or maybe 3 months), so I ran it that evening. yum update brought
down ~150 packages, and the update completed succesfully. The kernel
running on my home system was the same as the laptop. Guess what -
now the desktop has started hanging. The hangs seem to be the same
as those exhibited by the laptop - the system is working normally
one minute ... then no response.

5) The desktop is mainly used as a file server but I do do some
'playing' on this system and most of the hangs have occurred whilst
I'm actively using the destop, although I have experienced at least
two hangs that occurred with me not even using the desktop, the
system was merely functioning as a file server.

6) I've also just used yum update again which brought up 4 packages
including a new kernel 2.6.18-1.2239. If anything this seems to be
*more* unstable and hangs more frequently than any of the other kernels
that I've tried. But as I said even running the old, previously stable
kernel (2.6.17-1.2157) I still get hangs.


Hardware

The destop is a Compaq Desktop EN, with 384 Mb of memory, multiple
IDE disks bound together with the MD driver, and an XFS file system
on it. The network card is an e1000. Sound is enabled, but as the
bloodly thing has just crapped out again I'll have to wait before
I find out what the sound card is. I'll probably just include a list
of /proc/modules at the bottom ...

The laptop is a Dell Lattitude D610, with 512 Mb memory, a single
IDE disk, ext3 file system, and raw disk partitions (I removed the
lvm stuff). The network card is a built in 3com something.

None of the hardware on either system has been changed for several
months, and as both systems also run other operating systems without
problem I assume that the hangs are not caused by flaky hardware.

Also as the systems have very different hardware, I have to assume
that it's not anything periperally related (such as the network card
driver as both systems have diferent nics). This must be a problem
with some kind of core subsystem.

The problem has started recently (as the desktop previously ran
error free), but I can not find any other reports of this kind of
problem via any of the usual web channels - I regard this as VERY
strange. Normally when I experience problems like this I can come
up with some other examples on the net.


So has anybody got any ideas about

a) what could be causing the problem

b) why it has just started

c) what to do to stop it happening again.

These hangs are basically rendering both these systems unusable,
and I'll have to do *something* about the problem one way or another.




-- Chris


PS this is the /proc/modules from the destop - I can't get to the
laptop today ...


nfsd 216433 17 - Live 0xd8ce5000
exportfs 6081 1 nfsd, Live 0xd8a74000
lockd 62152 2 nfsd, Live 0xd8b6c000
nfs_acl 3905 1 nfsd, Live 0xd8a7c000
sunrpc 153597 12 nfsd,lockd,nfs_acl, Live 0xd8c87000
autofs4 21573 1 - Live 0xd8b39000
ip_conntrack_netbios_ns 3393 0 - Live 0xd8a7a000
ipt_REJECT 5697 1 - Live 0xd8a77000
xt_state 2625 8 - Live 0xd8a2d000
ip_conntrack 52085 2 ip_conntrack_netbios_ns,xt_state, Live 0xd8b4a000
nfnetlink 7513 1 ip_conntrack, Live 0xd8a28000
xt_tcpudp 3521 10 - Live 0xd8a2b000
iptable_filter 3393 1 - Live 0xd89ad000
ip_tables 12937 1 iptable_filter, Live 0xd8a66000
x_tables 14405 4 ipt_REJECT,xt_state,xt_tcpudp,ip_tables, Live 0xd8a61000
xfs 517221 1 - Live 0xd8b89000
dm_mirror 29073 0 - Live 0xd8a6b000
dm_mod 57433 1 dm_mirror, Live 0xd8ae8000
raid456 118353 1 - Live 0xd8a80000
xor 15689 1 raid456, Live 0xd8a3a000
video 17221 0 - Live 0xd8a34000
sbs 16257 0 - Live 0xd8a2f000
i2c_ec 5569 1 sbs, Live 0xd8a25000
container 4801 0 - Live 0xd89f6000
button 7249 0 - Live 0xd89ed000
battery 10565 0 - Live 0xd8a21000
asus_acpi 16857 0 - Live 0xd89fa000
ac 5701 0 - Live 0xd89f0000
ipv6 245985 16 - Live 0xd8aaa000
lp 13065 0 - Live 0xd89a8000
parport_pc 27493 1 - Live 0xd8979000
parport 37001 2 lp,parport_pc, Live 0xd89af000
floppy 57317 0 - Live 0xd89de000
uhci_hcd 23885 0 - Live 0xd899b000
e100 36425 0 - Live 0xd8965000
mii 5825 1 e100, Live 0xd8936000
e1000 119761 0 - Live 0xd89bf000
serio_raw 7493 0 - Live 0xd8924000
snd_intel8x0 32605 1 - Live 0xd8970000
snd_ac97_codec 91361 1 snd_intel8x0, Live 0xd8983000
snd_ac97_bus 2753 1 snd_ac97_codec, Live 0xd8931000
snd_seq_dummy 4293 0 - Live 0xd892e000
i2c_i810 5957 0 - Live 0xd892b000
snd_seq_oss 32705 0 - Live 0xd895c000
i2c_algo_bit 9033 1 i2c_i810, Live 0xd8927000
snd_seq_midi_event 8001 1 snd_seq_oss, Live 0xd8816000
i2c_core 21697 2 i2c_ec,i2c_algo_bit, Live 0xd8855000
snd_seq 51633 5 snd_seq_dummy,snd_seq_oss,snd_seq_midi_event, Live 0xd894e000
snd_seq_device 8781 3 snd_seq_dummy,snd_seq_oss,snd_seq, Live 0xd8862000
snd_pcm_oss 42849 0 - Live 0xd8874000
snd_mixer_oss 16833 1 snd_pcm_oss, Live 0xd885c000
snd_pcm 76485 3 snd_intel8x0,snd_ac97_codec,snd_pcm_oss, Live 0xd893a000
intel_rng 5185 0 - Live 0xd8811000
snd_timer 23237 2 snd_seq,snd_pcm, Live 0xd883e000
snd 52933 11 snd_intel8x0,snd_ac97_codec,snd_seq_oss,snd_seq,snd_seq_device,snd_
pcm_oss,snd_mixer_oss,snd_pcm,snd_timer, Live 0xd8866000
soundcore 10145 1 snd, Live 0xd883a000
snd_page_alloc 10569 2 snd_intel8x0,snd_pcm, Live 0xd8836000
pcspkr 3521 0 - Live 0xd8814000
raid1 22849 1 - Live 0xd8819000
ext3 129993 2 - Live 0xd8901000
jbd 58473 1 ext3, Live 0xd8845000




Chris Lidgates private email |
This email address is |
lidgaca AT lycos DOT com |