Subject: Re: [vserver] Reporting a bug
From: Herbert Poetzl <herbert@13thfloor.at>
Date: Wed, 28 Oct 2009 07:51:09 +0100

On Tue, Oct 27, 2009 at 10:08:33PM -0500, Ramkumar Cherukuri wrote:
> ---------- Forwarded message ----------
> From: Ramkumar Cherukuri <ramkumar.ch03@gmail.com>
> Date: Sun, Oct 25, 2009 at 3:09 PM
> Subject: Join in this mailing list and Reporting a bug
> To: vserver-subscribe@list.linux-vserver.org

> Hi

> I am new to this mailing list.

welcome to the ML then :)

> I got a problem with vserver kernel with new hardware. 
> I would like to report a bug and i need help to rectify my problem.

> Here i attached my bug report, if you need any test output or 
> files please let me know.

Good start, contains most of the information required to
identify an issue and/or narrow it down, but it would be
better to inline it (ASCII) instead of attaching it, so
that we can comment on it ... 

> I am looking forward to hear from you.

anyway, I copied the relevant parts and will comment on them

| Problem (rewritten, please correct me if necessary):
|
|  1. Sometimes when I restart/reboot the system, it starts booting
|     Linux but suddenly powers off like somebody had shut it down.
|  2. Sometimes the system gets stuck and I'm unable to use the
|     mouse or keyboard.
|  3. (Sometimes?) the Kernel does not identify the the nic (e1000).


| Description:
|
| Recently we got new DELL Optiplex 760 machines. We tried to installed
| one server for our network. First we installed CentOS 5.3 and later i
| installed linux vserver kernel. We want the system boot in to vserver
| kernel. If we boot the system with normal CentOS 5.3, It is working
| fine. But If we boot the system with VServer Kernel, with out getting
| home page the system goes to turn off as some one shut it down. This
| happens for 3 to 4 times every time we resart the system.

it seems to be a fairly new system, and I'd be interested
in the centos 5.3 kernel version everything is working
with ...

| Suddenly i don't know what happened to system i could not use my
| keyboard and mouse. Then i restarted it then is working fine.

it would help to have the bootup log (via serial console
or similar) in both cases, a 'working' bootup and a failing
one

| And the system could not identify the ethernet card e1000. Then i
| installed other ethernet card e10/100( below i gave both), then the
| new one is detected and it is working fine.

| ETHERNET CARDS
|
| Intel Corporation 82567LM-3 Gigabit Network Connection (rev 02)

this one requires the e1000e module

| Lite-On Communications Inc LNE100TX [Linksys EtherFast 10/100] (rev 25)

this one is handled by the tulip driver

| KERNEL VERSION

| Linux version 2.6.22.19-vs2.3.0.34.1 (mockbuild@x86.builders.hozac.com)
|  (gcc version 4.1.2 20070626 (Red Hat 4.1.2-14)) #1 SMP
|  Mon Mar 17 05:32:04 EDT 2008

| Linux 2.6.22.19-vs2.3.0.34.1 i686/0.30.216

| PROCESSOR INFORMATION
| model name	  : Intel(R) Core(TM)2 Duo CPU     E8400  @ 3.00GHz
| flags 	  : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca
|		    cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht
|		    tm pbe nx lm constant_tsc pni monitor ds_cpl vmx smx est
|		    tm2 ssse3 cx16 xtpr lahf_lm

the cpu flags state that the machine is 64bit capable,
but the util-vserver information shows 32bit tools/kernel?

| MODULES

| cpufreq_ondemand 12300 2 - Live 0xf8c1c000
| acpi_cpufreq 14472 0 - Live 0xf8c17000
| [rest zapped]

all in all, I'd suspect a mainline issue, which either
got fixed by the centos folks (for their kernel) or
by mainline, if they use a more recent version


I'd suggest to do the following:

 - remove CPU scaling and other desktop features
 - update the kernel to 2.6.31.x (with the appropriate
   Linux-VServer patch)
 - switch to 64bit (kernel and host userspace)

if the issue persists, please provide the bootup logs
and enable a watchdog (software) to get a trace when
the kernel 'hangs'

HTH,
Herbert


> Thank you

> ---Ram