Topic: [UNIQUE TOPIC] Kernel panic - not syncing (with RAID ?)

[RvLardin]

Hi to all,

Once again, I made the mistake of thinking that we are alone in our little corner. Since 6 monthes we lost several servers with this common tag ; "Kernel panic - not syncing".
On several hardware with Intel or AMD, until last week we were guessing that it has something to do with RAID, but today we lost our first server without any RAID (declared, but still exist in system since SME 7.0 is considering all LVM arrays in the same way). All our losts servers uses SATA (most of time large Maxtor HD). Other harware where various. There is nothing else in the logs that we can found to help us.

This topic is now usefull to use the force of the community to fix this *important* flaw. The main prob to fix that is that the prob could appends after a few hours or after a couple of monthes : so it is very difficult to know if it is fixed or not.

There's many bugtracks or thread on the same theme :

Bugs :
====
885 => seems to disapear if sound is "off" in BIOS
1901 => seems to be solved by using Kernel 2.6.9-34.0.2.EL
To be continued :  there's much more (at least 5/6 on my last search, our first was #1440) but the search is broken this morning in the BugTracker

Threads :
======
http://forums.contribs.org/index.php?topic=34243.0
http://forums.contribs.org/index.php?topic=34129.0       (IDE Drive)
http://forums.contribs.org/index.php?topic=33801.0
http://forums.contribs.org/index.php?topic=33793.0
http://forums.contribs.org/index.php?topic=33556.0
http://forums.contribs.org/index.php?topic=32023.0
http://forums.contribs.org/index.php?topic=30927.0
http://forums.contribs.org/index.php?topic=30807.0
http://forums.contribs.org/index.php?topic=30813.0

Please post your experiences / ideas here.

A+,
RV.

[gregswallow]

I haven't thad that problem, and I haven't looked at the bug reports, but I would think installing the newer kernel in smeupdates-testing would be something to try - that is almost the latest CentOS kernel.

FYI, there are problems like this with RHEL4 and CentOS, who supply the kernel rpm:
http://www.google.ca/search?q=site%3acentos%2eorg+%22Kernel+panic+%2d+not+syncing%22
http://www.google.ca/search?q=site%3aredhat%2ecom+%22Kernel+panic+%2d+not+syncing%22

[wjhobbs]

RV,

For the disks you are having problems with, are they SATA 150 or 300?

John

[JonB]

I had very similar issues with SATAII drives, as wjhobbs has alluded to in the previous message.

In one case I had problems installing SME onto a pair of SATAII (300) drives. It would install ok but would kernel panic on reboot.

In the second case I installed SME7 on a pair of SATAII drives ok and the server ran very stable until I tranferred a large amount of data onto the server. The server crashed with kernel panic.

After googling around a bit I found this appeared to be reasonably common with SATAII drives and the recommendation was to jumper the drives for SATAI (150). Once I did this and re-installed the server has been steady as a rock and I can transfer large amounts of data to my hearts content.

Jon

[chris burnat]

I

found this appeared to be reasonably common with SATAII drives and the recommendation was to jumper the drives for SATAI (150). Once I did this and re-installed the server has been steady as a rock and I can transfer large amounts of data to my hearts content.

Gooday John, This is a bit of a worry.  I have now deployed 4 servers using SATA configured at 3Gb, I take it this is SATA2?  My ignorance, I know.  Any way to jumber pack to 1.5Gb without reinstalling?

Or am I on the wrong track?
Thanks. chis

[CharlieBrady]

This topic is now usefull to use the force of the community to fix this *important* flaw.

This poll won't fix anything. If you want to see a fix, then open a bugzilla bug, and then someone might address it.

There's many bugtracks or thread on the same theme :

Bugs :
====
885 => seems to disapear if sound is "off" in BIOS
1901 => seems to be solved by using Kernel 2.6.9-34.0.2.EL

Neither of those match the symptoms you mention here.

Please report your problems to the bug tracker.

[Snoopyski]

Hello,

Wich Motherboard an Memory Brand Name do you use for your server ?

(Quality ??? Original ???)

I have a lot of server running SME 7 in RAID1 in SATA mode and I never get this error message. The only one time I get a similar message error is when I had a BAD (defect) SATA cable...

So, please check your Hardware...

(If that can be helpfull... I always use Western Digital Hard Drive... One or two years ago I removed Maxtor of my life !!!)

Thanks,

Snoopyski

[p-jones]

Seagate 250Gb ST3250824

[chris burnat]

I had very similar issues with SATAII drives, as wjhobbs has alluded to in the previous message. [Snip] .  Jon

John, I have taken the liberty of posting an item on Bugzilla on this subject, in this way it will be documented.  Hope its OK.

Others may wish to follow suite and fill bug reports about their particular experiences.  Items on Contribs tend to fade away, or sometime even vanish... Bugzilla is here to stay and provide the Devs with a centralised repository of information.

[Tillebeck]

I experience this problem on a single 1 cpu server with dual SATA 250GB disks.

Sometimes after updating the server it goes into Kernel Panic after reboot. I can fix it by hooking up a monitor and a keyboard. I can then access a boot menu when pressing any key after the nice pinguin shows up on a white screen. I choose to boot SME SERVER UP and that will boot the server just fine.

After choosing the SME SERVER UP in the boot menu I can boot the server over and over with no problems. Only after some sme server updates the server will go into Kernel Panic and I will have to choose the SME SERVER UP in the boot menu.

Hope my findings can be used.

BR. Anders

[chris burnat]

I experience this problem on a single 1 cpu server with dual SATA 250GB disks.

Sometimes after updating the server it goes into Kernel Panic after reboot. I can fix it by hooking up a monitor and a keyboard. I can then access a boot menu when pressing any key after the nice pinguin shows up on a white screen. I choose to boot SME SERVER UP and that will boot the server just fine.

After choosing the SME SERVER UP in the boot menu I can boot the server over and over with no problems. Only after some sme server updates the server will go into Kernel Panic and I will have to choose the SME SERVER UP in the boot menu.

Hope my findings can be used.

BR. Anders

Anders, your findings would be very useful in time if you reported them the the bugtracker, please go to Bugzilla and fill a NEW bug report.  As it is, there are already two bugs at least covering various aspects of this problem, Bug 2018 and 2028.  What you have noted appears different again.. I am sure someone there will ask you a number of relevant questions to try to pin down this issue. To lodge a new bug report, please go to:

Code: [Select]

http://bugs.contribs.org/enter_bug.cgi
you will have to log in to post.
Thanks. chris.

[Tillebeck]

Anders, your findings would be very useful in time if you reported them the the bugtracker, please go to Bugzilla and fill a NEW bug report. To lodge a new bug report, please go to:

Code: [Select]

http://bugs.contribs.org/enter_bug.cgi
you will have to log in to post.
Thanks. chris.

Thanks for the soft push
I have created a new bug on the issue:
http://bugs.contribs.org/show_bug.cgi?id=2047

BR. Anders