Topic: Wrong kernel booting after yum update?

[stian]

I see in the boot splash screen that it is the CentOS (2.6.9-34.0.2.ELsmp) kernel that is booting. Is that correct? I did a fresh install from 7.0stable and the yum update. I get an Kernel panic - not syncing: Attempted to kill init! when i boot with the new kernel.

I have an Areca hardware raid card that I must load a driver for during install of 7.0stable. Works fine until the yum update and reboot to the new kernel. Is that the problem? That the driver for the raid-card is not included in the new kernel? Is there any easy way to "copy" the driver over to the new kernel before rebooting?

If I choose to boot with the kernel SME Server (2.6.9-34.ELsmp) at the splash screen all works fine again, but is this "safe"?

[mercyh]

See related thread here

http://forums.contribs.org/index.php?topic=33825.0

and here

http://forums.contribs.org/index.php?topic=33780.0

This may not seem to be related until you look at the bug report on how the kernel is selected after a yum upgrade. See the bug report here.

http://bugs.contribs.org/show_bug.cgi?id=1930

The short answer to your question is, Yes it is safe and you should be booting the SME Server version of the kernel instead of the CentOS version.

Royce H.

[mercyh]

I just reread your post and saw this,

CentOS (2.6.9-34.0.2.ELsmp)

Kernel booted after install.

You then changed to

SME Server (2.6.9-34.ELsmp)

I would suggest that you try booting SME Server (2.6.9-34.0.2.ELsmp)
I am not sure if this will work but it might. (this would put you into the current version of the kernel.)

[stian]

The SME Server (2.6.9-34.0.2.ELsmp) kernel is not one of the options at my boot screen.

I only have:

CentOS (2.6.9-34.0.2.EL)
CentOS (2.6.9-34.0.2.ELsmp)
SME Server (2.6.9-34.EL)
SME Server (2.6.9-34.ELsmp)

[mercyh]

what does uname -a show?

If it is Linux <servername> 2.6.9-34.ELsmp I believe you are on the latest SME7 kernel.

The last option on your screen

CentOS (2.6.9-34.0.2.EL)
CentOS (2.6.9-34.0.2.ELsmp)
SME Server (2.6.9-34.EL)
SME Server (2.6.9-34.ELsmp)

Is correct and the latest kernel for a multi processor system.

If this boots and works correctly you should not need to do anything with the raid card drivers to get onto the current kernel.

I believe GRUB will boot this kernel from now until another update or a different kernel is manually selected at startup

[stian]

It shows:

Linux vinstra 2.6.9-34.ELsmp #1 SMP Wed Mar 8 00:27:03 CST 2006 i686 i686 i386 GNU/Linux

[stian]

Just found another post where someone has posted the output of uname -a:

Linux sme7 2.6.9-34.0.2.EL #1 Fri Jul 7 19:24:57 CDT 2006 i686 i686 i386 GNU/Linux

From this thread: http://forums.contribs.org/index.php?topic=33825.0

So it looks like he has a newer kernel than I do...

[mercyh]

He was booting the CentOS version of the kernel. If you go back into that post he ended up at the same place as you are now booting the

SME Server (2.6.9-34.ELsmp)  Kernel.

How do I know? That was me.  

(That just goes to show I should have known better then this post:

I just reread your post and saw this,

Quote:
CentOS (2.6.9-34.0.2.ELsmp)

Kernel booted after install.

You then changed to

Quote:
SME Server (2.6.9-34.ELsmp)


I would suggest that you try booting SME Server (2.6.9-34.0.2.ELsmp)
I am not sure if this will work but it might. (this would put you into the current version of the kernel.))

But I couldn't quite remember how the boot screen looked and didn't want to reboot my production server to see.


Good day and Good Luck with SME,

Royce Holdeman

[stian]

Thanks!

[byte]

I would suggest that you try booting SME Server (2.6.9-34.0.2.ELsmp)
I am not sure if this will work but it might. (this would put you into the current version of the kernel.)

See this bug here it would seem the Centos 2.6.9-34.0.2.ELsmp is not being renamed to SME Server...

http://bugs.contribs.org/show_bug.cgi?id=1996

[mercyh]

NOW MY HAIR IS TURNING GREY


CentOS (2.6.9-34.0.2.EL)
CentOS (2.6.9-34.0.2.ELsmp)
SME Server (2.6.9-34.EL)
SME Server (2.6.9-34.ELsmp)

byte,

Per your post we should be booting option #2-CentOS (2.6.9-34.0.2.ELsmp). Is the bug confirmed well enough that we KNOW this is correct?

If so stian does have a driver issue with his raid card.

Royce Holdeman

[stian]

Ok...

I have tried this I found on the areca faq site:
http://faq.areca.com.tw/modules/smartfaq/faq.php?faqid=224

I went through this before doing the yum update. Also did the sam for smp-kernel. It did not help. Is there anything I should modify to this to get it work on smeserver?

Also this:
http://faq.areca.com.tw/modules/smartfaq/faq.php?faqid=191

Is there any way to describe how i can do this?

[byte]

Is the bug confirmed well enough that we KNOW this is correct?

I would have said so but I'm no core dev.

I think from seeing various posts on here it is, but we really need people to report these to the Bug Tracker and/or open a new bug if it doesn't exist.

[mercyh]

Stian,

We need to wait for an answer from Byte on the kernel version before we get to far along with this. If the SME kernel should be at version 2.6.9-34.0.2 you have an issue that you are not running the latest kernel. I and I think Byte also was under the impression that the latest kernel in SME was 2.6.9-34.

Byte has much more experiance then I do and I am suspicious that you do also (I have never attempted to build a 3tb server)

In the meantime does your server boot and work without errors with the old SME 2.6.9-34 kernel?

Royce H.

[byte]

I and I think Byte also was under the impression that the latest kernel in SME was 2.6.9-34.

No, the latest in the default yum repo's are with 2.6.9-34.02 which when downloaded via YUM are left as Centos label and not SME Server.

Please report your findings via the bug tracker the more people who report the better outcome the bug will have.