Atomicorp

Hi,

I'm getting a Kernel Panic on our server when the box is re-booted. Reverting back to an older kernel fixes this issue.

My question is... should I remove the error prone kernels as recommended by one of the techs where the server is located?

doing a rpm -qa | grep kernel lists the following:

kernel-2.6.9-34.EL
kernel-2.6.9-67.0.1.EL
kernel-smp-2.6.9-67.0.22.EL
kernel-smp-2.6.9-55.0.6.EL
kernel-smp-2.6.9-67.0.4.EL
kernel-smp-2.6.9-67.0.1.EL
kernel-2.6.9-67.0.15.EL
kernel-smp-2.6.9-78.0.1.EL
kernel-2.6.9-78.0.1.EL
kernel-2.6.9-78.0.5.EL
kernel-2.6.9-78.0.8.EL
kernel-smp-2.6.9-42.0.3.EL
kernel-2.6.9-42.0.3.EL
kernel-2.6.9-55.0.2.EL
kernel-2.6.9-55.0.6.EL
kernel-smp-2.6.9-55.0.12.EL
kernel-2.6.9-67.0.4.EL
kernel-smp-2.6.9-67.0.7.EL
kernel-smp-2.6.9-67.0.20.EL
kernel-utils-2.4-14.1.117
kernel-smp-2.6.9-78.0.5.EL
kernel-smp-2.6.9-78.0.8.EL
kernel-smp-2.6.9-34.EL
kernel-2.6.9-55.0.12.EL
kernel-2.6.9-67.0.20.EL
kernel-2.6.9-67.0.22.EL
kernel-smp-2.6.9-55.0.2.EL
kernel-2.6.9-67.0.7.EL
kernel-smp-2.6.9-67.0.15.EL

and currently uname -r reports:

2.6.9-34.EL

which appears to be running fine.

Thanks

Woah, I'd start removing a bunch of unused kernels. Haven't you filled up /boot?

yeah, its pretty massive. I've had no errors saying that /boot is full.. actually.. it doesn't look like there is a separate partition for /boot. I'm guessing as I just ran df in conole and this came up:

Filesystem 1K-blocks Used Available Use% Mounted on
/dev/sda1 151771264 28384760 115676916 20% /
none 517352 0 517352 0% /dev/shm

My problem is that I only have remote access to the server so if I restart the box it may never come back online. I might need to pay to get the server techs to remove the erroneous kernels.

No problem there, you can set a kernel to boot once, and if it fails, it will automatically reboot you into another kernel. We documented it in the wiki here:

http://www.atomicorp.com/wiki/index.php/Kernel_Panic