What do these kernel errors in my logwatch mean?

Question

I was reading my logwatch this morning and I've noticed that there are a lot of kernel errors. I'm on CentOS 6 and the machine acts as a web server. Can anyone help me understand what these mean and if I need to take any action?

http://pastebin.com/raw.php?i=RE8sguXD

Fixing these sort of errors is not normally part of my job and as such I have virtually no experience with things like this. I'm particularly worried that the read and write errors may be indicating a disk has failed? Any help is very appreciated!

Something is wrong in the I/O subsystem, disk or controller probably.. — NickW, Apr 29 '13 at 08:25
Well, you know the disk is connected with a cable to the motherboard, on the motherboard there is usually some dedicated circuitry allowing other parts of the computer to talk to the disks. You probably need to do a proper check of your disks. — NickW, Apr 29 '13 at 09:25

score 1 · Accepted Answer · answered Apr 29 '13 at 09:16

1

Most likely you have either a bad disk, or a bad SCSI controller (which is what the disks connect to). If this is happening on a hardware RAID array, the RAID controller might be bad.

answered Apr 29 '13 at 09:16

Falcon Momot

25,244
15
63
92

Great, thanks. Yes the server has RAID so that would make sense. So these problems need fixing and shouldn't be left as is? What I'm asking is: do I need to take any action? – GhostInTheSecureShell Apr 29 '13 at 09:17
Absolutely. If there isn't already filesystem corruption, there will be if you leave it like that. This type of error often happens when the RAID controller (or drives) stop responding to commands. – Falcon Momot Apr 29 '13 at 09:25
1

Nailed it. The RAID controller was bad. Thank you. – GhostInTheSecureShell May 01 '13 at 08:57
Woot. Np. At least you get to keep the drives :) – Falcon Momot May 02 '13 at 01:13

What do these kernel errors in my logwatch mean?

1 Answers1