Debian Reports Non-optimal RAID Status

carlosinfl · 03-03-2011, 02:14 PM

I checked /var/log/messages today and it's flooding with:

Code:

Mar  3 14:58:36 mail mpt-statusd: detected non-optimal RAID status

The system is a fresh install and not built on RAID. It's a single disk on a VM (VMware) and I don't think 'mdadm' is even installed:

Code:

mail:/var/log# apt-cache policy mdadm
mdadm:
  Installed: (none)
  Candidate: 3.1.4-1+8efb9d1
  Version table:
     3.1.4-1+8efb9d1 0
        500 http://ftp.us.debian.org/debian/ squeeze/main amd64 Packages

I searched before I posted and found an old bug from 2009:

http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=539266

I can't understand why I'm seeing this on my new system with no RAID configuration and using Debian 6.0.

wpeckham · 03-03-2011, 03:49 PM

Interesting? RAID is not optimal when there are no RAID volumes, eh?
I have no answers, but am sure looking forward to seeing some.

Single volume raid: boggles the mind.

murmur101 · 03-04-2011, 09:16 AM

Got the same here..

debian 2.6.32-5-amd64
used atomic layout (no raid) in setup in a VM on an esxi 4.0
my /var/log/messages is peppered with " mpt-statusd: detected non-optimal RAID status"

I re-installed 5 times.. same thing

looking forward to see an answer to this one..

carlosinfl · 03-04-2011, 09:43 AM

I've looked all over and don't understand it. You configure the VM with x amount of disk space and that should be transparent to the O.S. (Debian 6.0) if it's local or remote RAID storage. If you have a RAID controller, it should present the storage as one solid disk and then you can carve it up as you feel fit.

Super confused.

I've tried RHEL 6, Arch Linux, Ubuntu Server 10.10, Slackware, & Gentoo and Debian is the only one with this problem. It's flooding my logs to the point that I can't even run this as a O.S. I also reinstalled 3 times.

carlosinfl · 03-07-2011, 12:52 PM

If you disable 'mpt-status' in init.d scripts, that seems to work:

Code:

/etc/init.d/mpt-status stop

Anyone know how to disable this after I reboot so it doesn't start again?

murmur101 · 03-08-2011, 02:10 AM

Hi,

I think this is correct

Code:

/etc/init.d/mpt-statusd stop

I disabled it using

Code:

update-rc.d-insserv -f mpt-statusd remove

works for me at least

carlosinfl · 03-10-2011, 01:32 PM

I don't know if there are any negative implementations to disabling this service or why Debian even elected to install / use this to begin with. Anyone know? Obviously I'm running this on a server with a hardware RAID controller from Dell (PERC 5i) so there is no need to run software RAID or 'mdadm'.

wmussatto · 03-23-2011, 03:18 PM

I may be missing something, but mpt-statusd deals with hardware raid. If you are getting message from your raid them perhaps the raid pair is either being rebuilt or needs to be? mdadm deals with software raid. I think at least some of the Dell Hardware Raid products use LSI chips or boards and that is what mpt-statusd is looking at.

I know that on our server I had to manually during bootup force the LSI card to start the rebuild of our array. It continued to do so through a series of reboots until the job was done. I got an email saying it was in "non-optimal status" and then one that said it was fixed. Hope this helps.