0

I have, for better or worse, been put in charge of a big-ass server running RHEL6. It has six 1.5 TB drives in a RAID5 configuration. I understand the theoretical concepts behind RAID, and I access this server exclusively through the command line. The server itself is housed in a data center to which access is possible--but not easy--for someone of my job tier. Thus routine physical inspection of the server is probably not an efficient avenue if software monitoring is acceptable.

When a hard drive fails, will I, as a command-line user, be alerted to this automatically during routine shell usage? If so, how? If not, what tool is recommended for simple monitoring and alerts?

1
  • 3
    How's the RAID configured? Software RAID, or a hardware controller? Jun 19, 2012 at 3:23

1 Answer 1

3

Assuming you are using hardware RAID, RHEL itself only sees the logical drive presented by the RAID. It is unaware of the underlying physical disks.

If it is a Dell server you can use OpenManage to run a script for an alert that will write to your console with wall or email you.

http://www.howtogeek.com/50555/setup-email-notifications-for-dell-server-hardware-alerts/

Here is some info on doing this on an HP server http://blog.mattbrock.co.uk/2010/01/25/monitoring-diskraid-hardware-in-centos-5-on-hp-dl360-servers/

If you have a different hardware provider they should have a similar functionality.

If you are using software raid there is a good article here http://my.sohost.eu/knowledgebase/4/Monitor-Linux-software-RAID-array-with-e-mail-notification.html about monitoring /proc/mdstat for changes with a cron and sending an email if it does.

0

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .