Sophie

Sophie

distrib > Mandriva > current > i586 > media > contrib-release-src > by-pkgid > 330d23739d2091aabf309a2b6b951b73 > files > 1

megactl-0.4.1-8mdv2010.0.src.rpm

MegaCTL for Mandriva
--------------------

Mandriva provides some integrated tools to periodically check the status of
all RAID arrays controlled by the RAID controller cards using LSI MegaRAID.

There are 2 tools :
 * megactl for SCSI disks
 * megasasctl for SAS disks

There are no simple way to define which tool should be used. Therefore, you
have to check yourself. To do that just start each command and check the
output.

To do that :
 * start a terminal,
 * you must be root (so su - if necessary),
 * type :
 megactl
 * if you get an error, you do not have a MegaRaid SCSI card, try the other
   command,
 * type :
 megasasctl
 * if you still get an error, you do not have any MegaRAID based card OR you
   have a MegaRAID card which is not supported yet by the open source tools.

Now that you now which MegaCTL command is good for you, you can modify the
good file.

If the megactl command was the one working for you, then edit the
/etc/megactl/megactl-cron.conf file and replace ENABLE=0 by ENABLE=1 in it.

If the megasasctl command was the one working for you, then edit the
/etc/megactl/megasasctl-cron.conf file and replace ENABLE=0 by ENABLE=1 in it.

Now cron will run a script each 10 minutes. On start the status of the RAID
is checked. If the RAID array is broken, then a log is written in the syslog
and a mail is sent to root. As long as the status of the RAID array is bad, a
log is written every 10 minutes in the syslog and a mail is sent every
2 hours to root.

As long as the status of the RAID array is bad, this last procedure is
repeated. A log and a mail are also provided when the status of the array has
changed. The maximum latency of messages and log on status change is
10 minutes.

Adam Cécile <acecile@mandriva.com>