On Thu, 2004-02-05 at 20:43, Ian Mortimer wrote:
> Hi Ross
>
> > thanks for this useful information Ian. I've had a look and the
> > information is there.
> > But can you tell me, have you done anything more? Like set it up so that
> > it mails you when there are errors, or set it to run consistency checks?
> > Thanks again,
>
> I have a script like this in /etc/cron.hourly
>
> #!/bin/bash
> status="$(< /proc/rd/status)"
> [[ $status = OK ]] && exit 0
> diff /proc/rd/c0/current_status /proc/rd/c0/initial_status |
> mail -s "Raid status on $(hostname -s) is $status" root
>
> No mail so far so I can't vouch for it's usefulness.
>
> You could run some more detailed checks on /proc/rd/c0/current_status
> or just check for changes in that file.
I use this script; seems to work with most Mylex cards; 170, 250, 1100.
Cokey
--
------------------------------------------------------------------
F. 'Cokey' de Percin, DBA Email:
CSC Work - cdeperci@xxxxxxx
Columbia, South Carolina Home - fdepercin@xxxxxxxxx
--
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.
Mailscanner thanks transtec Computers for their support.
#!/bin/sh
#
# This script checks the Mylex RAID hardware
#
# Note that the software RAID can be checked in a similar manner.
# The /proc/mdstat file(s?) can be checked as necessary
#
# Shouldn't need to touch these...
HOSTNAME=`hostname`
DATE=`date +%m/%d/%y:%H.%M`
# Person to send log activity to.
SYSADMIN=root
# Linux, FreeBSD, BSDI, Sun, etc.
MAIL=mail
# This is the /proc directory for the Hardware RAID info
RBASE="/proc/rd"
# This is where the Hardware RAID status file is
RSTAT=`cat $RBASE/status`
if [ "$RSTAT" = "OK" ]; then
echo "+++++++++++++++++++++++++++++++++++++++++++++" > /var/log/raidreport.log
echo "RAID hardware is OK" >> /var/log/raidreport.log
echo "+++++++++++++++++++++++++++++++++++++++++++++" >> /var/log/raidreport.log
#
# Note that this is turned OFF. I only want to see problems....
#
# cat /var/log/raidreport.log | $MAIL -s "$HOSTNAME $DATE Mylex RAID hardware in mode $RSTAT!!" $SYSADMIN
exit
fi
echo "+++++++++++++++++++++++++++++++++++++++++++++" > /var/log/raidreport.log
echo "WARNING, we have a RAID hardware failure" >> /var/log/raidreport.log
echo "WARNING, /proc/rd/status = $RSTAT" >> /var/log/raidreport.log
echo "+++++++++++++++++++++++++++++++++++++++++++++" >> /var/log/raidreport.log
CX=`ls -d $RBASE/c* | cut -f4 -d "/"`
for cx in `echo $CX`; do
RDSTAT=`grep Critical $RBASE/$cx/current_status`
if [ "$RDSTAT" != "" ]; then
echo "Controller $cx has status $RDSTAT" >> /var/log/raidreport.log
fi
done
cat /var/log/raidreport.log | $MAIL -s "$HOSTNAME $DATE Mylex RAID hardware in failure mode $RSTAT!!" $SYSADMIN
exit 1
