There is also a memory segment you can look at along with unirpcd and
the spooler daemon. This is what's in the shell script that's plugged
into our HA cluster software for restarting UV after a failover:
-----
# See if UV is running already
PSARGS=ax
olist=`/bin/su root -c "/usr/bin/ipcs -s"|grep 0xace |cut
-c11-21|sort|uniq`
mlist=`/bin/su root -c "/usr/bin/ipcs -m"|grep 0xaceb|cut
-c11-21|sort|uniq`
nlist=`/bin/su root -c "/usr/bin/ipcs -m"|grep 0xaceb|cut
-c11-21|sort|uniq`
llist=`/bin/su root -c "/usr/bin/ipcs -q"|grep 0xacea|cut
-c11-21|sort|uniq`
uvseg=`/bin/su root -c "/usr/bin/ipcs -m"|grep 0xaceca|cut
-c11-21|sort|uniq`
klist=`/bin/su root -c "/bin/ps $PSARGS"|grep usd|grep -v grep|cut
-c0-5`
dlist=`/bin/su root -c "/bin/ps $PSARGS"|grep uvdlockd|grep -v grep|cut
-c0-5`
vlist=`/bin/su root -c "/bin/ps $PSARGS"|grep unirpcd|grep -v grep|cut
-c0-5`
# Exit script if UV is already running
if [ "$uvseg" != "" ] && [ "$klist" != "" ] && [ "$dlist" != "" ] \
&& [ "$vlist" != "" ]; then
printf STDERR "LifeKeeper: restore successful for $TAG\n";
exit 0
fi
-----
IIRC, I adapted it from the uv.rc start/stop script in /etc/init.d. I
don't recall why I'm ignoring all the other memory segments aside from
uvseg. My guess is having one of the other segments missing isn't
necessarily an indication that UV is not running.
-John
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Jeff Schasny
Sent: Friday, January 22, 2010 11:37 AM
To: U2 Users List
Subject: Re: [U2] Any nagios plug-ins for UV?
I've been mulling what one could use as a sign of Universe being "alive"
and all I can come up with (on Unix/Linux anyway) is the presence of the
uvdlockd daemon. Anyone have any other ideas?
_______________________________________________
U2-Users mailing list
[email protected]
http://listserver.u2ug.org/mailman/listinfo/u2-users
