On 9 Jun 2014, at 9:03 pm, Francesco De Giorgi <francesco.degio...@exact-lab.it> wrote:
> On Sat, Jun 7, 2014 at 4:23 PM, Andrew Beekhof <and...@beekhof.net> wrote: >> >> On 7 Jun 2014, at 12:01 am, Francesco De Giorgi >> <francesco.degio...@exact-lab.it> wrote: >> >>> Hi all, >>> first post here. >>> >>> We are managing an HA Lustre filesystem with pacemaker 1.1.11 and >>> corosync 2.3.3 . >>> I was looking for a way to send email alerts to signal a Lustre target >>> migration, and I considered: >>> >>> - MailTo primitive >>> - cron script to parse crm_mon output >>> - nagios/icinga check >>> - ClusterMon primitive >>> >>> I compiled pacemaker-cli without smtp support, >> >> Ah, but you still have access to: >> >> -E, --external-agent=value A program to run when resource operations >> take place. >> -e, --external-recipient=value A recipient for your program (assuming you >> want the program to send something to someone). >> >> for crm_mon, and your script can do whatever you like :) >> > > Thanks Andrew, > I already tried the -E option on crm_mon and can't understand on what > conditions the external program is called. It should be whenever a resource (other than itself) is stopped or started. > > This is my ClusterMon resource > > primitive resMON ocf:pacemaker:ClusterMon \ > params user=root update=30 extra_options="-E /tmp/foo.sh" \ > operations $id="resMON-operations" \ > op monitor interval=10 timeout=20 on-fail=restart \ > op start timeout=20 interval=0 \ > op stop timeout=20 interval=0 > > clone resmon_clone resMON \ > meta globally-unique=false > > and I can see the crm_mon daemon running this way (under root user) > > /usr/sbin/crm_mon -p /tmp/ClusterMon_resMON.pid -d -i 0 -E /tmp/foo.sh > -h /tmp/ClusterMon_resMON.html > > /tmp/foo.sh is just a touch to a file in /tmp. > > But nothing happens. I straced the pid I found in > /tmp/ClusterMon_resMON.pid and, triggering a migration, found any > reference to my /tmp/foo.sh. > > Am I doing something wrong? > > Thanks again, > Francesco > > >>> so the ClusterMon would >>> be the last option. >>> >>> I was thinking to go with a MailTo primitive for each Lustre target, >>> and colocate them. Is this the easiest way, or something better >>> already exists? >>> >>> Thanks in advance, >>> Francesco >>> >>> _______________________________________________ >>> Pacemaker mailing list: Pacemaker@oss.clusterlabs.org >>> http://oss.clusterlabs.org/mailman/listinfo/pacemaker >>> >>> Project Home: http://www.clusterlabs.org >>> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf >>> Bugs: http://bugs.clusterlabs.org >> >> >> _______________________________________________ >> Pacemaker mailing list: Pacemaker@oss.clusterlabs.org >> http://oss.clusterlabs.org/mailman/listinfo/pacemaker >> >> Project Home: http://www.clusterlabs.org >> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf >> Bugs: http://bugs.clusterlabs.org >> > > > > -- > Francesco De Giorgi > eXact lab s.r.l. > +39 329 1529493 > www.exact-lab.it > > _______________________________________________ > Pacemaker mailing list: Pacemaker@oss.clusterlabs.org > http://oss.clusterlabs.org/mailman/listinfo/pacemaker > > Project Home: http://www.clusterlabs.org > Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf > Bugs: http://bugs.clusterlabs.org
signature.asc
Description: Message signed with OpenPGP using GPGMail
_______________________________________________ Pacemaker mailing list: Pacemaker@oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs: http://bugs.clusterlabs.org
