[Linux-HA] trouble with MON

[Andrew Beekhof]

On 10/12/06, patrice pariel <ppariel at ch-lagny77.fr> wrote:
> hello
> i 'm able to do failover with heartbeat + drbd : it works fine now  ;-)
> but i don't managed to monitor one service...

use 2.0.x of Heartbeat in crm-mode.  it can do the monitoring for you.

> In fact, if i like to do a failover, i need to run /etc/init.d/heartbeat
> stop .... manually.
> But heartbeat supervize several services (apache2; vsftpd, postfix) and
> if one of them have a trouble, the ressource stay on the same node
> without restarting the ressources .
> So, i ve heard mon can help me to supervize all services ...does anyone
> can help me to do that.
> I 've read man + faq...i think i ve understand how it works but it fails.
>
> my mon.cf  # for apache2
> ***********
> cfbasedir   = /etc/mon
> alertdir   = /usr/lib/mon/alert.d
> mondir     = /usr/lib/mon/mon.d
> statedir    = /var/state/mon
> pidfile     = /var/run/mon/mon.pid
> maxprocs    = 20
> histlength  = 100
> randstart   = 60s
> serverbind  = 127.0.0.1
> logdir     = /var/log/mon
> dtlogging  = yes
>
> hostgroup www HEARTBEAT.cluster.com
> watch www
>         service http
>         interval 5s
>              monitor http.monitor -p 80
>              depend boxes:Ping_Check
>              period wd {Mon-Sun} hr {0am-24pm}
>              alertafter 2m
>              alertevery 2h
>              alert mail.alert ppariel at ch-lagny77.fr
>
>
> QUESTION
> ***********
> 1 /what do i put for hostgroup : server name (localhost) ? CNAME of the
> ressource ? CN for the cluster?
>
> 2/ MURPHY:/home/patrice# /etc/init.d/mon restart
> Restarting monitor daemon: monWarning: hostgroup www has no watch
> assigned to it!
> ...
> => why does it say that ? the mon.cf seems good for me ?
>
> regards
>
>
>
> _______________________________________________
> Linux-HA mailing list
> Linux-HA at lists.linux-ha.org
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems
>
>
>