Hi,
On Fri, Apr 06, 2012 at 11:48:18AM +0200, Julien Cornuwel wrote:
> Hi,
>
> I have a working cluster on which I decided to go further and manage
> the radius process with corosync instead of just the IPaddr.
> And there comes the troubles...
> The radius server is actually started and the monitor_hook (nagios
> check) says it works fine...
>
> Reading the logfile (full log here:
> http://pastebin.com/raw.php?i=MJySbCjx), I found this :
> info: RA output: (Radius:start:stderr)
> /usr/lib/ocf/resource.d//heartbeat/anything: line 60: kill: (9057) -
> No such process
> And the PID of radiusd is indeed different from what is in /var/run/rad.pid.
What probably happens is that radiusd does a fork later, as all
unix daemon programs do. From the RA meta-data:
binfile* (string): Full path name of the binary to be executed
The full name of the binary to be executed. This is expected
to keep running
with the same pid and not just do something and exit.
Looks like "anything" is not suitable for radiusd. You can try
the lsb (i.e. /etc/init.d/...) RA.
Thanks,
Dejan
> I tried putting this path in /etc/raddb/radius.conf and things got
> weird : radius entered a start-stop loop.
>
> Can you see an error in my configuration ? Or have an idea on how to
> convince radiusd to behave ?
>
>
>
> #########################################################################
>
> # crm configure show
> node meaban1.infra
> node meaban2.infra
> primitive ClusterIP ocf:heartbeat:IPaddr2 \
> params ip="172.25.1.53" cidr_netmask="24" \
> op monitor interval="5s"
> primitive Radius ocf:heartbeat:anything \
> params binfile="/usr/sbin/radiusd"
> monitor_hook="/usr/local/nagios/libexec/check_radius.sh &>/dev/null"
> pidfile="/var/run/rad.pid" \
> op monitor interval="5s" timeout="20s" \
> meta target-role="Started"
> group Cluster Radius ClusterIP
> location prefer-node1 Cluster 50: meaban1.infra
> property $id="cib-bootstrap-options" \
> dc-version="1.0.12-unknown" \
> cluster-infrastructure="openais" \
> expected-quorum-votes="2" \
> stonith-enabled="false" \
> no-quorum-policy="ignore"
>
> # crm_mon -1
> ============
> Last updated: Fri Apr 6 11:46:35 2012
> Stack: openais
> Current DC: meaban1.infra - partition with quorum
> Version: 1.0.12-unknown
> 2 Nodes configured, 2 expected votes
> 1 Resources configured.
> ============
>
> Online: [ meaban1.infra meaban2.infra ]
>
>
> Failed actions:
> Radius_monitor_5000 (node=meaban2.infra, call=15, rc=1,
> status=complete): unknown error
> Radius_start_0 (node=meaban2.infra, call=19, rc=1,
> status=complete): unknown error
> Radius_monitor_5000 (node=meaban1.infra, call=11, rc=1,
> status=complete): unknown error
> Radius_start_0 (node=meaban1.infra, call=16, rc=1,
> status=complete): unknown error
>
> _______________________________________________
> 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
