[Pacemaker] Problems setting up clone resource syslog-ng - Ineed help

[Senftleben, Stefan, ITSC]

Great - that worked, I executed following commands, to use the RA of syslog-ng:

export OCF_ROOT=/usr/lib/ocf
export OCF_RESKESY_configfile=/etc/syslog-ng/syslog-ng.conf
export OCF_RESKEY_syslog_ng_binary=/usr/sbin/syslog-ng


Additionally I added the "syslog_ng_binary" parameter within "crm configure edit":

primitive syslog-ng ocf:heartbeat:syslog-ng \
        params configfile="/etc/syslog-ng/syslog-ng.conf" syslog_ng_binary="/usr/sbin/syslog-ng" \
        op monitor interval="60s" timeout="60s" depth="0"
clone syslog-ng-clone syslog-ng \
        meta globally-unique="false" target-role="Started"


Afterwards I had to fire that command:
crm_resource -P

Thank you to all of you!

Stefan
___________________________________________________________________________________

-----Ursprüngliche Nachricht-----
Von: Michael Schwartzkopff [mailto:misch at clusterbau.com] 
Gesendet: Mittwoch, 17. November 2010 13:23
An: The Pacemaker cluster resource manager
Betreff: Re: [Pacemaker] Problems setting up clone resource syslog-ng - Ineed help

On Wednesday 17 November 2010 13:09:01 Senftleben, Stefan, ITSC wrote:
> Hi,
> 
> okay, these infos I found in syslog. But no lrmd-log entries canbe found.
> 
> 
> root at lxds05:~# cat /var/log/syslog|grep pengi|grep syslog Nov 17 
> 12:47:22 lxds05 pengine: [1089]: ERROR: unpack_rsc_op: Hard error -
> syslog-ng:0_monitor_0 failed with rc=6: Preventing syslog-ng-clone 
> from re-starting anywhere in the cluster Nov 17 12:47:22 lxds05 pengine:
> [1089]: ERROR: unpack_rsc_op: Hard error - syslog-ng:0_monitor_0 
> failed with rc=6: Preventing syslog-ng-clone from re-starting anywhere 
> in the cluster Nov 17 12:47:22 lxds05 pengine: [1089]: notice: clone_print:
> Clone Set: syslog-ng-clone Nov 17 12:47:22 lxds05 pengine: [1089]: notice:
> short_print:      Stopped: [ syslog-ng:0 syslog-ng:1 ] Nov 17 12:47:22
> lxds05 pengine: [1089]: info: native_color: Resource syslog-ng:0 
> cannot run anywhere Nov 17 12:47:22 lxds05 pengine: [1089]: info: native_color:
> Resource syslog-ng:1 cannot run anywhere Nov 17 12:47:22 lxds05 pengine:
> [1089]: notice: LogActions: Leave resource syslog-ng:0  (Stopped) Nov 
> 17
> 12:47:22 lxds05 pengine: [1089]: notice: LogActions: Leave resource
> syslog-ng:1  (Stopped) Nov 17 12:48:00 lxds05 pengine: [1089]: ERROR:
> unpack_rsc_op: Hard error - syslog-ng:0_monitor_0 failed with rc=6:
> Preventing syslog-ng-clone from re-starting anywhere in the cluster
> 
> 
> Question:
> 	I executed following commands in /usr/lib/ocf/resource.d/heartbeat/ :
> OCF_ROOT=/usr/lib/ocf
> OCF_FUNCTIONS_DIR=${OCF_ROOT}/resource.d/heartbeat
> 
> 	Executing the RA manually:
> root at lxds05:/usr/lib/ocf/resource.d/heartbeat# ./syslog-ng
> ./syslog-ng: Zeile 27: /resource.d/heartbeat/.ocf-shellfuncs: No such 
> file or directory ./syslog-ng: Zeile 216: ocf_log: Kommando nicht gefunden.
> 
> What command is needed to run the RA manually?
> 
> Thanks
> Stefan

Hi,

The lines above the ERROR messages in the syslog would be interesting. Please check on both nodes for output of the resource agent.

export OCF_ROOT=/usr/lib/ocf
/usr/lib/ocf/resource.d/heartbeat/syslog-ng meta-data

this will tell you that the configfile parameter is mandatory.

so have to set something like
export OCF_RESKESY_configfile=/etc/syslog-ng.conf

Greetings,


--
Dr. Michael Schwartzkopff
Guardinistr. 63
81375 München

Tel: (0163) 172 50 98