[Pacemaker] crm resource restart fails to restart the service

Pavlos Parissis pavlos.parissis at gmail.com
Wed Nov 17 12:42:29 UTC 2010


On 17 November 2010 13:35, Vadym Chepkov <vchepkov at gmail.com> wrote:
>
> On Nov 17, 2010, at 7:26 AM, Dan Frincu wrote:
>
>> Hi,
>>
>> root at cluster1:/# pgrep mysql
>> 961
>> 1127
>> root at cluster1:/# crm resource restart mysqld
>> root at cluster1:/# pgrep -fl mysql
>> 961
>> 1127
>>
>> The restart command doesn't actually restart the process, I have tried this with another custom built OCF compliant RA and have the same issue.
>>
>> # rpm -qa '(pacemaker|corosync|resource-agents)'
>> pacemaker-1.0.9.1-1.el5
>> resource-agents-1.0.3-2.el5
>> corosync-1.2.7-1.1.el5
>>
>> # crm configure show mysqld
>> primitive mysqld ocf:heartbeat:mysql \
>>       params binary="/usr/bin/mysqld_safe" config="/etc/my.cnf" enable_creation="0" datadir="/mysql/database" user="root" test_user="monitor" test_passwd="monitor" test_table="cluster.monitor" \
>>       op monitor interval="10s" timeout="5s" \
>>       op start interval="0s" \
>>       op stop interval="0s" \
>>       meta target-role="Started"
>>
>> Ideas?
>
>
> RA doesn't support restart action? Most luckily you get OCF_ERR_UNIMPLEMENTED in the log
>
> Vadym
>
that is correct
[root at node-01 heartbeat]# pwd
/usr/lib/ocf/resource.d/heartbeat
[root at node-01 heartbeat]# grep usage mysql
# An example usage in /etc/ha.d/haresources:
# See usage() function below for more details...
usage() {
        usage: $0 (start|stop|validate-all|meta-data|monitor)
  usage|help)   usage
 *)             usage


but in my case it supports restart



More information about the Pacemaker mailing list