GDS

From pressy's brainbackup
Jump to: navigation, search

GDS

Depending on your own GDS resource script, you might have a start script that exits after startup, so the cluster has no more process or forks to monitor and this will lead to a "failed to stay up".

In your start script, tell the cluster not to monitor parent and childs:

#########
# test to stay up
#########
sleep 60 & 
/usr/cluster/bin/pmfadm -s ${RESOURCEGROUP_NAME},${RESOURCE_NAME},0.svc
/usr/cluster/bin/pmfadm -s cluster-rg,cluster-rs,0.svc