CRS-0184: Cannot communicate with the CRS daemon,
Hunted!
After a scheduled downtime when I was starting our RAC database I found Cluster ready services(CRS) was not starting up in one node, then I found that all the mount points (shared disk drive) were not visible in nodes. I rebooted the DB servers.
Two Luns were not were not available.
after database server reboot of rac nodes, mount points were became visible,
but still one of the nodes is having problem in starting up CRS i.e.
Error:
CRS-0184: Cannot communicate with the CRS daemon
Logs:
$ less /u01/app/crs/log/nodename/crsd/crsd.log
output:
: (10076b610) no listener at (ADDRESS=(PROTOCOL=ipc)
(KEY=OCSSD_LL_nodename_))
: [ CSSCLNT][1]clsssInitNative: failed to connect to
(ADDRESS=(PROTOCOL=ipc)(KEY=OCSSD_LL_nodename_)), rc 9
: [ CRSRTI][1] CSS is not ready. Received status 3 from CSS.
Waiting for good status ..
Reason:
sometimes when CRS server reboot it try to create sockets under /tmp/.oracle or /var/tmp/.oracle and there are already previous socket files...which are protecting to create new sockets.
Solution:
loing with root user and remove all files under /tmp/.oracle or /var/tmp/.oracle restart CRS on faulty node or even reboot that machine!, start the crs and it will come up.
