Red Hat Bugzilla – Bug 193603
rgmanager init script should fail it it can't start
Last modified: 2009-04-16 16:20:33 EDT
Description of problem:
Without any cluster services (ccsd, cman, lock_gulm) running, an attempt to
start the rgmanger init script passed as if it worked. The log in the console
messages it a bit more descriptive though.
[root@taft-04 init.d]# service rgmanager start
Starting Cluster Service Manager: [ OK ]
[root@taft-04 init.d]# echo $?
[root@taft-04 init.d]# clustat
Could not connect to cluster service
[root@taft-04 init.d]# ps -ef | grep clurgmgrd
root 4694 4214 0 12:02 pts/0 00:00:00 grep clurgmgrd
May 30 12:01:42 taft-04 clurgmgrd: <notice> Resource Group Manager Starting
May 30 12:01:42 taft-04 clurgmgrd: <info> Loading Service Data
May 30 12:01:42 taft-04 rgmanager: clurgmgrd startup succeeded
May 30 12:01:42 taft-04 clurgmgrd: <crit> #5: Couldn't connect to ccsd!
May 30 12:01:42 taft-04 clurgmgrd: <crit> #8: Couldn't initialize services
Version-Release number of selected component (if applicable):
Created attachment 135688 [details]
Adds 30 second wait for status report from rgmanager on startup
This will allow the admin to see if rgmanager started correctly or not from
init script output, and has 30 second timeout by default. This will not
override $RGMGR_OPTS if the user has set them.
The proper status is now reported, but still missing the proper return code in
[root@link-07 tmp]# service rgmanager start
Starting Cluster Service Manager: [FAILED]
[root@link-07 tmp]# echo $?
Fixes in CVS.
An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on the solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.