Description of problem: Service is running on node link-08. I am typing on link-12, 'clusvcadm -s FOO'. Command reports success. The service was not stopped. This works properly if the command is issued from the node on which the service is currently running. So maybe the problem here is my incorrect expectations. If so, it should be clearly stated in the man pages that you can't expect this to work from other nodes. It should also not report success when it failed. It would be a confusing expectation in that the user can enable and disable services from anywhere in the cluster, but not stop them? [root@link-12 cluster]# clusvcadm -s FOO Member link-12 stopping FOO...success [root@link-12 cluster]# clustat Member Status: Quorate, Group Member Member Name State ID ------ ---- ----- -- link-11 Online 0x000000000000000b link-10 Online 0x000000000000000a link-08 Online 0x0000000000000008 link-12 Online 0x000000000000000c Resource Group Owner (Last) State -------- ----- ----- ------ ----- FOO link-08 started Version-Release number of selected component (if applicable): [root@link-12 cluster]# clusvcadm -v 1.9.20 How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info:
Reproduced. Easy fix.
Fixes in CVS
Verified in 1.9.23. [root@link-10 ~]# clustat Member Status: Quorate, Group Member Member Name State ID ------ ---- ----- -- link-12 Online 0x000000000000000c link-11 Online 0x000000000000000b link-08 Online 0x0000000000000008 link-10 Online 0x000000000000000a Resource Group Owner (Last) State -------- ----- ----- ------ ----- FOOGOOPOO link-12 started [root@link-10 ~]# clusvcadm -s FOOGOOPOO Member link-10 stopping FOOGOOPOO...success [root@link-10 ~]# clustat Member Status: Quorate, Group Member Member Name State ID ------ ---- ----- -- link-12 Online 0x000000000000000c link-11 Online 0x000000000000000b link-08 Online 0x0000000000000008 link-10 Online 0x000000000000000a Resource Group Owner (Last) State -------- ----- ----- ------ ----- FOOGOOPOO (link-12 ) stopped