Description of problem: Also, "rgmanager" attr of <node> is not accurate (always '0') Version-Release number of selected component (if applicable): rgmanager-2.0.4-0 How reproducible: Always Steps to Reproduce: 1. execute `clustat -x` 2. look for "groupmember" attr of <quorum> 3. look for "rgmanager" attr of <node>
Devel ACK for RHEL 5.
This request was evaluated by Red Hat Product Management for inclusion in a Red Hat Enterprise Linux release. Product Management has requested further review of this request by Red Hat Engineering. This request is not yet committed for inclusion in release.
Created attachment 137948 [details] implements behavior [root@war utils]# ./clustat Member Status: Quorate Member Name ID Status ------ ---- ---- ------ war.lab.boston.redhat.com 1 Online, Local, rgmanager pestilence.lab.boston.redhat.com 2 Online Service Name Owner (Last) State ------- ---- ----- ------ ----- service:test war.lab.boston.redhat.com started [root@war utils]# ./clustat Member Status: Quorate Member Name ID Status ------ ---- ---- ------ war.lab.boston.redhat.com 1 Online, Local, rgmanager pestilence.lab.boston.redhat.com 2 Online, rgmanager Service Name Owner (Last) State ------- ---- ----- ------ ----- service:test war.lab.boston.redhat.com started [root@war utils]# ./clustat -x <?xml version="1.0"?> <clustat version="4.1.1"> <quorum quorate="1"/> <nodes> <node name="war.lab.boston.redhat.com" state="1" local="1" estranged="0" rgmanager="1" nodeid="0x00000001"/> <node name="pestilence.lab.boston.redhat.com" state="1" local="0" estranged="0" rgmanager="1" nodeid="0x00000002"/> </nodes> <groups> <group name="service:test" state="112" state_str="started" owner="war.lab.boston.redhat.com" last_owner="pestilence.lab.boston.redhat.com" restarts="0" last_transition="1160165965" last_transition_str="Fri Oct 6 16:19:25 2006"/> </groups> </clustat>
Created attachment 137949 [details] corrected fix - removes memory leak
Created attachment 137953 [details] corrected corrected patch. I'm going home now.
Tested third patch with: window 1: while :; do clusvcadm -d test; clusvcadm -e test; done window 2: while :; do clustat ; done No leaks found.
QE ack for RHEL5.
Created attachment 138389 [details] Ancillary patch which adds back the groupmember attr (not just rgmanager attr)
[root@war utils]# ./clustat -x <?xml version="1.0"?> <clustat version="4.1.1"> <quorum quorate="1" groupmember="0"/> <nodes> <node name="war.lab.boston.redhat.com" state="1" local="1" estranged="0" rgmanager="0" nodeid="0x00000001"/> <node name="pestilence.lab.boston.redhat.com" state="0" local="0" estranged="0" rgmanager="0" nodeid="0x00000002"/> </nodes> </clustat> [root@war utils]# service rgmanager start Starting Cluster Service Manager: [ OK ] [root@war utils]# clustat Member Status: Quorate Member Name ID Status ------ ---- ---- ------ war.lab.boston.redhat.com 1 Online, Local, rgmanager pestilence.lab.boston.redhat.com 2 Offline Service Name Owner (Last) State ------- ---- ----- ------ ----- service:test war.lab.boston.redhat.com started [root@war utils]# clustat -x <?xml version="1.0"?> <clustat version="4.1.1"> <quorum quorate="1" groupmember="1"/> <nodes> <node name="war.lab.boston.redhat.com" state="1" local="1" estranged="0" rgmanager="1" nodeid="0x00000001"/> <node name="pestilence.lab.boston.redhat.com" state="0" local="0" estranged="0" rgmanager="0" nodeid="0x00000002"/> </nodes> <groups> <group name="service:test" state="112" state_str="started" owner="war.lab.boston.redhat.com" last_owner="none" restarts="0" last_transition="1160691076" last_transition_str="Thu Oct 12 18:11:16 2006"/> </groups> </clustat>
Thanks Lon
Closed/Upstream isn't a valid state for RHEL5 bugs at this point in the process. It gives no indication if the fix is actually available in the RHEL5 tree or not, or if it's been verified to work as expected in a RHEL5 installation. Moving back to Assigned. Please provide information about which version of the package the fix is actually present in.
It was fixed and committed to upstream/public CVS prior to inclusion in any Red Hat package build. For Red Hat Packages, it was fixed in rgmanager-2.0.11 (and subsequent builds); built on 12-Oct-2006 for FC6 (2.0.11-0) and 13-Oct-2006 for RHEL5 (2.0.11-1.el5).
[root@cobalt 4.91]# find . -name "rgmanager*" -print ./i386/os/Cluster/rgmanager-2.0.13-1.i386.rpm ./i386/debug/rgmanager-debuginfo-2.0.13-1.i386.rpm ./x86_64/os/Cluster/rgmanager-2.0.13-1.x86_64.rpm ./x86_64/debug/rgmanager-debuginfo-2.0.13-1.x86_64.rpm ./ia64/os/Cluster/rgmanager-2.0.13-1.ia64.rpm ./ia64/debug/rgmanager-debuginfo-2.0.13-1.ia64.rpm ./source/SRPMS/rgmanager-2.0.13-1.src.rpm [root@cobalt 4.91]# pwd /mnt/redhat/rel-eng/RHEL5-Server-20061020.1/4.91
Moving all RHCS ver 5 bugs to RHEL 5 so we can remove RHCS v5 which never existed.