Re: [Linux-HA] remove resource WITHOUT moving the other resources
On Sun, Aug 14, 2011 at 11:04 PM, Julian D. Seifert ala...@julian-seifert.de wrote: Hi, Thank you for your response, I have some follow-up questions. Now what I am looking for is a way to completely delete/remove openvzve_itv without affecting the other resources. is-managed-default=false and/or resource-stickiness=INFINITY When trying to use the first method, setting is-managed-default to false, how would I go on about that? I think I'd have to crm configure property is-managed-default=false Right (How would I list all set properties?) crm configure show After that I should be able to delete the stuff from the configuration without it causing pacemaker to act upon it, right? Right I noticed that only deleting the group entry of openvzve_itv causes the cluster to want to move all resources to the other node. (Even after setting said property) That doesn't sound right. Can you send logs if you still have them? I'm curious as to why the cluster would decide to do this in the first place, how he decides that this would be a good idea. regards, Julian D. Seifert -- ___ Linux-HA mailing list Linux-HA@lists.linux-ha.org http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems ___ Linux-HA mailing list Linux-HA@lists.linux-ha.org http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
Re: [Linux-HA] remove resource WITHOUT moving the other resources
Hi, Thank you for your response, I have some follow-up questions. Now what I am looking for is a way to completely delete/remove openvzve_itv without affecting the other resources. is-managed-default=false and/or resource-stickiness=INFINITY When trying to use the first method, setting is-managed-default to false, how would I go on about that? I think I'd have to crm configure property is-managed-default=false (How would I list all set properties?) After that I should be able to delete the stuff from the configuration without it causing pacemaker to act upon it, right? I noticed that only deleting the group entry of openvzve_itv causes the cluster to want to move all resources to the other node. (Even after setting said property) I'm curious as to why the cluster would decide to do this in the first place, how he decides that this would be a good idea. regards, Julian D. Seifert -- signature.asc Description: OpenPGP digital signature ___ Linux-HA mailing list Linux-HA@lists.linux-ha.org http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
Re: [Linux-HA] remove resource WITHOUT moving the other resources
On Thu, Aug 11, 2011 at 5:28 AM, ala...@julian-seifert.de wrote: Hi List, I have a little problem with my 2 node pacemaker cluster. (Active/Passive Setup). vpsnode01-rz (current active node) and vpsnode01-nk (passive) It's a bunch of OpenVZ containers grouped together and colocated to where the DRBD-Master resource is running. I have a problem with a newly created Container openvzve_itv. 1. It's recognized as running on the passive node. (Which it is NOT and that wouldn't be possible anyway as the VEs are located on the DRBD storage backends) Then the agent is broken. 2. I am NOT able to remove the openvzve_itv resource WITHOUT ptest suggesting that EVERY other resource will switch nodes. To clarify what I just describe I pasted the crm_mon output and the current configuration: http://pastebin.ubuntu.com/653023/ Now what I am looking for is a way to completely delete/remove openvzve_itv without affecting the other resources. is-managed-default=false and/or resource-stickiness=INFINITY I tried deleting everything in the configuration (crm configuration edit) related to openvzve_itv (namely the resource itself and its entry in the openvz group) but as I mentioned earlier when running ptest vvv nograph after that and before commiting of course it suggests that ALL resources will get moved. PTEST: crm(live)configure# ptest vvv nograph ptest[7623]: 2011/08/10_21:27:50 notice: unpack_config: On loss of CCM Quorum: Ignore ptest[7623]: 2011/08/10_21:27:50 notice: unpack_rsc_op: Hard error - openvzve_itv_start_0 failed with rc=5: Preventing openvzve_itv from re-starting on vpsnode01-nk ptest[7623]: 2011/08/10_21:27:50 WARN: unpack_rsc_op: Processing failed op openvzve_itv_start_0 on vpsnode01-nk: not installed (5) ptest[7623]: 2011/08/10_21:27:50 WARN: unpack_rsc_op: Processing failed op openvzve_itv_stop_0 on vpsnode01-nk: unknown error (1) ptest[7623]: 2011/08/10_21:27:50 notice: clone_print: Clone Set: connected ptest[7623]: 2011/08/10_21:27:50 notice: short_print: Started: [ vpsnode01-nk vpsnode01-rz ] ptest[7623]: 2011/08/10_21:27:50 notice: group_print: Resource Group: openvz ptest[7623]: 2011/08/10_21:27:50 notice: native_print: fs_openvz (ocf::heartbeat:Filesystem): Started vpsnode01-rz ptest[7623]: 2011/08/10_21:27:50 notice: native_print: ip_openvz (ocf::heartbeat:IPaddr2): Started vpsnode01-rz ptest[7623]: 2011/08/10_21:27:50 notice: native_print: openvzve_plesk953 (ocf::heartbeat:ManageVE): Started vpsnode01-rz ptest[7623]: 2011/08/10_21:27:50 notice: native_print: openvzve_stream3 (ocf::heartbeat:ManageVE): Started vpsnode01-rz ptest[7623]: 2011/08/10_21:27:50 notice: native_print: openvzve_mail (ocf::heartbeat:ManageVE): Started vpsnode01-rz ptest[7623]: 2011/08/10_21:27:50 notice: native_print: openvzve_dns1 (ocf::heartbeat:ManageVE): Started vpsnode01-rz ptest[7623]: 2011/08/10_21:27:50 notice: native_print: openvzve_itv (ocf::heartbeat:ManageVE): Started vpsnode01-nk (unmanaged) FAILED ptest[7623]: 2011/08/10_21:27:50 notice: clone_print: Master/Slave Set: ms_drbd_openvz ptest[7623]: 2011/08/10_21:27:50 notice: short_print: Masters: [ vpsnode01-rz ] ptest[7623]: 2011/08/10_21:27:50 notice: short_print: Slaves: [ vpsnode01-nk ] ptest[7623]: 2011/08/10_21:27:50 notice: common_apply_stickiness: openvzve_plesk953 can fail 99 more times on vpsnode01-rz before being forced off ptest[7623]: 2011/08/10_21:27:50 WARN: common_apply_stickiness: Forcing openvzve_itv away from vpsnode01-nk after 100 failures (max=100) ptest[7623]: 2011/08/10_21:27:50 notice: RecurringOp: Start recurring monitor (10s) for openvzve_plesk953 on vpsnode01-nk ptest[7623]: 2011/08/10_21:27:50 notice: RecurringOp: Start recurring monitor (10s) for openvzve_stream3 on vpsnode01-nk ptest[7623]: 2011/08/10_21:27:50 notice: RecurringOp: Start recurring monitor (10s) for openvzve_mail on vpsnode01-nk ptest[7623]: 2011/08/10_21:27:50 notice: RecurringOp: Start recurring monitor (10s) for openvzve_dns1 on vpsnode01-nk ptest[7623]: 2011/08/10_21:27:50 notice: RecurringOp: Start recurring monitor (20s) for drbd_openvz:1 on vpsnode01-rz ptest[7623]: 2011/08/10_21:27:50 ERROR: create_notification_boundaries: Creating boundaries for ms_drbd_openvz ptest[7623]: 2011/08/10_21:27:50 ERROR: create_notification_boundaries: Creating boundaries for ms_drbd_openvz ptest[7623]: 2011/08/10_21:27:50 notice: RecurringOp: Start recurring monitor (20s) for drbd_openvz:1 on vpsnode01-rz ptest[7623]: 2011/08/10_21:27:50 ERROR: create_notification_boundaries: Creating boundaries for ms_drbd_openvz ptest[7623]: 2011/08/10_21:27:50 ERROR: create_notification_boundaries: Creating boundaries for ms_drbd_openvz ptest[7623]: 2011/08/10_21:27:50 notice: LogActions: Leave resource ping:0 (Started vpsnode01-nk) ptest[7623]: 2011/08/10_21:27:50 notice: LogActions: Leave
[Linux-HA] remove resource WITHOUT moving the other resources
Hi List, I have a little problem with my 2 node pacemaker cluster. (Active/Passive Setup). vpsnode01-rz (current active node) and vpsnode01-nk (passive) It's a bunch of OpenVZ containers grouped together and colocated to where the DRBD-Master resource is running. I have a problem with a newly created Container openvzve_itv. 1. It's recognized as running on the passive node. (Which it is NOT and that wouldn't be possible anyway as the VEs are located on the DRBD storage backends) 2. I am NOT able to remove the openvzve_itv resource WITHOUT ptest suggesting that EVERY other resource will switch nodes. To clarify what I just describe I pasted the crm_mon output and the current configuration: http://pastebin.ubuntu.com/653023/ Now what I am looking for is a way to completely delete/remove openvzve_itv without affecting the other resources. I tried deleting everything in the configuration (crm configuration edit) related to openvzve_itv (namely the resource itself and its entry in the openvz group) but as I mentioned earlier when running ptest vvv nograph after that and before commiting of course it suggests that ALL resources will get moved. PTEST: crm(live)configure# ptest vvv nograph ptest[7623]: 2011/08/10_21:27:50 notice: unpack_config: On loss of CCM Quorum: Ignore ptest[7623]: 2011/08/10_21:27:50 notice: unpack_rsc_op: Hard error - openvzve_itv_start_0 failed with rc=5: Preventing openvzve_itv from re-starting on vpsnode01-nk ptest[7623]: 2011/08/10_21:27:50 WARN: unpack_rsc_op: Processing failed op openvzve_itv_start_0 on vpsnode01-nk: not installed (5) ptest[7623]: 2011/08/10_21:27:50 WARN: unpack_rsc_op: Processing failed op openvzve_itv_stop_0 on vpsnode01-nk: unknown error (1) ptest[7623]: 2011/08/10_21:27:50 notice: clone_print: Clone Set: connected ptest[7623]: 2011/08/10_21:27:50 notice: short_print: Started: [ vpsnode01-nk vpsnode01-rz ] ptest[7623]: 2011/08/10_21:27:50 notice: group_print: Resource Group: openvz ptest[7623]: 2011/08/10_21:27:50 notice: native_print: fs_openvz (ocf::heartbeat:Filesystem):Started vpsnode01-rz ptest[7623]: 2011/08/10_21:27:50 notice: native_print: ip_openvz (ocf::heartbeat:IPaddr2): Started vpsnode01-rz ptest[7623]: 2011/08/10_21:27:50 notice: native_print: openvzve_plesk953 (ocf::heartbeat:ManageVE): Started vpsnode01-rz ptest[7623]: 2011/08/10_21:27:50 notice: native_print: openvzve_stream3(ocf::heartbeat:ManageVE): Started vpsnode01-rz ptest[7623]: 2011/08/10_21:27:50 notice: native_print: openvzve_mail (ocf::heartbeat:ManageVE): Started vpsnode01-rz ptest[7623]: 2011/08/10_21:27:50 notice: native_print: openvzve_dns1 (ocf::heartbeat:ManageVE): Started vpsnode01-rz ptest[7623]: 2011/08/10_21:27:50 notice: native_print: openvzve_itv (ocf::heartbeat:ManageVE): Started vpsnode01-nk (unmanaged) FAILED ptest[7623]: 2011/08/10_21:27:50 notice: clone_print: Master/Slave Set: ms_drbd_openvz ptest[7623]: 2011/08/10_21:27:50 notice: short_print: Masters: [ vpsnode01-rz ] ptest[7623]: 2011/08/10_21:27:50 notice: short_print: Slaves: [ vpsnode01-nk ] ptest[7623]: 2011/08/10_21:27:50 notice: common_apply_stickiness: openvzve_plesk953 can fail 99 more times on vpsnode01-rz before being forced off ptest[7623]: 2011/08/10_21:27:50 WARN: common_apply_stickiness: Forcing openvzve_itv away from vpsnode01-nk after 100 failures (max=100) ptest[7623]: 2011/08/10_21:27:50 notice: RecurringOp: Start recurring monitor (10s) for openvzve_plesk953 on vpsnode01-nk ptest[7623]: 2011/08/10_21:27:50 notice: RecurringOp: Start recurring monitor (10s) for openvzve_stream3 on vpsnode01-nk ptest[7623]: 2011/08/10_21:27:50 notice: RecurringOp: Start recurring monitor (10s) for openvzve_mail on vpsnode01-nk ptest[7623]: 2011/08/10_21:27:50 notice: RecurringOp: Start recurring monitor (10s) for openvzve_dns1 on vpsnode01-nk ptest[7623]: 2011/08/10_21:27:50 notice: RecurringOp: Start recurring monitor (20s) for drbd_openvz:1 on vpsnode01-rz ptest[7623]: 2011/08/10_21:27:50 ERROR: create_notification_boundaries: Creating boundaries for ms_drbd_openvz ptest[7623]: 2011/08/10_21:27:50 ERROR: create_notification_boundaries: Creating boundaries for ms_drbd_openvz ptest[7623]: 2011/08/10_21:27:50 notice: RecurringOp: Start recurring monitor (20s) for drbd_openvz:1 on vpsnode01-rz ptest[7623]: 2011/08/10_21:27:50 ERROR: create_notification_boundaries: Creating boundaries for ms_drbd_openvz ptest[7623]: 2011/08/10_21:27:50 ERROR: create_notification_boundaries: Creating boundaries for ms_drbd_openvz ptest[7623]: 2011/08/10_21:27:50 notice: LogActions: Leave resource ping:0 (Started vpsnode01-nk) ptest[7623]: 2011/08/10_21:27:50 notice: LogActions: Leave resource ping:1 (Started vpsnode01-rz) ptest[7623]: 2011/08/10_21:27:50 notice: LogActions: Move resource fs_openvz(Started vpsnode01-rz - vpsnode01-nk) ptest[7623]: 2011/08/10_21:27:50 notice: LogActions: Move
