Ciao,I don't think.. in V1 style is working, the behavior change with V2 style.
In attachment you will find a small ha-log file (zip format).
ha-log.rtf.gz
Description: GNU Zip compressed data
I don't know if the output of "ciblint" could help [r...@afsitfs3 crm]# ciblint -LERROR: <nvpair name="short-resource-names"...>: [short-resource-names] is not a legal name for the <crm_config> section ERROR: <nvpair name="transition-idle-timeout"...>: [transition-idle- timeout] is not a legal name for the <crm_config> section WARNING: STONITH disabled <nvpair name="stonith-enabled" value="false">. STONITH is STRONGLY recommended.
WARNING: No STONITH resources configured. STONITH is not available.INFO: See http://linux-ha.org/ciblint/stonith for more information on this topic. INFO: See http://linux-ha.org/ciblint/crm_config#stonith-enabled for more information on this topic.
WARNING: resource afs_6 has failcount 2 on node afsitfs3.roma1.infn.it INFO: Resource Filesystem_4 running on node afsitfs3.roma1.infn.it INFO: Resource Filesystem_2 running on node afsitfs3.roma1.infn.it INFO: Resource drbddisk_1 running on node afsitfs3.roma1.infn.it INFO: Resource drbddisk_3 running on node afsitfs3.roma1.infn.it WARNING: Resource afs_6 not running anywhere.INFO: Resource IPaddr_141_108_26_31 running on node afsitfs3.roma1.infn.it
Thanks cristina On Apr 14, 2009, at 1:00 PM, Dejan Muhamedagic wrote:
Hi, On Tue, Apr 14, 2009 at 10:56:23AM +0200, Cristina Bulfon wrote:Ciao,thanks for the answer ... Dejan has already pointed me out regarding theIP. That IP is the alias IP for the AFS server, and I was using also with IPaddr2 because at the beginning,while I was configuring AFS, I had probem with network communication and Ithought to redirect the trafficon that IP. I've solved that problem and I forgot to delete the entry inharesource file beacuse that configuration work fine with V1... Anyway I correct the haresource file as follows afsitfs3.roma1.infn.it \ drbddisk::afs_fs Filesystem::/dev/drbd1::/vicepa/::xfs \ drbddisk::afs_sw Filesystem::/dev/drbd2::/usr/afs::ext3 \ 141.108.26.31 afs and create the cib.xml I don't have anymore the error but the AFS start/stop continuouslyProbably an afs issue. What do you see in the logs? Dejancristina On Apr 14, 2009, at 10:38 AM, Andrew Beekhof wrote:On Fri, Apr 10, 2009 at 12:25, Cristina Bulfon <[email protected]> wrote:Dejan,I've followed your advice and I've moved to V2, first the software hasbeen updated to version 2.1.4. I just modified the following files - ha.cf, added the line crm yes- cib.xml has been produced using the python script and my haresourcesafsitfs3.roma1.infn.it IPaddr2::141.108.26.31/24/eth0:0 afsitfs3.roma1.infn.it drbddisk::afs_fs Filesystem::/dev/drbd1::/vicepa::xfs afsitfs3.roma1.infn.it drbddisk::afs_sw Filesystem::/dev/drbd2::/usr/afs::ext3 afsitfs3.roma1.infn.it 141.108.26.31 afs With this kind of configuration I've got a lot of error and the AFS resource doesn't workLooks to me like the ip address is the one that doesn't work. Did youactually read the output you pasted below?You might want to double check the nic and netmask attributes, they'reprobably swapped around.- crm_verify -L -x /var/lib/heartbeat/crm/cib.xmlcrm_verify[30489]: 2009/04/10_12:20:01 ERROR: unpack_rsc_op: Hard error:IPaddr2_1_monitor_0 failed with rc=2.crm_verify[30489]: 2009/04/10_12:20:01 ERROR: unpack_rsc_op: PreventingIPaddr2_1 from re-starting on afsitfs4.roma1.infn.itcrm_verify[30489]: 2009/04/10_12:20:01 ERROR: unpack_rsc_op: Hard error:IPaddr2_1_monitor_0 failed with rc=2.crm_verify[30489]: 2009/04/10_12:20:01 ERROR: unpack_rsc_op: PreventingIPaddr2_1 from re-starting on afsitfs3.roma1.infn.it I've attached both cib.xml, ha-log and ha.cf Thanks for helping me cristina On Apr 8, 2009, at 5:50 PM, Cristina Bulfon wrote:Dejan, thanks so much for the explanation :-) c. On Apr 8, 2009, at 5:46 PM, Dejan Muhamedagic wrote:Ciao, On Wed, Apr 08, 2009 at 04:17:45PM +0200, Cristina Bulfon wrote:Ciao Dejan, thanks for the answer.Do you mean that I have to use heartbeat V2 plus CRM and there is away to check the HBA without using hbaping ?Unlike Heartbeat v1, CRM/v2 can monitor resources. I suppose that in your case, a failing HBA would cause drbd or Filesystem monitor action to fail, which would result in either a failover or restart, depending on the configuration. Thanks, DejanJust to be sure if I have understood correctly. I am newby on heartbeat V2 thanks cristina On Mar 31, 2009, at 2:00 PM, Dejan Muhamedagic wrote:Ciao,On Tue, Mar 31, 2009 at 01:48:47PM +0200, Cristina Bulfon wrote:Ciao,in our heartbeat cluster we have simulated the breaking of the HBAbyunplugging the fiber from HBA on the primary node. The resourcedidn'tswitch to the secondary node and on the log file on primary nodereported the following messages:Feb 19 14:33:33 afsitfs3 kernel: qla2xxx 0000:0a:01.0: LOOP DOWNdetected (2 e678 16ed).Feb 19 14:33:38 afsitfs3 kernel: qla2xxx 0000:0a:01.1: LOOP DOWNdetected (2 8633 16fc).Feb 19 14:33:46 afsitfs3 kernel: qla2x00: FAILOVER device 2 from200500a0b832d169 -> 200400a0b832d16a - LUN 10, reason=0x2 Feb 19 14:33:46 afsitfs3 kernel: qla2x00: FROM HBA 0 to HBA 1Feb 19 14:33:52 afsitfs3 kernel: qla2x00: FAILOVER device 2 from200400a0b832d16a -> 200500a0b832d16a - LUN 10, reason=0x2 Feb 19 14:33:52 afsitfs3 kernel: qla2x00: FROM HBA 1 to HBA 1Feb 19 14:33:55 afsitfs3 kernel: qla2x00: FAILOVER device 2 from200500a0b832d16a -> 200400a0b832d169 - LUN 10, reason=0x2 Feb 19 14:33:55 afsitfs3 kernel: qla2x00: FROM HBA 1 to HBA 0Feb 19 14:33:58 afsitfs3 kernel: qla2x00: FAILOVER device 2 from200400a0b832d169 -> 200500a0b832d169 - LUN 10, reason=0x2 Feb 19 14:33:58 afsitfs3 kernel: qla2x00: FROM HBA 0 to HBA 0Feb 19 14:34:01 afsitfs3 kernel: qla2x00: FAILOVER device 2 from200500a0b832d169 -> 200400a0b832d16a - LUN 10, reason=0x2 In some way I expected this kind of messages but I do not understand why the secondary node doesn't take the control of the resources.In the ha.cf there is not nothing related to HBA and the haresourcesfile is afsitfs3.roma1.infn.it IPaddr2::Y.Y.Y.Y/24/eth0:0 afsitfs3.roma1.infn.it drbddisk::r0 Filesystem::/dev/drbd1::/vicepa::xfs afsitfs3.roma1.infn.it drbddisk::r1 Filesystem::/dev/drbd2::/usr/afs::ext3 afsitfs3.roma1.infn.it Y.Y.Y.Y afsThere's no resource monitoring with v1. For that you have to go with v2/Pacemaker (aka CRM).Also tried to use hbaping compiling the hbaapi_src_2.2 but withoutsuccess.. got problem during the compilations and I didn't understand if Ihave to use libHBAAPI.so from hbaapi or from HBA vendor.That could work with ipfail, perhaps. Thanks, DejanOur FC controller isLogic PCI to Fibre Channel Host Adapter for QLA2342: Firmware version 3.03.25 IPX, Driver version 8.02.14.01-foThanks in advance cristina _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems_______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems_______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems_______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems_______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems_______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems_______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems_______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems_______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
_______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
