>> On 06/10/2010 07:04 PM, Caspar Smit wrote: >>> Don't get me wrong, the failover IS WORKING. Only the connection >>> re-instatement to MS iSCSI Initiator isn't when using the RAs. >> >> Can you try setting an explicit SCSI ID and SN on your logical units >> please, and see if that improves the situation? As in: >> >> primitive res_lu_lun1 ocf:heartbeat:iSCSILogicalUnit \ >> params target_iqn="iqn.2001-04.com.example:storage.example.iscsivg01" >> lun="1" \ >> path="/dev/iscsivg01/lun1" \ >> scsi_id="iscsivg01.lun1" scsi_sn="4711" > > I already tested this, but did a re-check to no avail. > Still only node01->node02 failover works. > > I thought maybe the DC role of pacemaker had something to do with it > because when node01 is DC the services (failover-ip, iscsitarget, > iscsilogicalunit, portblock) start/stop almost instantly on node1 (when > monitoring with crm_mon) start/stop on node02 takes a second each. > Switching DC didn't change anything. > > Still only node01->node02 failover works. > > Now to test open-iscsi.
Ok, I can now DEFINATLY say this is an MS iSCSI Initiator issue. I tested with open-iscsi and now the connection re-instatement works both ways, tested like 10 failovers and no disconnect! :-) Used same pacemaker config like before. Still weird why MS initiator only works one way. Guess I have to ditch MS in my setup. Kind regards, Caspar > > Kind regards, > Caspar Smit > >> >> Thanks. >> >> Cheers, >> Florian >> >> _______________________________________________________ >> Linux-HA-Dev: [email protected] >> http://lists.linux-ha.org/mailman/listinfo/linux-ha-dev >> Home Page: http://linux-ha.org/ >> > > _______________________________________________________ > Linux-HA-Dev: [email protected] > http://lists.linux-ha.org/mailman/listinfo/linux-ha-dev > Home Page: http://linux-ha.org/ > _______________________________________________________ Linux-HA-Dev: [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha-dev Home Page: http://linux-ha.org/
