Hi Phil, On Tue, Apr 19, 2011 at 3:36 PM, Phil Hunt <phil.h...@orionhealth.com> wrote: > Hi > I have iscsid running, no iscsi.
Good. You don't want the system to auto-connect the iSCSI disks on boot, pacemaker will do that for you. > > > > Here is the crm status: > ============ > Last updated: Tue Apr 19 12:39:03 2011 > Stack: openais > Current DC: CentClus2 - partition with quorum > Version: 1.0.10-da7075976b5ff0bee71074385f8fd02f296ec8a3 > 2 Nodes configured, 2 expected votes > 1 Resources configured. > ============ > > Online: [ CentClus1 CentClus2 ] > > Resource Group: CL_group > ClusterIP (ocf::heartbeat:IPaddr2): Started CentClus2 > FS_disk (ocf::heartbeat:Filesystem): Stopped > ISCSI_disk (ocf::heartbeat:iscsi): Stopped > VG_disk (ocf::heartbeat:LVM): Stopped > PM_ping (ocf::pacemaker:ping): Stopped > IP_ping (ocf::heartbeat:IPaddr2): Stopped > > The resources are listed in top-down start order. So you're starting ClusterIP, but then try to start the filesystem when you still haven't connected to the iSCSI disk or started the volume group. > > Here is the crm config: > node CentClus1 .... > group CL_group ClusterIP FS_disk ISCSI_disk VG_disk PM_ping IP_ping .... > order ISCSI_startup inf: ISCSI_disk VG_disk FS_disk You have conflicting orders, there. A resource group defines an order, so the other order statement seems unnecessary. If you remove that order constraint, and change your group line so that things start in the correct order, does it come up? group CL_group ClusterIP ISCSI_disk VG_disk FS_disk PM_ping I left off IP_ping, because it's exactly the same as ClusterIP. Was it meant to be something else? Regards, Mark _______________________________________________ Pacemaker mailing list: Pacemaker@oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs: http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker