On Thu, Jan 13, 2011 at 06:29:57PM +0100, RaSca wrote: > Il giorno Gio 13 Gen 2011 17:32:36 CET, Dejan Muhamedagic ha scritto: > [...] > >Bad filesystem? That's what it says anyway. > > I'm loosing my mind in this problem. The patch seemed to work the > first time, but on reboot the resource still hangs.
iscsiadm fails with the following message: > Jan 13 19:15:46 debian-squeeze-nodo1 lrmd: [1274]: info: RA output: > (www_db-iscsi:start:stderr) iscsiadm: > Jan 13 19:15:46 debian-squeeze-nodo1 lrmd: [1274]: info: RA output: > (www_db-iscsi:start:stderr) no records found! Try to figure out what does that mean, my iscsi is a bit rusty. > And it doesn't > comes up until I give a command like this one: > > iscsiadm -m node -p 10.0.0.100:3260 -T > "iqn.2010-12.local.rascanet:db.rascanet.iscsi" -l Hmm, really can't say what's up. That command is what is run by the iscsi RA too in the start action. > Then, the resource comes up and switch correctly. Attached there is > a log extract of a startup. > > Let me just ask this thing, on my topology I've got this boot order: > > open-iscsi daemon > THEN > iscsitarget daemon > THEN > heartbeat > > Does it seems right? I think so, but cannot say for sure. At any rate, your distribution should've set the boot order properly. Thanks, Dejan > Resource Agents of iscsitarget and open-iscsi > may populate the target and the iscsi, and on failover and shutdown > clean their exports. > I know that those might be 2 clusters instead of one, but the > customer wants iscsi, wants failover and has got just two machines. > > Thanks for your patience, > > -- > Raoul Scarazzini > Mia Mamma Usa Linux: Niente รจ impossibile da capire, se lo spieghi bene! > ra...@miamammausalinux.org _______________________________________________ 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
