Greetings! I have 3 OSTs that will not activate. Here are the errors... LustreError: 5203:0:(llog_obd.c:194:llog_setup()) obd feline-OST0013-osc ctxt 2 lop_setup=ffffffffa0298c20 failed -2 LustreError: 5203:0:(osc_request.c:3740:osc_llog_init()) osc ''feline-OST0013-osc'' tgt ''feline-MDT0000'' cnt 1 catid 000001021d6739a8 rc=-2 LustreError: 5203:0:(lov_log.c:243:lov_llog_init()) error osc_llog_init idx 19 osc ''feline-OST0013 osc'' tgt ''feline-MDT0000'' (rc=-2) LustreError: 14278:0:(mds_lov.c:918:__mds_lov_synchronize()) feline-OST0013_UUID failed at update_mds: -2 LustreError: 14287:0:(mds_lov.c:960:__mds_lov_synchronize()) feline-OST0013_UUID sync failed -2, deactivating Running lctl dl shows the OST inactive on the MDS only (OSS shows active). Running lctl --device 24 conf_param feline-OST0013.osc.active=1 completes successfully and shows "Lustre Permanently reactivating feline-OST0013" in dmesg. About 30 seconds later it goes through the above errors again and deactivates. Any ideas on how to fix this? Dan -- Sent from my Palm Pre -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.lustre.org/pipermail/lustre-discuss/attachments/20100415/86426e07/attachment.html
On 2010-04-15, at 14:34, Dan wrote:> I have 3 OSTs that will not activate. Here are the errors... > > LustreError: 5203:0:(llog_obd.c:194:llog_setup()) obd feline-OST0013- > osc ctxt 2 lop_setup=ffffffffa0298c20 failed -2I believe this is a known bug, but I can''t search bugzilla right now. The way to fix it, IIRC, is to mount the MDT as ldiskfs and delete the CATALOGS file.> LustreError: 5203:0:(osc_request.c:3740:osc_llog_init()) osc ''feline- > OST0013-osc'' tgt ''feline-MDT0000'' cnt 1 catid 000001021d6739a8 rc=-2 > LustreError: 5203:0:(lov_log.c:243:lov_llog_init()) error > osc_llog_init idx 19 osc ''feline-OST0013 osc'' tgt ''feline- > MDT0000'' (rc=-2) > LustreError: 14278:0:(mds_lov.c:918:__mds_lov_synchronize()) feline- > OST0013_UUID failed at update_mds: -2 > LustreError: 14287:0:(mds_lov.c:960:__mds_lov_synchronize()) feline- > OST0013_UUID sync failed -2, deactivating > > Running lctl dl shows the OST inactive on the MDS only (OSS shows > active). Running lctl --device 24 conf_param feline- > OST0013.osc.active=1 completes successfully and shows "Lustre > Permanently reactivating feline-OST0013" in dmesg. About 30 seconds > later it goes through the above errors again and deactivates. > > Any ideas on how to fix this? > > > Dan > > > > > > > -- Sent from my Palm Pre > > _______________________________________________ > Lustre-discuss mailing list > Lustre-discuss at lists.lustre.org > http://lists.lustre.org/mailman/listinfo/lustre-discussCheers, Andreas -- Andreas Dilger Principal Engineer, Lustre Group Oracle Corporation Canada Inc.
Thanks a lot Andreas!! Fixed in a second. After a round of lfsck things are looking great. I''m always surprised at how durable and capable the Lustre system is. Thanks for all you effort, Lustre Team!! Dan -- Sent from my Palm Pre On Apr 15, 2010 4:16 PM, Andreas Dilger <andreas.dilger at oracle.com> wrote: On 2010-04-15, at 14:34, Dan wrote: > I have 3 OSTs that will not activate. Here are the errors... > > LustreError: 5203:0:(llog_obd.c:194:llog_setup()) obd feline-OST0013- > osc ctxt 2 lop_setup=ffffffffa0298c20 failed -2 I believe this is a known bug, but I can''t search bugzilla right now. The way to fix it, IIRC, is to mount the MDT as ldiskfs and delete the CATALOGS file. > LustreError: 5203:0:(osc_request.c:3740:osc_llog_init()) osc ''feline- > OST0013-osc'' tgt ''feline-MDT0000'' cnt 1 catid 000001021d6739a8 rc=-2 > LustreError: 5203:0:(lov_log.c:243:lov_llog_init()) error > osc_llog_init idx 19 osc ''feline-OST0013 osc'' tgt ''feline- > MDT0000'' (rc=-2) > LustreError: 14278:0:(mds_lov.c:918:__mds_lov_synchronize()) feline- > OST0013_UUID failed at update_mds: -2 > LustreError: 14287:0:(mds_lov.c:960:__mds_lov_synchronize()) feline- > OST0013_UUID sync failed -2, deactivating > > Running lctl dl shows the OST inactive on the MDS only (OSS shows > active). Running lctl --device 24 conf_param feline- > OST0013.osc.active=1 completes successfully and shows "Lustre > Permanently reactivating feline-OST0013" in dmesg. About 30 seconds > later it goes through the above errors again and deactivates. > > Any ideas on how to fix this? > > > Dan > > > > > > > -- Sent from my Palm Pre > > _______________________________________________ > Lustre-discuss mailing list > Lustre-discuss at lists.lustre.org > http://lists.lustre.org/mailman/listinfo/lustre-discuss Cheers, Andreas -- Andreas Dilger Principal Engineer, Lustre Group Oracle Corporation Canada Inc. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.lustre.org/pipermail/lustre-discuss/attachments/20100415/76a517f1/attachment.html