On Wednesday we had an OSS hang due to a disk controller lockup on one of our lustre installs. Since then we've seen this error on the MDS:
lund-adm1 kernel: LustreError: 6923:0:(mds_open.c:1295:mds_open()) ASSERTION(!mds_inode_is_orphan(dchild->d_inode)) failed: dchild b668006:144d828c (ffff8104651d9d20) inode ffff81046e94a138/191266822/340624012 Message from syslogd@ at Fri Dec 17 02:05:54 2010 ... lund-adm1 kernel: LustreError: 6923:0:(mds_open.c:1295:mds_open()) LBUG We did an e2fsck on all OSS and the MDS which cleaned up some orphan inode on the MDS and all of the OSS's were clean. I also searched through BZ but all I found on the error was a bug from 1.4. Is a full lfsck needed here? Any suggestions would be appreciated. -- Gary Molenkamp SHARCNET Systems Administrator University of Western Ontario Compute/Calcul Canada http://www.computecanada.org [email protected] http://www.sharcnet.ca (519) 661-2111 x88429 (519) 661-4000 _______________________________________________ Lustre-discuss mailing list [email protected] http://lists.lustre.org/mailman/listinfo/lustre-discuss
