Thomas Bennett
2007-Jan-15 20:30 UTC
[Lustre-discuss] Lustre problem with 2.6.12 kernel, MDS does not start
Hi, While running the llmount.sh example provided with the lustre, it fails when trying to start up the mds server. The first error I noticed in /var/log/messages/ is Jan 15 13:23:01 goldenclient kernel: LustreError: 18067:0:(lvfs.h:130:ll_lookup_one_len()) bad inode returned 25002/2176617687 Below are some details of the installation, stdout for llmount and the complete output from /var/log/messages Any help/suggestions welcome. Lustre version: 1.4.8 Kernel version: 2.6.12 (patched with quilt) Linux: CentOS 4.3 llmount stdout: # ./llmount.sh loading module: libcfs srcdir None devdir libcfs loading module: lnet srcdir None devdir lnet loading module: ksocklnd srcdir None devdir klnds/socklnd loading module: lvfs srcdir None devdir lvfs loading module: obdclass srcdir None devdir obdclass loading module: ptlrpc srcdir None devdir ptlrpc loading module: ost srcdir None devdir ost loading module: ldiskfs srcdir None devdir ldiskfs loading module: fsfilt_ldiskfs srcdir None devdir lvfs loading module: obdfilter srcdir None devdir obdfilter loading module: mdc srcdir None devdir mdc loading module: osc srcdir None devdir osc loading module: lov srcdir None devdir lov loading module: mds srcdir None devdir mds loading module: llite srcdir None devdir llite NETWORK: NET_goldenclient_tcp NET_goldenclient_tcp_UUID tcp goldenclient OSD: OST_goldenclient OST_goldenclient_UUID obdfilter /tmp/ost1-goldenclient 400000 ldiskfs no 0 256 OST mount options: errors=remount-ro MDSDEV: mds1 mds1_UUID /tmp/mds1-goldenclient ldiskfs no recording clients for filesystem: FS_fsname_UUID Recording log mds1 on mds1 LOV: lov_mds1 19815_lov_mds1_3a6905033b mds1_UUID 1 1048576 0 0 [u''OST_goldenclient_UUID''] mds1 OSC: OSC_goldenclient_OST_goldenclient_mds1 19815_lov_mds1_3a6905033b OST_goldenclient_UUID End recording log mds1 on mds1 Recording log goldenclient on mds1 Recording log client on mds1 MDSDEV: mds1 mds1_UUID /tmp/mds1-goldenclient ldiskfs 400000 no MDS mount options: errors=remount-ro,user_xattr,acl, MDS failed to start. Check the syslog for details. (May need to run lconf --write-conf) Active Internet connections (w/o servers) Proto Recv-Q Send-Q Local Address Foreign Address State PID/Program name tcp 0 0 192.168.2.26:22 192.168.2.27:32774 ESTABLISHED 11263/2 tcp 0 0 192.168.2.26:22 192.168.2.27:32797 ESTABLISHED 14167/0 tcp 0 0 127.0.0.1:7634 127.0.0.1:55341 TIME_WAIT - tcp 0 0 127.0.0.1:7634 127.0.0.1:55340 TIME_WAIT - tcp 0 0 127.0.0.1:7634 127.0.0.1:55343 TIME_WAIT - tcp 0 0 127.0.0.1:7634 127.0.0.1:55342 TIME_WAIT - tcp 0 0 127.0.0.1:7634 127.0.0.1:55339 TIME_WAIT - tcp 0 0 127.0.0.1:7634 127.0.0.1:55338 TIME_WAIT - /var/log/messages Jan 15 13:22:58 goldenclient kernel: Lustre: 17434:0:(module.c:382:init_libcfs_module()) maximum lustre stack 8192 Jan 15 13:22:58 goldenclient kernel: Lustre: OBD class driver Build Version: 1.4.8-19700101020000-PRISTINE-.usr.src.kernels.linux-2.6.12-lustre.-2.6.12-lustre, info@clusterfs.com Jan 15 13:22:58 goldenclient kernel: Lustre: Added LNI 192.168.2.26@tcp[8/256] Jan 15 13:22:58 goldenclient kernel: Lustre: Accept secure, port 988 Jan 15 13:22:58 goldenclient kernel: Lustre: Filtering OBD driver; info@clusterfs.com Jan 15 13:22:58 goldenclient kernel: Lustre: Lustre Lite Client File System; info@clusterfs.com Jan 15 13:22:59 goldenclient kernel: kjournald starting. Commit interval 5 seconds Jan 15 13:22:59 goldenclient kernel: LDISKFS FS on loop0, internal journal Jan 15 13:22:59 goldenclient kernel: LDISKFS-fs: mounted filesystem with ordered data mode. Jan 15 13:22:59 goldenclient kernel: Lustre: 17714:0:(filter.c:575:filter_init_server_data()) OST_goldenclient: initializing new last_rcvd Jan 15 13:22:59 goldenclient kernel: Lustre: OST OST_goldenclient now serving /dev/loop0 (788ac747-0908-4c38-ba37-28d73e6b2ab8) with recovery enabled Jan 15 13:23:00 goldenclient kernel: kjournald starting. Commit interval 5 seconds Jan 15 13:23:00 goldenclient kernel: LDISKFS FS on loop1, internal journal Jan 15 13:23:00 goldenclient kernel: LDISKFS-fs: mounted filesystem with ordered data mode. Jan 15 13:23:00 goldenclient kernel: Lustre: 17907:0:(mds_fs.c:239:mds_init_server_data()) mds1: initializing new last_rcvd Jan 15 13:23:00 goldenclient kernel: Lustre: MDT mds1 now serving /dev/loop1 (7a1f7412-617f-4e16-90ff-f749924889b0) with recovery enabled Jan 15 13:23:01 goldenclient kernel: Lustre: MDT mds1 has stopped. Jan 15 13:23:01 goldenclient kernel: Lustre: 18067:0:(handler.c:1828:fsoptions_to_mds_flags()) ignoring unsupported acl mount option Jan 15 13:23:01 goldenclient kernel: kjournald starting. Commit interval 5 seconds Jan 15 13:23:01 goldenclient kernel: LDISKFS FS on loop1, internal journal Jan 15 13:23:01 goldenclient kernel: LDISKFS-fs: mounted filesystem with ordered data mode. Jan 15 13:23:01 goldenclient kernel: LustreError: 18067:0:(lvfs.h:130:ll_lookup_one_len()) bad inode returned 25002/2176617687 Jan 15 13:23:01 goldenclient kernel: LustreError: 18067:0:(mds_fs.c:472:mds_fs_setup()) cannot create LOGS directory: rc = -2 Jan 15 13:23:01 goldenclient kernel: LustreError: 18067:0:(handler.c:1929:mds_setup()) mds1: MDS filesystem method init failed: rc = -2 Jan 15 13:23:01 goldenclient kernel: LustreError: 18069:0:(obd_config.c:288:class_cleanup()) Device 3 not setup -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mail.clusterfs.com/pipermail/lustre-discuss/attachments/20070115/aa7dccd7/attachment.html