https://jira.whamcloud.com This file is an XML representation of an issue en-us 9.4.14 940014 05-12-2023 https://jira.whamcloud.com/browse/LU-2568 Lustre <p>here is a log from failed mdt start:</p> <p>Feb 12 22:04:14 tstmds0a01 kernel: LustreError: 5302:0:(llog_lvfs.c:616:llog_lvfs_create()) error looking up logfile 0x7a4801b:0x2790e5b9: rc -116<br/> Feb 12 22:04:14 tstmds0a01 kernel: LustreError: 5302:0:(llog_cat.c:174:llog_cat_id2handle()) error opening log id 0x7a4801b:2790e5b9: rc -116<br/> Feb 12 22:04:14 tstmds0a01 kernel: LustreError: 5302:0:(llog_obd.c:318:cat_cancel_cb()) Cannot find handle for log 0x7a4801b<br/> Feb 12 22:04:14 tstmds0a01 kernel: LustreError: 5248:0:(llog_obd.c:391:llog_obd_origin_setup()) llog_process() with cat_cancel_cb failed: -116<br/> Feb 12 22:04:14 tstmds0a01 kernel: LustreError: 5248:0:(llog_obd.c:218:llog_setup_named()) obd mdd_obd-content-MDT0000 ctxt 12 lop_setup=ffffffff88625e70 failed -116<br/> Feb 12 22:04:14 tstmds0a01 kernel: LustreError: 5248:0:(mds_log.c:182:mds_changelog_llog_init()) changelog llog setup failed -116<br/> Feb 12 22:04:14 tstmds0a01 kernel: LustreError: 5248:0:(mdd_device.c:196:mdd_changelog_llog_init()) no changelog context<br/> Feb 12 22:04:14 tstmds0a01 kernel: LustreError: 5248:0:(mdd_device.c:271:mdd_changelog_init()) Changelog setup during init failed -22<br/> Feb 12 22:04:15 tstmds0a01 kernel: Lustre: content-MDT0000: used disk, loading<br/> Feb 12 22:04:15 tstmds0a01 kernel: LustreError: 5248:0:(mdt_handler.c:1889:mdt_llog_ctxt_clone()) Can't get mdd ctxt -2<br/> Feb 12 22:04:15 tstmds0a01 kernel: LustreError: 5248:0:(obd_config.c:522:class_setup()) setup content-MDT0000 failed (-2)</p> LU-2568

MDT unable to start with corrupted llog files.

Bug Major Resolved Cannot Reproduce WC Triage Alexander Zarochentsev Thu, 3 Jan 2013 17:16:30 +0000 Thu, 9 Jan 2020 06:20:26 +0000 Thu, 9 Jan 2020 06:20:26 +0000 0 2 <p>Xyratex has a fix for this issue I will upload it later.</p> <p>Zam, this doesn't look as master bug, is it some older Lustre version than 2.3?</p> <p>Yes, it is older bug. but looks like it is still in master. </p> <p>The issue was with missing llog files and their inode numbers were re-used for other objects.<br/> llog_lvfs_create returns ESTATE in that case &#8211; it is the error seen at the top level.</p> <p>The key fix was:</p> <div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"> <pre class="code-java">diff --git a/lustre/obdclass/llog_lvfs.c b/lustre/obdclass/llog_lvfs.c index 0987020..60bad4c 100644 --- a/lustre/obdclass/llog_lvfs.c +++ b/lustre/obdclass/llog_lvfs.c @@ -615,6 +615,10 @@ <span class="code-keyword">static</span> <span class="code-object">int</span> llog_lvfs_create(struct llog_ctxt *ctxt, struct llog_handle **res, rc = PTR_ERR(dchild); CERROR(<span class="code-quote">"error looking up logfile "</span>LPX64<span class="code-quote">":0x%x: rc %d\n"</span>, logid-&gt;lgl_oid, logid-&gt;lgl_ogen, rc); + <span class="code-keyword">if</span> (rc == -ESTALE) + /* handle reused inode same way as + non-existing one */ + GOTO(out, rc = -ENOENT); GOTO(out, rc); } </pre> </div></div> <p>I still think it is actual for the master branch. but I haven't tried to re-create it on master.</p> <p>the llog_lvfs_create is not used anymore in master, llogs are OSD-based now, do you have any reproducer for this? I suppose it shouldn't be problem now if the reason was inode re-use because now llog object is fid-based, but we need to check that </p> <p>Close old ticket.</p> Development Rank 1|hzvek7: Rank (Obsolete) 6000 Severity