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-1493 Lustre <p>System did assert with following trace :</p> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre>LustreError: 12908:0:(quota_context.c:683:dqacq_completion()) ASSERTION(count &lt; *hardlimit) failed: id(10912) flag(22) type(u) isblk(b) count(134217728) qd_qunit(134217728) hardlimit(131072). LustreError: 12908:0:(quota_context.c:683:dqacq_completion()) LBUG Pid: 12908, comm: mdt_01 Call Trace: [&lt;ffffffffa04df855&gt;] libcfs_debug_dumpstack+0x55/0x80 [libcfs] [&lt;ffffffffa04dfe95&gt;] lbug_with_loc+0x75/0xe0 [libcfs] [&lt;ffffffffa07d6d06&gt;] dqacq_completion+0x1606/0x1610 [lquota] [&lt;ffffffffa07eb84d&gt;] ? quota_search_lqs+0x9d/0x5f0 [lquota] [&lt;ffffffffa04e09ae&gt;] ? cfs_free+0xe/0x10 [libcfs] [&lt;ffffffff81085bf0&gt;] ? getnstimeofday+0x60/0xf0 [&lt;ffffffffa07d4738&gt;] schedule_dqacq+0xa08/0x19d0 [lquota] [&lt;ffffffffa04e09ae&gt;] ? cfs_free+0xe/0x10 [libcfs] [&lt;ffffffffa07d15e8&gt;] ? check_cur_qunit+0x448/0xb70 [lquota] [&lt;ffffffffa07dc8ca&gt;] ? quota_is_set+0x6a/0x2f0 [lquota] [&lt;ffffffffa07d79c9&gt;] qctxt_adjust_qunit+0x109/0x350 [lquota] [&lt;ffffffff81085cea&gt;] ? do_gettimeofday+0x1a/0x50 [&lt;ffffffffa07e299d&gt;] mds_quota_adjust+0x2ad/0x3b0 [lquota] [&lt;ffffffffa09f2001&gt;] ? mdd_lov_create_finish+0x61/0xd0 [mdd] [&lt;ffffffffa0a07982&gt;] mdd_create+0x6c2/0x1db0 [mdd] [&lt;ffffffffa04ee649&gt;] ? cfs_hash_bd_add_locked+0x29/0x90 [libcfs] [&lt;ffffffffa05a96de&gt;] ? lu_object_find_at+0x3fe/0x770 [obdclass] [&lt;ffffffffa09e5f08&gt;] ? mdd_version_get+0x68/0xa0 [mdd] [&lt;ffffffffa0a902bc&gt;] cml_create+0xbc/0x280 [cmm] [&lt;ffffffffa0a4d746&gt;] ? mdt_version_save+0x96/0x170 [mdt] [&lt;ffffffffa0a65747&gt;] mdt_reint_open+0x1f67/0x2d90 [mdt] [&lt;ffffffff81003ace&gt;] ? common_interrupt+0xe/0x13 [&lt;ffffffffa0a0e586&gt;] ? md_ucred+0x26/0x60 [mdd] [&lt;ffffffffa0a305f5&gt;] ? mdt_ucred+0x15/0x20 [mdt] [&lt;ffffffffa0a4786f&gt;] ? mdt_root_squash+0x2f/0x450 [mdt] [&lt;ffffffffa0a4cabf&gt;] mdt_reint_rec+0x3f/0x100 [mdt] [&lt;ffffffffa069cd74&gt;] ? lustre_msg_get_flags+0x34/0xa0 [ptlrpc] [&lt;ffffffffa0a44f64&gt;] mdt_reint_internal+0x6d4/0x9f0 [mdt] [&lt;ffffffffa0a32cde&gt;] ? mdt_intent_fixup_resent+0x4e/0x270 [mdt] [&lt;ffffffffa0a455e5&gt;] mdt_intent_reint+0x245/0x600 [mdt] [&lt;ffffffffa04ef615&gt;] ? cfs_hash_bd_lookup_intent+0xe5/0x130 [libcfs] [&lt;ffffffffa069e170&gt;] ? lustre_swab_ldlm_intent+0x0/0x20 [ptlrpc] [&lt;ffffffffa0a3d630&gt;] mdt_intent_policy+0x3c0/0x6b0 [mdt] [&lt;ffffffff810f18c6&gt;] ? __perf_event_task_sched_out+0x36/0x50 [&lt;ffffffffa0587441&gt;] ? class_handle_hash+0xa1/0x280 [obdclass] [&lt;ffffffffa0655afa&gt;] ldlm_lock_enqueue+0x2da/0xa50 [ptlrpc] [&lt;ffffffffa0674495&gt;] ? ldlm_export_lock_get+0x15/0x20 [ptlrpc] [&lt;ffffffffa04ee682&gt;] ? cfs_hash_bd_add_locked+0x62/0x90 [libcfs] [&lt;ffffffffa067c577&gt;] ldlm_handle_enqueue0+0x447/0x1090 [ptlrpc] [&lt;ffffffffa0a313a1&gt;] ? mdt_unpack_req_pack_rep+0x51/0x5d0 [mdt] [&lt;ffffffffa0a3d0ca&gt;] mdt_enqueue+0x4a/0x110 [mdt] [&lt;ffffffffa0a37865&gt;] mdt_handle_common+0x8d5/0x1810 [mdt] [&lt;ffffffffa069a4f4&gt;] ? lustre_msg_get_opc+0x94/0x100 [ptlrpc] [&lt;ffffffffa0a38875&gt;] mdt_regular_handle+0x15/0x20 [mdt] [&lt;ffffffffa06ab239&gt;] ptlrpc_main+0xc79/0x19d0 [ptlrpc] [&lt;ffffffff810017bc&gt;] ? __switch_to+0x1ac/0x320 [&lt;ffffffffa06aa5c0&gt;] ? ptlrpc_main+0x0/0x19d0 [ptlrpc] [&lt;ffffffff810041aa&gt;] child_rip+0xa/0x20 [&lt;ffffffffa06aa5c0&gt;] ? ptlrpc_main+0x0/0x19d0 [ptlrpc] [&lt;ffffffffa06aa5c0&gt;] ? ptlrpc_main+0x0/0x19d0 [ptlrpc] [&lt;ffffffff810041a0&gt;] ? child_rip+0x0/0x20 </pre> </div></div> <p>After the MDS went back to production it did failed again with the same trace, same assertion (with same parameters), but req-&gt;rq_peer didn't reference the same client.</p> <p>Any idea ?</p> <p>Thanks</p> LU-1493

assertion in dqacq_completion() (count < *hardlimit) failed

Bug Minor Resolved Fixed Niu Yawei Alexandre Louvet Thu, 7 Jun 2012 05:30:40 +0000 Wed, 25 Jul 2012 12:05:17 +0000 Wed, 25 Jul 2012 12:05:17 +0000 Lustre 2.1.1 Lustre 2.3.0 Lustre 2.1.3 0 3 <p>Niu</p> <p>Could you please comment on this one?</p> <p>Thanks</p> <p>Peter</p> <p>Hi, Alexandre, is this easy to be reproduced? Do you know what kind of operations caused this?</p> <p>Hi Niu,</p> <p>Unfortunately I doesn't have a reproducer nor an idea of what was running on the remote node at time of assert.<br/> Since then, the system was rebooted and we did run fsck + quotacheck and we haven't see the issue again, but I don't know if the code that did create this issue has run again or not.</p> <p>Right now I don't have more data about this issue.</p> <p>Alex.</p> <p>Looks there is a race in the quota code, which could cause unexpected quota release, and trigger the assert in dqacq_completion(). </p> <p>The patch for master: <a href="http://review.whamcloud.com/3074" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/3074</a></p> <p>Landed for 2.1.3 and 2.3</p> Development Rank 1|hzv6h3: Rank (Obsolete) 4583 Severity