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-13462 Lustre <p>MDS deadlocked</p> <p>Similar to <a href="https://jira.whamcloud.com/browse/LU-13073" title="Multiple MDS deadlocks (in lod_qos_prep_create) after OSS crash" class="issue-link" data-issue-key="LU-13073"><del>LU-13073</del></a></p> <div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"> <pre class="code-java"> 12287155.058187] LNet: Service thread pid 15312 was inactive <span class="code-keyword">for</span> 550.48s. The thread might be hung, or it might only be slow and will resume later. Dumping the stack trace <span class="code-keyword">for</span> debugging purposes: [12287155.109703] LNet: Skipped 2 previous similar messages [12287155.125609] [&lt;ffffffffa5f87398&gt;] call_rwsem_down_read_failed+0x18/0x30 [12287155.130583] [&lt;ffffffffc144acfc&gt;] osd_read_lock+0x5c/0xe0 [osd_ldiskfs] [12287155.130612] [&lt;ffffffffc16f28ea&gt;] lod_read_lock+0x3a/0xd0 [lod] [12287155.130625] [&lt;ffffffffc17779aa&gt;] mdd_read_lock+0x3a/0xd0 [mdd] [12287155.130632] [&lt;ffffffffc177d730&gt;] mdd_xattr_get+0x70/0x5c0 [mdd] [12287155.130648] [&lt;ffffffffc15e6ea6&gt;] mdt_stripe_get+0xd6/0x400 [mdt] [12287155.130657] [&lt;ffffffffc15e7a2d&gt;] mdt_attr_get_complex+0x46d/0x850 [mdt] [12287155.130665] [&lt;ffffffffc15e800c&gt;] mdt_getattr_internal+0x1fc/0xf60 [mdt] [12287155.130673] [&lt;ffffffffc15ebd60&gt;] mdt_getattr_name_lock+0x950/0x1c30 [mdt] [12287155.130681] [&lt;ffffffffc15f3c05&gt;] mdt_intent_getattr+0x2b5/0x480 [mdt] [12287155.130691] [&lt;ffffffffc15f0a18&gt;] mdt_intent_policy+0x2e8/0xd00 [mdt] [12287155.130736] [&lt;ffffffffc0f2dd26&gt;] ldlm_lock_enqueue+0x366/0xa60 [ptlrpc] [12287155.130769] [&lt;ffffffffc0f56587&gt;] ldlm_handle_enqueue0+0xa47/0x15a0 [ptlrpc] [12287155.130815] [&lt;ffffffffc0fde882&gt;] tgt_enqueue+0x62/0x210 [ptlrpc] [12287155.130853] [&lt;ffffffffc0fe31da&gt;] tgt_request_handle+0xaea/0x1580 [ptlrpc] [12287155.130887] [&lt;ffffffffc0f8880b&gt;] ptlrpc_server_handle_request+0x24b/0xab0 [ptlrpc] [12287155.130921] [&lt;ffffffffc0f8c13c&gt;] ptlrpc_main+0xafc/0x1fc0 [ptlrpc] [12287155.130925] [&lt;ffffffffa5cc1da1&gt;] kthread+0xd1/0xe0 [12287155.130929] [&lt;ffffffffa6375c37&gt;] ret_from_fork_nospec_end+0x0/0x39 [12287155.130947] [&lt;ffffffffffffffff&gt;] 0xffffffffffffffff </pre> </div></div> LU-13462

MDS deadlocks in osd_read_lock()

Bug Major Resolved Fixed Yang Sheng Mahmoud Hanafi Sat, 18 Apr 2020 02:54:58 +0000 Mon, 25 Jul 2022 18:38:52 +0000 Mon, 25 Jul 2022 18:38:52 +0000 Lustre 2.12.2 0 6 <p>Mahmoud</p> <p>Could you please supply details of the kernel version that you are running?</p> <p>Yang Sheng</p> <p>Could you please advise</p> <p>Thanks</p> <p>Peter</p> <p>Hi, Mahmoud,</p> <p>Could you please provide more info? What do you mean for similar <a href="https://jira.whamcloud.com/browse/LU-13073" title="Multiple MDS deadlocks (in lod_qos_prep_create) after OSS crash" class="issue-link" data-issue-key="LU-13073"><del>LU-13073</del></a>? </p> <p>Thanks,<br/> Yangsheng</p> <p>The stack trace for hung threads is the same as <a href="https://jira.whamcloud.com/browse/LU-13073" title="Multiple MDS deadlocks (in lod_qos_prep_create) after OSS crash" class="issue-link" data-issue-key="LU-13073"><del>LU-13073</del></a>. But in our case we didn't have a OSS crash</p> <p>Our kernel is: 3.10.0-957.21.3.el7_lustre212.x86_64</p> <p> </p> <p>Then have any possible to provide sysrq-t info? From stack trace i don't think it same as lu-13073.</p> <p>Attached the stack trace.</p> <p>Hi, Mahmoud,</p> <p>The log you attached really duplicated with <a href="https://jira.whamcloud.com/browse/LU-13073" title="Multiple MDS deadlocks (in lod_qos_prep_create) after OSS crash" class="issue-link" data-issue-key="LU-13073"><del>LU-13073</del></a>. But it is different with which you pasted stackstrace. You pasted log shows thread stuck on osd_read_lock. This most was caused by some local filesystem issue. But the <a href="https://jira.whamcloud.com/browse/LU-13073" title="Multiple MDS deadlocks (in lod_qos_prep_create) after OSS crash" class="issue-link" data-issue-key="LU-13073"><del>LU-13073</del></a> is not. It is a long outstanding issue caused by OSS. </p> <p>Thanks,<br/> YangSheng</p> Related LU-13073 Development Rank 1|i00y7b: Rank (Obsolete) 9223372036854775807 Severity