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-6817 Lustre <p>Concurrent write and truncates to a file cause clients to try to set HS_DIRTY on the file even if the file was never archived.</p> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre># while true; do echo XXX &gt;&gt; f0; done &amp; # while true; do truncate --size=13 f0; done &amp; t login: [109280.708508] LustreError: 12321:0:(llite_lib.c:1647:ll_setattr_raw()) [0x200000401:0x1:0x0]HSM set dirty failed: rc2 = -22 [109280.712174] LustreError: 12321:0:(llite_lib.c:1647:ll_setattr_raw()) Skipped 18 previous similar messages [109314.024737] LustreError: 17287:0:(llite_lib.c:1647:ll_setattr_raw()) [0x200000401:0x1:0x0]HSM set dirty failed: rc2 = -22 [109314.026936] LustreError: 17287:0:(llite_lib.c:1647:ll_setattr_raw()) Skipped 39 previous similar messages </pre> </div></div> LU-6817

(llite_lib.c:1647:ll_setattr_raw()) [0x200000401:0x342:0x0]HSM set dirty failed: rc2 = -22

Bug Minor Resolved Fixed Zhenyu Xu John Hammond hsm llite Wed, 8 Jul 2015 20:28:19 +0000 Tue, 18 Aug 2015 13:58:26 +0000 Tue, 18 Aug 2015 13:58:26 +0000 Lustre 2.8.0 Lustre 2.8.0 0 5 <p>I'm suddenly seeing a lot of errors like this logged all the time. doesn't require overlapping write/trunc to make it happen as far as I can tell.</p> <p>This error was introduced by recent commit 6acf93339ad3297f2e5c659f2269c05df6198f74 in master for <a href="https://jira.whamcloud.com/browse/LU-5823" title="Replace some obsolete obd operations with CLIO ioctl interface" class="issue-link" data-issue-key="LU-5823"><del>LU-5823</del></a> llite: Remove access of stripe in ll_setattr_raw (I think).</p> <p>Bobijam</p> <p>Could you please look into this issue?</p> <p>Thanks</p> <p>Peter</p> <p>This message is harmless. The root cause of this problem is that LLIF_DATA_MODIFIED is used to identify if a file has been restored. Based on the chance of this happening, I would like to fix this problem by:</p> <p>1. clear LLIF_DATA_MODIFIED before sending MDS_DATA_MODIFIED;</p> <div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"> <pre class="code-java"> <span class="code-keyword">if</span> (!hsm_import &amp;&amp; attr-&gt;ia_valid &amp; ATTR_SIZE) { /* If we are changing file size, file content is * modified, flag it. */ attr-&gt;ia_valid |= MDS_OPEN_OWNEROVERRIDE; op_data-&gt;op_bias |= MDS_DATA_MODIFIED; &gt;&gt;&gt;&gt;&gt;&gt; New code ll_file_clear_flag(lli, LLIF_DATA_MODIFIED); </pre> </div></div> <p>This way it can reduce the chance of this happening.</p> <p>2. Since it's still a legal race for write and truncate happening on the same time, we should change CERROR of this message to CDEBUG(D_INFO, ...).</p> <p>Bobi Jam (bobijam@hotmail.com) uploaded a new patch: <a href="http://review.whamcloud.com/15541" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/15541</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-6817" title="(llite_lib.c:1647:ll_setattr_raw()) [0x200000401:0x342:0x0]HSM set dirty failed: rc2 = -22" class="issue-link" data-issue-key="LU-6817"><del>LU-6817</del></a> llite: lower message level for ll_setattr_raw()<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: 1<br/> Commit: eefae9e4afc12796a39f4cb9b1f1dd33e264ae47</p> <p>Oleg Drokin (oleg.drokin@intel.com) merged in patch <a href="http://review.whamcloud.com/15541/" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/15541/</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-6817" title="(llite_lib.c:1647:ll_setattr_raw()) [0x200000401:0x342:0x0]HSM set dirty failed: rc2 = -22" class="issue-link" data-issue-key="LU-6817"><del>LU-6817</del></a> llite: lower message level for ll_setattr_raw()<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: <br/> Commit: 5291d454d49cd6048b815382fc5647712c277c1d</p> <p>Landed for 2.8</p> Development Rank 1|hzxhqv: Rank (Obsolete) 9223372036854775807 Severity