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-16673 Lustre <p>This issue was created by maloo for Andreas Dilger &lt;adilger@whamcloud.com&gt;</p> <p>This issue relates to the following test suite run with aarch64 servers:<br/> <a href="https://testing.whamcloud.com/test_sets/38335ace-3096-4f6a-a18a-f252f4d7bf6c" class="external-link" target="_blank" rel="nofollow noopener">https://testing.whamcloud.com/test_sets/38335ace-3096-4f6a-a18a-f252f4d7bf6c</a></p> <p>test_125 failed with the following error:</p> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre>setfacl /mnt/lustre/d125.sanity failed </pre> </div></div> <p>Test session details:<br/> clients: 2.15.54.79 - 4.18.0-348.23.1.el8_lustre331.aarch64<br/> servers: 2.15.54.79 - 4.18.0-348.23.1.el8_lustre331.aarch64</p> <p>Also failed the following subtests (listing all here for searching):<br/> test_150f<br/> test_150g<br/> test_151<br/> test_154a<br/> test_154b<br/> test_156</p> <p>VVVVVVV DO NOT REMOVE LINES BELOW, Added by Maloo for auto-association VVVVVVV<br/> sanity test_125 - setfacl /mnt/lustre/d125.sanity failed</p> LU-16673

sanity test_125: failures with aarch64 servers

Bug Minor Resolved Fixed Xinliang Liu Maloo Mon, 27 Mar 2023 18:52:07 +0000 Mon, 21 Aug 2023 19:54:01 +0000 Sat, 19 Aug 2023 12:12:51 +0000 Lustre 2.16.0 0 5 <p>It looks like there are a handful of consistent sanity subtest failures when running with aarch64 servers.</p> <p>You are likely already aware of this, since the test results are coming from a linaro test cluster, but I filed the ticket to allow tracking the test failures, and for submitting patches to fix them.</p> <p>Yeah, we are aware of this. We tracking the sanity.sh test suite failed tests on this ticket: <a href="https://linaro.atlassian.net/browse/STOR-123" class="external-link" target="_blank" rel="nofollow noopener">https://linaro.atlassian.net/browse/STOR-123</a></p> <p>Recently, I am looking at this. </p> <p>Looks like the user USER0 for setfacl operation is nonexistent when specifying a valid user test 125 pass.</p> <p>USER0=${USER0:-"sanityusr"}</p> <p>setfacl -R -m u:$USER0:rwx $DIR/$tdir</p> <div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"> <pre class="code-java"> $ sudo USER0=openeuler RUNAS_ID=<span class="code-quote">"1000"</span> ~/lustre-release/lustre/tests/auster  -rv sanity --only 125 == sanity test 125: don't <span class="code-keyword">return</span> EPROTO when a dir has a non-<span class="code-keyword">default</span> striping and ACLs ========================================================== 09:03:07 (1680166987) setfacl -R -m u:openeuler:rwx /mnt/lustre/d125.sanity drwxrwxr-x+ 2 root root 4096 Mar 30 09:03 /mnt/lustre/d125.sanity PASS 125 (1s) </pre> </div></div> <p>Should add a check for user USER0 in the test.</p> <p>"xinliang &lt;xinliang.liu@linaro.org&gt;" uploaded a new patch: <a href="https://review.whamcloud.com/c/fs/lustre-release/+/50501" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/c/fs/lustre-release/+/50501</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-16673" title="sanity test_125: failures with aarch64 servers" class="issue-link" data-issue-key="LU-16673"><del>LU-16673</del></a> tests: add checking for user USER0<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: 1<br/> Commit: 7559cd4ae840828720b5ae5f1ccf927ca96280e2</p> <p>And for those tests 150f,150g,151,156, failing might relate to 64K page size, because these tests pass on aarch64 4k page size cluster.</p> <p><a href="https://jira.whamcloud.com/secure/ViewProfile.jspa?name=xinliang" class="user-hover" rel="xinliang">xinliang</a>, out of curiosity, are you formatting the OSTs with 64KB blocksize, or still 4KB blocksize (assuming ldiskfs)? There are definitely some interesting benefits from 64KB blocksize:</p> <ul class="alternate" type="square"> <li>larger directories with fewer htree levels</li> <li>256x fewer block groups due to 16x bits per bitmap x 16x larger blocks</li> <li>more efficient allocation for large files</li> </ul> <p>This would mean less space efficiency for files &lt; 64KB in size and for the fraction of a block at the end of each file, but for files &lt;= 64KB it would be possible to use DoM to store them on the MDT.</p> <p>Our aarch64 CI clusters run on CentOS 8 which use 64KB PAGE_SIZE and thus use 64KB blocksize for ldiskfs.</p> <p>It sounds a good idea to store files &lt; 64KB on MDT with DoM for 64KB PAGE_SIZE cluster.</p> <p> </p> <p>We also do test locally on openEuler which use 4K PAGE_SIZE.</p> <p> </p> <p>BTW, RHEL 9 uses 4K PAGE_SIZE.</p> <p>"Oleg Drokin &lt;green@whamcloud.com&gt;" merged in patch <a href="https://review.whamcloud.com/c/fs/lustre-release/+/50501/" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/c/fs/lustre-release/+/50501/</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-16673" title="sanity test_125: failures with aarch64 servers" class="issue-link" data-issue-key="LU-16673"><del>LU-16673</del></a> tests: add checking for user USER0<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: <br/> Commit: b7b55534b9b5d4d58025082ae5403938853b168c</p> <p>Landed for 2.16</p> Related LU-16541 Development Rank 1|i03hbz: Rank (Obsolete) 9223372036854775807 Severity