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-472 Lustre <div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"> <pre class="code-java"># Skip these tests # bug number: 17466 18857 ALWAYS_EXCEPT=<span class="code-quote">"61d 33a 33b $REPLAY_SINGLE_EXCEPT"</span> </pre> </div></div> <p>Bug 17466 was fixed in Dec, 2008. The test should be reviewed and removed from ALWAYS_EXCEPT.</p> LU-472

Revive replay-single 61d

Bug Minor Resolved Fixed Li Wei Li Wei Wed, 29 Jun 2011 05:51:28 +0000 Thu, 9 Jan 2020 07:40:03 +0000 Tue, 5 Nov 2019 20:13:03 +0000 Lustre 2.1.0 Lustre 2.11.0 0 2 <p>Since replay-single 61d's debut in Jul, 2008, the test has been led astray from <a href="https://bugzilla.lustre.org/show_bug.cgi?id=16002#c25" class="external-link" target="_blank" rel="nofollow noopener">its original mission</a>. The historical changes described below should be reverted or fixed.</p> <p><a href="http://git.whamcloud.com/?p=fs/lustre-release.git;a=commitdiff;h=02d8c2fa5c63271d01818d24c86a80d7d63a6e00#patch6" class="external-link" target="_blank" rel="nofollow noopener">llog_setup_named()</a> and <a href="http://git.whamcloud.com/?p=fs/lustre-release.git;a=commitdiff;h=5efe4c568f6f6b25cc56b1de507491b3f6ff7d45" class="external-link" target="_blank" rel="nofollow noopener">llog_setup()</a>. Originally, the fault injection simulated what might happen when the "CATALOGS" internal file was corrupted. After the changes were applied, the fault started to be injected at the first call of llog_setup_named()/llog_setup() even when lop_setup was not implemented. In other words, a fault could be injected at locations that would never fail! Another issue in these changes is that replacing llog_ctxt_destroy() with llog_ctxt_put() causes lop_cleanup to be invoked upon lop_setup failures. In general, interfaces like lop_setup should clean things up before returning an error; lop_cleanup should not be called upon lop_setup failures.</p> <p><a href="http://git.whamcloud.com/?p=fs/lustre-release.git;a=commitdiff;h=2fe8b2600cd7aea21149fbb9d0829129a274a472" class="external-link" target="_blank" rel="nofollow noopener">test_61d()</a>. This change looks like a result of the fictional fault injection described above. The test should stop and start the MDT rather than the MGS.</p> <p>Johann and Andreas, any comments?</p> <p>Hi Li Wei,<br/> Yes, i agree with you, this test should start/stop the MDS and not the MGS. We used to only test combo MGS &amp; MDS, that's probably why we have never encountered such a problem before.</p> <p>This patch fixes the test:<br/> <a href="http://review.whamcloud.com/1112" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/1112</a><br/> This patch tries to fix problems revealed by the newly revived test:<br/> <a href="http://review.whamcloud.com/1113" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/1113</a></p> <p>This was enabled in patch <a href="https://review.whamcloud.com/27404" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/27404</a> "<a href="https://jira.whamcloud.com/browse/LU-9590" title="replay-single test cases 61d 89 73b remove from ALWAYS_EXCEPT" class="issue-link" data-issue-key="LU-9590"><del>LU-9590</del></a> tests: remove replay-single tests from ALWAYS_EXCEPT".</p> Related LU-9590 Development Rank 1|hzvy6v: Rank (Obsolete) 9723 Severity