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-1920 Lustre <p>Please write and attach the test plan for how to test this feature.</p> LU-1920

Create and attach Test Plan for quota testing

Technical task LU-1842 Critical Resolved Fixed Cliff White Jodi Levi Wed, 12 Sep 2012 18:26:19 +0000 Mon, 20 Jan 2014 18:06:07 +0000 Mon, 20 Jan 2014 18:06:07 +0000 Lustre 2.4.0 0 5 <p> Quota test plan<br/> ***************</p> <p>1. Correctness (both ldiskfs and zfs)<br/> --------------</p> <ul class="alternate" type="square"> <li>sanity-quota with SLOW=yes</li> <li>online OST addition</li> <li>failover tests (MDT &amp; OST) with quota on and enforced</li> <li>running all acc-sm tests with quota enabled and enforced</li> </ul> <p>2. Upgrade (ldiskfs only)<br/> ----------</p> <ul class="alternate" type="square"> <li>create a filesystem with 1.8/2.1, enable quota, set some quota limits and create some files</li> <li>upgrade to 2.4 (require tunefs.lustre --quota and lctl conf_param lustre.quota... to be run)</li> <li>check limits and usage</li> <li>remove / create more files and check behavior</li> </ul> <p>3. Client Interoperability (both ldiskfs and zfs)<br/> --------------------------<br/> 2.3 client compatible, clients prior to 2.3 aren't yet (due to EINPROGRESS support)</p> <ul class="alternate" type="square"> <li>run s-q with 2.3 client and 2.4 servers (require to use s-q version from 2.4)</li> <li>compatibility with older clients to be tested once EINPROGRESS patches are landed to b2_! and b1_8.</li> </ul> <p>4. Impact on performance (both ldiskfs and zfs)<br/> ------------------------<br/> Run ior and mdtest on hyperion with:</p> <ul class="alternate" type="square"> <li>fresh filesystem with no quota settings. This should provide us with reference numbers.</li> <li>quota enabled via conf_param. Impact on performance should be null.</li> <li>quota enforcement enabled with a large limit for the user (via setquota). Impact on performance should be close to null.</li> <li>quota enforcement enabled with a limit close to expected usage (should still fit). Impact on performance to be compared with prior lustre release (2.1, 2.2 or 2.3)</li> <li>quota enforcement enabled with a limit smaller than usage (EDQUOT error expected). We should again compare with a prior lustre release.</li> <li>quota disabled via conf_param. Impact on performance should be NULL.</li> </ul> <p>5. DNE support (both ldiskfs and zfs)<br/> ---------------</p> <ul class="alternate" type="square"> <li>sanity-quota with SLOW=yes with multiple MDTs</li> <li>space rebalancing for inodes (never exercised when one single MDT)</li> <li>impact on metadata performance with remote directory (on MDT1) while master still runs on MDT0</li> <li>online MDT addition</li> </ul> <p>Niu, could you please comment on the above TP? Thanks in advance.</p> <p>the test plan looks comprehensive to me. thanks.</p> <p>Does this need to be stressed at scale? If no: does this have a feature that needs to be enabled or disabled during normal SWL testing?</p> <blockquote> <p>Does this need to be stressed at scale?</p></blockquote> <p>Yes, it would be great to run the performance test on hyperion.</p> <blockquote> <p>If no: does this have a feature that needs to be enabled or disabled during normal SWL testing?</p></blockquote> <p>I don't think SWL has any special code for quota. That said, we can probably run it on a filesystem with quota enabled.</p> <p>Johann,<br/> Can you please add the details for what the performance test should entail? Also, how is Quota enabled on Hyperion?<br/> Thank you!</p> <p>I have updated the document. I think we need to run the same test with a lustre version &lt;2.4 in order to compare.</p> <p>For the online OST addition testing, here are the test steps, please comment if it is not enough.</p> <p>1. set up lustre with one OST-0, enable quota and create a file then use up all the blocks<br/> 2. online OST addition to the file system<br/> 3. check if the quota is enabled on the new OST-1<br/> 4. create a new file make sure it is on the new OST-1 and then write, should expect EDQUOT <br/> 5. delete the first file on OST-0 and write again on OST-1, should expect success</p> <p>Sarah, it looks good to me. Thanks.</p> <p>Have we finally run the whole test plan? No major issues found?</p> <p>It appears the tests were run twice on the 2.3.59 tag on both ldiskfs and ZFS and all passed.<br/> Sarah can comment more.</p> <p>That's good news.</p> <p>Sarah, could you please share with me the performance numbers?</p> <p>Hi Johann,</p> <p>Cliff is working on the performance part(section 4), I don't have the permission to access Hyperion. Section 5(DNE part) is now skipped, and for the rest of part have all been tested(by autotest or manually).</p> <p>Just to make it clear that before downgrade, need run "tune2fs -O ^quota" on all MDT and OST devices.</p> <p>Now that DNE is landed, can we retest quota with DNE?</p> <p>Reassign to Cliff for feedback, since those tests need to be run on hyperion, i guess.<br/> I am also interested in the performance results.<br/> BTW, have we run all the quota tests with ZFS as suggested in the test plan?</p> <p>Hi Johann,</p> <p>I've run all tests with ZFS except DNE.</p> <p>This is one year old and I believe done. Reopen if needed</p> Development Project Rank 1|hzv36f: Rank (Obsolete) 4014