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-2176 Lustre <p>There seems to be a very strange problem with racer and lustre over zfs.<br/> When you run SLOW=yes REFORMAT=yes FSTYPE=zfs DURATIOn=2700 sh racer.sh<br/> the dd stuff at the start of racer mostly never finish by themselves, even though we see heavy activity of zfs threads and some errors about ENOSPC in the logs.</p> <p>the actual creates and other stuff usually gets run after the timeout is already finished and everything is being killed.</p> <p>Contrast this with an ldiskfs run whre shortly after start we get a lot of messages about "file_create: SIZE=XXX" going all the way until the time is up.</p> <p>Alex is already looking into this one.</p> LU-2176

ZFS: running racer grounds everything to a standstill

Bug Major Resolved Cannot Reproduce Alex Zhuravlev Oleg Drokin LB performance zfs Sun, 14 Oct 2012 23:42:02 +0000 Thu, 30 Apr 2020 11:41:33 +0000 Thu, 30 Apr 2020 11:41:33 +0000 Lustre 2.4.0 0 5 <p>I'd suggest to make few runs with <a href="http://review.whamcloud.com/4403" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/4403</a> and attach the stats here.</p> <p>1) ldiskfs with default settings<br/> 2) zfs with default settings<br/> 3) zfs with OSTSIZE=500000 (rm -f /tmp/lustre-ost* to recreate them with a different size)</p> <p><a href="http://review.whamcloud.com/4403" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/4403</a> should provide a bit more info including amount of data read/written.</p> <p>from what I've seen locally racer does work with ZFS, not that great as with ldiskfs, but quite acceptable, IMHO.</p> <p>I'd suggest Oleg to see actual numbers locally and, hopefully, close the ticket <img class="emoticon" src="https://jira.whamcloud.com/images/icons/emoticons/smile.png" height="16" width="16" align="absmiddle" alt="" border="0"/></p> <p>SLOW=yes DURATION=2700 sh racer.sh:</p> <p>ldiskfs </p> {OST|MDS}SIZE=default: 4740321 opens, 44659 getattrs, 131489 readdirs, 654MB in 145047 reads, 10312MB in 84271 writes<br/> zfs {OST|MDS} <p>SIZE=default: 1287981 opens, 201684 getattrs, 41838 readdirs, 41MB in 5827 reads, 825MB in 36025 writes<br/> zfs </p> {OST|MDS}SIZE=500000: 1830863 opens, 180514 getattrs, 50876 readdirs, 98MB in 17732 reads, 3640MB in 80152 writes<br/> zfs,4GB,{OST|MDS} <p>SIZE=500000: 2941173 opens, 132408 getattrs, 84917 readdirs, 216MB in 7664 reads, 4726MB in 62449 writes</p> <p>given zfs's requirements are significantly higher, I think the numbers are OK at the moment.</p> <p>Oleg, is this still an actual issue?</p> <p>Close old issue, I think ZFS is running this today w/o problems.</p> Related LU-2887 Development Rank 1|hzvad3: Rank (Obsolete) 5214 Severity