<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class="">
Hi Luke,
<div class=""><br class="">
</div>
<div class="">I\u2019ve got an off the wall suggestion for you, which may or may not work depending on whether or not you have any UID conflicts with old and new UIDs \u2026 this won\u2019t actually speed things up but it will eliminate the \u201cdowntime\u201d for your users. And
the big caveat is that there can\u2019t be any UID conflicts \u2026 i.e. someone\u2019s new UID can\u2019t be someone else\u2019s old UID.</div>
<div class=""><br class="">
</div>
<div class="">Given that \u2026 what if you set an ACL to allow access to both their old and new UIDs \u2026 then change their UID to the new UID \u2026 then chown the files to the new UID and remove the ACL? More work for you, but no downtime for them.</div>
<div class=""><br class="">
</div>
<div class="">We actually may need to do something similar as we will need to change Windows-assigned UID\u2019s based on SIDs to \u201ccorrect\u201d UIDs at some point in the future on one of our storage systems.</div>
<div class=""><br class="">
</div>
<div class="">If someone has a better way to solve your problem, I hope they\u2019ll post it to the list, as it may help us as well. HTHAL. Thanks\u2026</div>
<div class=""><br class="">
</div>
<div class="">Kevin</div>
<div class=""><br class="">
<div>
<blockquote type="cite" class="">
<div class="">On Jun 30, 2017, at 10:20 AM, <a href="mailto:hpc-luke@uconn.edu" class="">
hpc-luke@uconn.edu</a> wrote:</div>
<br class="Apple-interchange-newline">
<div class="">
<div class="">Hello, <br class="">
<br class="">
<span class="Apple-tab-span" style="white-space:pre"></span>We're trying to change most of our users uids, is there a clean way to<br class="">
migrate all of one users files with say `mmapplypolicy`? We have to change the<br class="">
owner of around 273539588 files, and my estimates for runtime are around 6 days.<br class="">
<br class="">
<span class="Apple-tab-span" style="white-space:pre"></span>What we've been doing is indexing all of the files and splitting them up by<br class="">
owner which takes around an hour, and then we were locking the user out while we<br class="">
chown their files. I made it multi threaded as it weirdly gave a 10% speedup<br class="">
despite my expectation that multi threading access from a single node would not<br class="">
give any speedup.<br class="">
<br class="">
<span class="Apple-tab-span" style="white-space:pre"></span>Generally I'm looking for advice on how to make the chowning faster. Would<br class="">
spreading the chowning processes over multiple nodes improve performance? Should<br class="">
I not stat the files before running lchown on them, since lchown checks the file<br class="">
before changing it? I saw mention of inodescan(), in an old gpfsug email, which<br class="">
speeds up disk read access, by not guaranteeing that the data is up to date. We<br class="">
have a maintenance day coming up where all users will be locked out, so the file<br class="">
handles(?) from GPFS's perspective will not be able to go stale. Is there a<br class="">
function with similar constraints to inodescan that I can use to speed up this<br class="">
process?<br class="">
<br class="">
Thank you for your time,<br class="">
<br class="">
Luke<br class="">
Storrs-HPC<br class="">
University of Connecticut<br class="">
_______________________________________________<br class="">
gpfsug-discuss mailing list<br class="">
gpfsug-discuss at <a href="http://spectrumscale.org" class="">spectrumscale.org</a><br class="">
<a href="http://gpfsug.org/mailman/listinfo/gpfsug-discuss" class="">http://gpfsug.org/mailman/listinfo/gpfsug-discuss</a><br class="">
</div>
</div>
</blockquote>
</div>
<br class="">
</div>
<br class="">
<br class="">
<div class="">
<div class="">\u2014</div>
<div class="">Kevin Buterbaugh - Senior System Administrator</div>
<div class="">Vanderbilt University - Advanced Computing Center for Research and Education</div>
<div class=""><a href="mailto:Kevin.Buterbaugh@vanderbilt.edu" class="">Kevin.Buterbaugh@vanderbilt.edu</a> - (615)875-9633</div>
<div class=""><br class="">
</div>
<br class="Apple-interchange-newline">
</div>
<br class="">
</body>
</html>