https://tracker.ceph.com/https://tracker.ceph.com/favicon.ico2016-06-06T13:50:13ZCeph CephFS - Bug #16164: mds: enforce a dirfrag limit on entrieshttps://tracker.ceph.com/issues/16164?journal_id=720962016-06-06T13:50:13ZSage Weilsage@newdream.net
<ul><li><strong>Project</strong> changed from <i>Ceph</i> to <i>CephFS</i></li></ul> CephFS - Bug #16164: mds: enforce a dirfrag limit on entrieshttps://tracker.ceph.com/issues/16164?journal_id=721052016-06-06T16:36:19ZPatrick Donnellypdonnell@redhat.com
<ul><li><strong>Assignee</strong> set to <i>Patrick Donnelly</i></li></ul><p>I'm taking a look at this one.</p> CephFS - Bug #16164: mds: enforce a dirfrag limit on entrieshttps://tracker.ceph.com/issues/16164?journal_id=721182016-06-06T18:49:19ZPatrick Donnellypdonnell@redhat.com
<ul><li><strong>Status</strong> changed from <i>New</i> to <i>In Progress</i></li></ul> CephFS - Bug #16164: mds: enforce a dirfrag limit on entrieshttps://tracker.ceph.com/issues/16164?journal_id=721912016-06-07T21:57:45ZGreg Farnumgfarnum@redhat.com
<ul></ul><p>Hmm, I was talking to m0zes (whose situation kicked off this bug) and it turns out the objects actually causing the issue are the stray directories rather than the large user directories.</p>
<p>He says he'd actually be okay with disallowing deletes as well, if it prevents the issue from recurring in his cluster. I wouldn't want to do that by default, but it <strong>might</strong> be pretty simple to set up with a default-off config option. How do others feel about that? I imagine it can hook in in pretty much the same ways as the rename logic will.</p> CephFS - Bug #16164: mds: enforce a dirfrag limit on entrieshttps://tracker.ceph.com/issues/16164?journal_id=722082016-06-08T05:10:07ZXiaoxi Chenxiaoxchen@ebay.com
<ul></ul><p>Greg Farnum wrote:</p>
<blockquote>
<p>Hmm, I was talking to m0zes (whose situation kicked off this bug) and it turns out the objects actually causing the issue are the stray directories rather than the large user directories.</p>
<p>He says he'd actually be okay with disallowing deletes as well, if it prevents the issue from recurring in his cluster. I wouldn't want to do that by default, but it <strong>might</strong> be pretty simple to set up with a default-off config option. How do others feel about that? I imagine it can hook in in pretty much the same ways as the rename logic will.</p>
</blockquote>
<p>Could I have more background on m0zes 's issue? This ticketed is from our previous bug report <a class="external" href="http://tracker.ceph.com/issues/16010#change-71438">http://tracker.ceph.com/issues/16010#change-71438</a></p> CephFS - Bug #16164: mds: enforce a dirfrag limit on entrieshttps://tracker.ceph.com/issues/16164?journal_id=722582016-06-08T13:42:03ZGreg Farnumgfarnum@redhat.com
<ul></ul><p>He's got more info in <a class="external" href="http://tracker.ceph.com/issues/16177">http://tracker.ceph.com/issues/16177</a>.</p>
<p>Basically, a CephFS user created directories large enough to start causing problems with reading the omap entries fast enough for OSD recovery to work (just reading the entries out of leveldb took too long); and then deleted the entries so quickly that the stray directories got <strong>really</strong> large.</p> CephFS - Bug #16164: mds: enforce a dirfrag limit on entrieshttps://tracker.ceph.com/issues/16164?journal_id=730102016-06-17T20:46:36ZPatrick Donnellypdonnell@redhat.com
<ul><li><strong>Status</strong> changed from <i>In Progress</i> to <i>17</i></li></ul><p>PR here: <a class="external" href="https://github.com/ceph/ceph/pull/9789">https://github.com/ceph/ceph/pull/9789</a></p> CephFS - Bug #16164: mds: enforce a dirfrag limit on entrieshttps://tracker.ceph.com/issues/16164?journal_id=737322016-06-30T11:47:45ZJohn Sprayjcspray@gmail.com
<ul><li><strong>Status</strong> changed from <i>17</i> to <i>Pending Backport</i></li><li><strong>Backport</strong> set to <i>jewel</i></li></ul> CephFS - Bug #16164: mds: enforce a dirfrag limit on entrieshttps://tracker.ceph.com/issues/16164?journal_id=737452016-06-30T16:35:42ZNathan Cutlerncutler@suse.cz
<ul><li><strong>Copied to</strong> <i><a class="issue tracker-9 status-3 priority-4 priority-default closed" href="/issues/16560">Backport #16560</a>: jewel: mds: enforce a dirfrag limit on entries</i> added</li></ul> CephFS - Bug #16164: mds: enforce a dirfrag limit on entrieshttps://tracker.ceph.com/issues/16164?journal_id=750192016-07-19T23:06:44ZGreg Farnumgfarnum@redhat.com
<ul><li><strong>Category</strong> set to <i>Performance/Resource Usage</i></li></ul> CephFS - Bug #16164: mds: enforce a dirfrag limit on entrieshttps://tracker.ceph.com/issues/16164?journal_id=770872016-08-23T07:39:38ZLoïc Dacharyloic@dachary.org
<ul><li><strong>Status</strong> changed from <i>Pending Backport</i> to <i>Resolved</i></li></ul>