https://tracker.ceph.com/https://tracker.ceph.com/favicon.ico2020-01-26T10:43:38ZCeph rgw - Backport #43823: nautilus: rgw: push rgw bucket listing with prefixes and delimiters logic/filtering to the cls layer to improve performancehttps://tracker.ceph.com/issues/43823?journal_id=1570272020-01-26T10:43:38ZNathan Cutlerncutler@suse.cz
<ul><li><strong>Copied from</strong> <i><a class="issue tracker-2 status-3 priority-4 priority-default closed" href="/issues/41051">Feature #41051</a>: rgw: push rgw bucket listing with prefixes and delimiters logic/filtering to the cls layer to improve performance</i> added</li></ul> rgw - Backport #43823: nautilus: rgw: push rgw bucket listing with prefixes and delimiters logic/filtering to the cls layer to improve performancehttps://tracker.ceph.com/issues/43823?journal_id=1570342020-01-26T11:28:45ZNathan Cutlerncutler@suse.cz
<ul><li><strong>Status</strong> changed from <i>New</i> to <i>Need More Info</i></li></ul><p>10 commits - asking a developer to do this one</p> rgw - Backport #43823: nautilus: rgw: push rgw bucket listing with prefixes and delimiters logic/filtering to the cls layer to improve performancehttps://tracker.ceph.com/issues/43823?journal_id=1600202020-02-28T21:03:12ZJ. Eric Ivancichivancich@redhat.com
<ul><li><strong>Status</strong> changed from <i>Need More Info</i> to <i>In Progress</i></li><li><strong>Assignee</strong> set to <i>J. Eric Ivancich</i></li></ul> rgw - Backport #43823: nautilus: rgw: push rgw bucket listing with prefixes and delimiters logic/filtering to the cls layer to improve performancehttps://tracker.ceph.com/issues/43823?journal_id=1612512020-03-18T13:42:51ZMichael Bisig
<ul></ul><p>Hi all,<br />Thanks for the feature and the work to backport it to Nautilus!<br />We are currently experiencing such a bucket index listing limitation in one of our buckets which has millions of objects and we were in contact with Paul for this case as well.</p>
<p>We hope this feature will improve the performance for our usecase.</p>
<p>What is the target backport release of this feature? <br />Thanks in advance for your time.</p> rgw - Backport #43823: nautilus: rgw: push rgw bucket listing with prefixes and delimiters logic/filtering to the cls layer to improve performancehttps://tracker.ceph.com/issues/43823?journal_id=1759092020-09-24T14:21:35ZDan van der Ster
<ul></ul><p>I assume this is not going to be backported to nautilus? If so, it also needs <a class="issue tracker-1 status-3 priority-4 priority-default closed" title="Bug: s3tests that list buckets with prefix/delimiter fail on sharded buckets (Resolved)" href="https://tracker.ceph.com/issues/44353">#44353</a> AFAICT</p> rgw - Backport #43823: nautilus: rgw: push rgw bucket listing with prefixes and delimiters logic/filtering to the cls layer to improve performancehttps://tracker.ceph.com/issues/43823?journal_id=1761382020-09-28T19:14:51ZJ. Eric Ivancichivancich@redhat.com
<ul><li><strong>Status</strong> changed from <i>In Progress</i> to <i>New</i></li></ul><p>Backporting this to nautilus is not a priority for me.</p> rgw - Backport #43823: nautilus: rgw: push rgw bucket listing with prefixes and delimiters logic/filtering to the cls layer to improve performancehttps://tracker.ceph.com/issues/43823?journal_id=1762312020-09-29T15:46:46ZNathan Cutlerncutler@suse.cz
<ul><li><strong>Status</strong> changed from <i>New</i> to <i>Need More Info</i></li></ul><p>large-scale backport already assigned to Eric</p> rgw - Backport #43823: nautilus: rgw: push rgw bucket listing with prefixes and delimiters logic/filtering to the cls layer to improve performancehttps://tracker.ceph.com/issues/43823?journal_id=1762362020-09-29T16:00:05ZJ. Eric Ivancichivancich@redhat.com
<ul></ul><p>Nathan Cutler wrote:</p>
<blockquote>
<p>large-scale backport already assigned to Eric</p>
</blockquote>
<p>I'm unclear what are you looking for, Nathan.... Are you looking for more information from me?</p> rgw - Backport #43823: nautilus: rgw: push rgw bucket listing with prefixes and delimiters logic/filtering to the cls layer to improve performancehttps://tracker.ceph.com/issues/43823?journal_id=1765932020-10-03T01:54:15ZNathan Cutlerncutler@suse.cz
<ul><li><strong>Assignee</strong> deleted (<del><i>J. Eric Ivancich</i></del>)</li></ul><p>Eric, I guess you wanted to unassign yourself so I just did that for you.</p> rgw - Backport #43823: nautilus: rgw: push rgw bucket listing with prefixes and delimiters logic/filtering to the cls layer to improve performancehttps://tracker.ceph.com/issues/43823?journal_id=1765952020-10-03T01:59:42ZNathan Cutlerncutler@suse.cz
<ul></ul><p>Eric Ivancich wrote:</p>
<blockquote>
<p>Nathan Cutler wrote:</p>
<blockquote>
<p>large-scale backport already assigned to Eric</p>
</blockquote>
<p>I'm unclear what are you looking for, Nathan.... Are you looking for more information from me?</p>
</blockquote>
<p>This is a non-trivial backport that needs to be done carefully. So, to clarify a little what I'm trying to achieve. When a backport issue is in "New" status, backporters will be tempted to run the backporting script on it. So we get multiple people trying to run the script, seeing the conflicts, and moving on. Sometimes even the same person runs the script on the issue multiple times.</p>
<p>To avoid wasting people's time, I change the status to "Need More Info" to signify that the backport is non-trivial.</p> rgw - Backport #43823: nautilus: rgw: push rgw bucket listing with prefixes and delimiters logic/filtering to the cls layer to improve performancehttps://tracker.ceph.com/issues/43823?journal_id=1833012021-01-26T21:27:13ZJ. Eric Ivancichivancich@redhat.com
<ul></ul><p>Nathan Cutler wrote:</p>
<blockquote>
<p>Eric Ivancich wrote:</p>
<blockquote>
<p>Nathan Cutler wrote:</p>
<blockquote>
<p>large-scale backport already assigned to Eric</p>
</blockquote>
<p>I'm unclear what are you looking for, Nathan.... Are you looking for more information from me?</p>
</blockquote>
<p>This is a non-trivial backport that needs to be done carefully. So, to clarify a little what I'm trying to achieve. When a backport issue is in "New" status, backporters will be tempted to run the backporting script on it. So we get multiple people trying to run the script, seeing the conflicts, and moving on. Sometimes even the same person runs the script on the issue multiple times.</p>
<p>To avoid wasting people's time, I change the status to "Need More Info" to signify that the backport is non-trivial.</p>
</blockquote>
<p>Nathan,</p>
<p>I don't think this backport will ever be done; I don't think it's worth the effort. I'd like to close this as Rejected. And then I'd like to remove the backport entry from <a class="external" href="https://tracker.ceph.com/issues/41051">https://tracker.ceph.com/issues/41051</a> and mark that as resolved.</p>
<p>Would you be OK with the above?</p>
<p>Thanks,</p>
<p>Eric</p> rgw - Backport #43823: nautilus: rgw: push rgw bucket listing with prefixes and delimiters logic/filtering to the cls layer to improve performancehttps://tracker.ceph.com/issues/43823?journal_id=1833132021-01-27T08:17:55ZNathan Cutlerncutler@suse.cz
<ul></ul><blockquote>
<p>I'd like to close this as Rejected. And then I'd like to remove the backport entry from <a class="external" href="https://tracker.ceph.com/issues/41051">https://tracker.ceph.com/issues/41051</a> and mark that as resolved.</p>
</blockquote>
<p>Eric, as far as I'm concerned, you are welcome to mark any backport issue as Rejected when and as you see fit.</p>
<p>However, when a backport ticket is placed in "Rejected" state, the backport should remain in the Backport field of the parent issue, so please don't remove it. If it is removed, the script sees that a backport issue exists for a backport that was not called for and complains, which then causes me to re-add the backport to the Backport field to appease the script ;-)</p> rgw - Backport #43823: nautilus: rgw: push rgw bucket listing with prefixes and delimiters logic/filtering to the cls layer to improve performancehttps://tracker.ceph.com/issues/43823?journal_id=1833142021-01-27T08:18:06ZNathan Cutlerncutler@suse.cz
<ul><li><strong>Assignee</strong> set to <i>J. Eric Ivancich</i></li></ul> rgw - Backport #43823: nautilus: rgw: push rgw bucket listing with prefixes and delimiters logic/filtering to the cls layer to improve performancehttps://tracker.ceph.com/issues/43823?journal_id=1837782021-01-27T20:42:42ZJ. Eric Ivancichivancich@redhat.com
<ul><li><strong>Status</strong> changed from <i>Need More Info</i> to <i>Rejected</i></li></ul><p>The engineering effort required for this non-trivial backport is not worth it.</p>