| opendevreview | Merged openstack/swift master: container-server: do [end_]marker filtering in SQL query https://review.opendev.org/c/openstack/swift/+/888310 | 00:03 |
|---|---|---|
| opendevreview | Merged openstack/swift master: proxy: add new metrics to account/container_info cache for skip/miss https://review.opendev.org/c/openstack/swift/+/885798 | 01:23 |
| opendevreview | Alistair Coles proposed openstack/swift master: container-server: use LIMIT in get_own_shard_range() query https://review.opendev.org/c/openstack/swift/+/888575 | 13:31 |
| opendevreview | Alistair Coles proposed openstack/swift master: ContainerBroker: clarify get_shard_ranges docstring https://review.opendev.org/c/openstack/swift/+/889095 | 13:31 |
| reid_g | Good morning. I'm looking at cleaning up those containers I mentioned yesterday. I failed to mention that only of of the container replicas is being quarantined. If the account doesn't exist in the cluster. Shouldn't all three copies be quarantined? | 14:08 |
| reid_g | From swift-container-info: "Deleted: False", "Delete Timestamp: 2023-04-29T13:03:59.150870 (1682773439.15087)", "Object Count: 4" | 14:12 |
| reid_g | timburke suggestion was to turn off the container-replicators for a bit and let the cleanup happen but if the other 2 aren't being quarantined would they be cleaned up? Also shouldn't those 4 objects have been deleted if the account/container were deleted? | 14:15 |
| opendevreview | Alistair Coles proposed openstack/swift master: proxy: encapsulate Getter resp, node and parts_iter https://review.opendev.org/c/openstack/swift/+/886994 | 15:34 |
| timburke | reid_g, yes, i would've expected all three replicas to quarantine eventually -- we're sure the container-updater's running on the other nodes? might be worth doing a targeted run. or maybe they did their reporting and so shouldn't need to again :-/ | 18:51 |
| timburke | the four objects may or may not have been deleted, depending on the order of events, which nodes were down when, etc. | 18:54 |
| timburke | the good news is that it's only 4 -- i'd go looking for them manually (list them directly on the container-server node, then do swift-get-nodes for each of them to figure out where they should be. hop on those nodes and start looking for .data/.ts files) | 18:54 |
| opendevreview | Tim Burke proposed openstack/swift master: Add non-ascii meta values to ssync probe test https://review.opendev.org/c/openstack/swift/+/884954 | 19:34 |
| opendevreview | Pete Zaitcev proposed openstack/swift master: doc/SAIO: Update the Fedora section https://review.opendev.org/c/openstack/swift/+/890455 | 20:00 |
| opendevreview | Pete Zaitcev proposed openstack/swift master: docs/SAIO: Make reconciler use the SAIO convention for logs https://review.opendev.org/c/openstack/swift/+/890457 | 20:03 |
| reid_g | I've noticed that some have .lock and .pending files along with the .db. Would those prevent cleanup? | 20:24 |
| reid_g | They are in the same dir as the .db and .pending has a timestamp of a few months ago. The .lock a couple years | 20:25 |
| opendevreview | Tim Burke proposed openstack/pyeclib master: Add release notes to PyPI project URLs https://review.opendev.org/c/openstack/pyeclib/+/888190 | 22:46 |
| opendevreview | Tim Burke proposed openstack/pyeclib master: Drop support for liberasurecode<1.4.0 https://review.opendev.org/c/openstack/pyeclib/+/839643 | 23:31 |
| opendevreview | Jianjian Huo proposed openstack/swift master: container-server: use LIMIT in get_own_shard_range() query https://review.opendev.org/c/openstack/swift/+/888575 | 23:47 |
| opendevreview | Jianjian Huo proposed openstack/swift master: Container-server: add container full namespaces GET interface. https://review.opendev.org/c/openstack/swift/+/890470 | 23:47 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!