| opendevreview | Matthew Oliver proposed openstack/swift master: Refactor memcache config and MemcacheRing loading https://review.opendev.org/c/openstack/swift/+/820648 | 03:57 |
|---|---|---|
| opendevreview | Matthew Oliver proposed openstack/swift master: Refactored errorlimited out of proxy + global cache https://review.opendev.org/c/openstack/swift/+/820313 | 03:57 |
| opendevreview | Matthew Oliver proposed openstack/swift master: Add global error_limiting to object-updater https://review.opendev.org/c/openstack/swift/+/820486 | 03:57 |
| opendevreview | Matthew Oliver proposed openstack/swift master: db: shard up the DatabaseBroker pending files https://review.opendev.org/c/openstack/swift/+/830551 | 06:17 |
| *** timburke_ is now known as timburke | 12:57 | |
| timburke | reminder that the ops feedback session is going to start soon in https://www.openstack.org/ptg/rooms/havana | 12:57 |
| timburke | etherpad for it is at https://etherpad.opendev.org/p/swift-zed-ops-feedback | 13:01 |
| DHE | are there rules for this session? | 13:32 |
| timburke | https://opendev.org/x/slogging | 13:54 |
| timburke | DHE, sorry, only just saw this -- it's intended to be *very* open-ended -- feel free to break in at any time :-) | 13:55 |
| DHE | cool, thanks. but I'll probably be keeping quiet because I'm also busy with work on my side so... | 13:55 |
| DHE | (I did add something to the etherpad with my name on it, and will respond if it comes up)00 | 13:56 |
| kota | oh, slogging was moved to opendev codebase. | 14:06 |
| kota | seems like the composite ring supports to the source builder version ahead, maybe? https://github.com/openstack/swift/blob/master/swift/common/ring/composite_builder.py#L283-L298 | 14:22 |
| * kota should recall the reason... probably it should be safe operation to composite... | 14:23 | |
| kota | should be for safe | 14:24 |
| DHE | ultimately that is the issue. restoring ring files from the `backups` directory are going to have lower version numbers and set off this check. also the ring composer does not backup its files to this directory either | 14:25 |
| DHE | my "solution" is to edit the builder file (which is just json) to change the expected version numbers. | 14:25 |
| kota | yah, that's what in my idea for temporary operational fix we can do currently. | 14:29 |
| DHE | I am surprised that the ring composer doesn't seem to see more love (though this meeting isn't necessarily representative of the userbase) | 14:33 |
| kota | looking... https://review.opendev.org/c/openstack/swift/+/441921/10..21/swift/common/ring/composite_builder.py#b187 <- thread was the discussion for adding "<" | 14:38 |
| kota | it looks like we want to prevent to try to rebalance if the source component changes nothing | 14:39 |
| kota | might be missing the rolling back case... | 14:42 |
| timburke | maybe useful: https://github.com/fusionfoto/swift-metadata-sync (unfortunately, the original swiftstack repo isn't still around) | 15:01 |
| kota | IIRC, mid-cycle at Denver was the place to discuss storage-policy and ec mostly, maybe... 2014? | 15:03 |
| kota | ah, that's the start line of the ec work and we had a lot of work then. | 15:05 |
| timburke | https://review.opendev.org/c/openstack/swift/+/743002 | 15:25 |
| timburke | oh... the '.ismount' sentinel only works if the "mountpoint" is a symlink... https://github.com/openstack/swift/blob/master/swift/common/utils.py#L4414-L4422 | 16:26 |
| timburke | https://hub.docker.com/r/openstackswift/saio/tags?page=1&name=v2 | 16:36 |
| kota | that was great PTG | 16:53 |
| opendevreview | Tim Burke proposed openstack/swift master: encryption: Cache some cipher context initialization https://review.opendev.org/c/openstack/swift/+/837021 | 19:16 |
| opendevreview | Tim Burke proposed openstack/swift master: api-ref: Document `reverse` param https://review.opendev.org/c/openstack/swift/+/837024 | 19:45 |
| opendevreview | Tim Burke proposed openstack/swift master: tests: Fix swiftclient/requests log level adjustment https://review.opendev.org/c/openstack/swift/+/837036 | 21:47 |
| opendevreview | Tim Burke proposed openstack/swift master: db: Close ReplConnection sockets on errors/timeouts https://review.opendev.org/c/openstack/swift/+/837038 | 22:09 |
| opendevreview | Tim Burke proposed openstack/swift master: diskfile: Reduce tracebacks when consolidating hashes https://review.opendev.org/c/openstack/swift/+/837043 | 23:33 |
| opendevreview | Tim Burke proposed openstack/swift master: db: Tolerate some ENOENTs in quarantine https://review.opendev.org/c/openstack/swift/+/837045 | 23:59 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!