| *** threestrands has joined #openstack-swift | 00:00 | |
| kota_ | morning | 01:05 |
|---|---|---|
| openstackgerrit | Tim Burke proposed openstack/swift master: Add some functional CORS tests https://review.opendev.org/533028 | 02:29 |
| openstackgerrit | Tim Burke proposed openstack/swift master: s3api: Pass through CORS headers https://review.opendev.org/710330 | 02:29 |
| openstackgerrit | Tim Burke proposed openstack/swift master: s3api: Allow CORS preflight requests https://review.opendev.org/710355 | 02:29 |
| openstackgerrit | Tim Burke proposed openstack/swift master: WIP: s3api: Allow MPUs via CORS requests https://review.opendev.org/720098 | 02:29 |
| *** psachin has joined #openstack-swift | 03:33 | |
| *** gyee has quit IRC | 04:13 | |
| *** evrardjp has quit IRC | 04:37 | |
| *** evrardjp has joined #openstack-swift | 04:37 | |
| *** threestrands has quit IRC | 05:54 | |
| *** ccamacho has joined #openstack-swift | 07:11 | |
| *** rdejoux has quit IRC | 07:23 | |
| *** rpittau|afk is now known as rpittau | 07:28 | |
| *** rdejoux has joined #openstack-swift | 07:32 | |
| *** ccamacho has quit IRC | 08:20 | |
| *** tkajinam has quit IRC | 08:21 | |
| *** ccamacho has joined #openstack-swift | 08:40 | |
| *** ccamacho has quit IRC | 08:40 | |
| *** ccamacho has joined #openstack-swift | 08:45 | |
| *** rdejoux has quit IRC | 08:47 | |
| *** dtantsur|afk is now known as dtantsur | 08:49 | |
| openstackgerrit | Charles Hsu proposed openstack/python-swiftclient master: Support v3 application credentials auth. https://review.opendev.org/699457 | 09:42 |
| *** ccamacho has quit IRC | 09:44 | |
| *** ccamacho has joined #openstack-swift | 10:05 | |
| *** rpittau is now known as rpittau|bbl | 10:23 | |
| *** rdejoux has joined #openstack-swift | 10:29 | |
| *** rdejoux has quit IRC | 10:34 | |
| *** rdejoux has joined #openstack-swift | 10:41 | |
| *** rdejoux has quit IRC | 12:33 | |
| *** rpittau|bbl is now known as rpittau | 12:49 | |
| *** rdejoux has joined #openstack-swift | 12:54 | |
| *** rcernin has quit IRC | 13:37 | |
| *** tkajinam has joined #openstack-swift | 14:43 | |
| *** dtantsur is now known as dtantsur|brb | 14:55 | |
| *** gyee has joined #openstack-swift | 15:04 | |
| *** ianychoi has quit IRC | 15:13 | |
| timss | Hi folks. We replaced a faulty drive today on one of our object nodes and shortly after it was reformatted and data started getting replicated back (3x replica) a set of drives on multiple other servers experienced XFS corruption. | 15:46 |
| timss | We've had this happen before (2y ago), but not at this rate (9 drives across 4 servers in total), and ended up categorizing it as a freak incident back then after successfully repairing enough drives to ensure little to no risk of data loss. | 15:46 |
| timss | How can XFS become corrupt from simple rsyncing of simple files (and not blocks, or?), even across multiple servers? No appearent issues with the underlying HW either, although heavy I/O being a trigger is not excluded. Repairing the filesystems has proven difficult without discarding dirty metadata (xfs_repair -L), meaning even with replica there's a fair chance of permanent data loss due to the | 15:46 |
| timss | number of drives and no time to copy to handoff. | 15:46 |
| timss | Embarrassing to say this installation hasn't receieved enough love and is still running on Ubuntu 16.04 with Swift 2.7.0 (Ubuntu packages). The servers are HPE with a RAID controller running in JBOD. | 15:46 |
| *** ianychoi has joined #openstack-swift | 15:47 | |
| *** tkajinam has quit IRC | 15:53 | |
| *** dtantsur|brb is now known as dtantsur | 16:11 | |
| *** psachin has quit IRC | 16:17 | |
| *** rpittau is now known as rpittau|afk | 16:33 | |
| *** evrardjp has quit IRC | 16:37 | |
| *** evrardjp has joined #openstack-swift | 16:37 | |
| *** dtantsur is now known as dtantsur|afk | 16:53 | |
| timburke | timss, are your auditors running? if they aren't, that might explain the behavior -- if clients put a bunch of data in then only read a fraction of it, large parts of the filesystem might not be touched until the replicators find an out-of-date node | 17:44 |
| timburke | but otherwise, the auditors should be walking the drives (including reading all data off disk and confirming the MD5) and make FS corruption obvious :-/ | 17:46 |
| timburke | fwiw, newest swift still supports xenial -- you might want to upgrade to at least 2.22.0 (2.23.0 and later need a newer version of eventlet that might be more risk than you want to take on) | 17:49 |
| *** mikecmpbll has joined #openstack-swift | 17:50 | |
| timburke | oh, fyi: tc election results are in http://lists.openstack.org/pipermail/openstack-discuss/2020-April/014135.html | 17:52 |
| clayg | doh! i forgot to vote 😞 | 18:01 |
| *** mikecmpb_ has joined #openstack-swift | 19:12 | |
| *** mikecmpbll has quit IRC | 19:12 | |
| *** mikecmpbll has joined #openstack-swift | 19:16 | |
| *** mikecmpb_ has quit IRC | 19:17 | |
| seongsoocho | morning ! | 20:56 |
| kota_ | morning | 20:56 |
| mattoliverau | Morning | 20:56 |
| timburke | o/ | 20:56 |
| timburke | almost meeting time :-) | 20:57 |
| *** mikecmpbll has quit IRC | 21:13 | |
| *** mikecmpbll has joined #openstack-swift | 21:15 | |
| *** mikecmpbll has quit IRC | 21:16 | |
| *** mikecmpbll has joined #openstack-swift | 21:17 | |
| *** mikecmpbll has quit IRC | 21:23 | |
| openstackgerrit | Romain LE DISEZ proposed openstack/swift master: proxy: stop sending frags to PyECLib with a Queue https://review.opendev.org/704892 | 21:37 |
| *** jv has quit IRC | 22:03 | |
| *** jv has joined #openstack-swift | 22:03 | |
| *** ccamacho has quit IRC | 22:11 | |
| *** rcernin has joined #openstack-swift | 22:45 | |
| *** tkajinam has joined #openstack-swift | 22:59 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!