| *** CR7_ has quit IRC | 00:08 | |
| *** hrou has joined #openstack-swift | 00:28 | |
| *** haomaiwang has joined #openstack-swift | 00:31 | |
| *** wbhuber has joined #openstack-swift | 00:32 | |
| *** haomaiwang has quit IRC | 00:35 | |
| *** chlong has joined #openstack-swift | 00:50 | |
| *** blmartin has joined #openstack-swift | 01:15 | |
| *** ozialien has quit IRC | 01:37 | |
| *** ozialien has joined #openstack-swift | 01:37 | |
| *** wbhuber has quit IRC | 01:48 | |
| *** yuan has joined #openstack-swift | 01:49 | |
| *** haomaiwang has joined #openstack-swift | 01:50 | |
| *** baojg has joined #openstack-swift | 02:01 | |
| *** chlong has quit IRC | 02:18 | |
| *** david-ly_ has joined #openstack-swift | 02:31 | |
| *** david-lyle has quit IRC | 02:33 | |
| *** blmartin has quit IRC | 02:47 | |
| *** baojg has quit IRC | 02:48 | |
| *** baojg has joined #openstack-swift | 02:51 | |
| *** sanchitmalhotra has joined #openstack-swift | 02:51 | |
| *** silor has joined #openstack-swift | 02:57 | |
| *** DericHorn-HP has joined #openstack-swift | 03:06 | |
| *** marzif has joined #openstack-swift | 03:22 | |
| *** DericHorn-HP has quit IRC | 03:29 | |
| *** chlong has joined #openstack-swift | 03:32 | |
| *** david-lyle has joined #openstack-swift | 03:37 | |
| *** david-ly_ has quit IRC | 03:40 | |
| *** ppai has joined #openstack-swift | 03:42 | |
| *** baojg has quit IRC | 03:44 | |
| *** baojg has joined #openstack-swift | 03:47 | |
| *** jrichli has quit IRC | 03:53 | |
| *** mahatic has joined #openstack-swift | 04:09 | |
| *** ppai has quit IRC | 04:32 | |
| *** mahatic has quit IRC | 04:32 | |
| *** ppai has joined #openstack-swift | 04:34 | |
| openstackgerrit | Emile Snyder proposed openstack/swift: Fix 'swift-ring-builder write_builder' after you remove a device https://review.openstack.org/233488 | 04:43 |
|---|---|---|
| *** chlong has quit IRC | 04:45 | |
| *** klrmn2 has quit IRC | 04:49 | |
| *** hrou has quit IRC | 04:51 | |
| *** nadeem has joined #openstack-swift | 04:54 | |
| *** chlong has joined #openstack-swift | 04:57 | |
| *** trifon has joined #openstack-swift | 05:01 | |
| *** eranrom has joined #openstack-swift | 05:01 | |
| *** silor has quit IRC | 05:01 | |
| *** silor has joined #openstack-swift | 05:10 | |
| *** marzif has quit IRC | 05:10 | |
| *** trifon has quit IRC | 05:18 | |
| *** SkyRocknRoll has joined #openstack-swift | 05:39 | |
| *** peterlisak has joined #openstack-swift | 05:43 | |
| *** baojg has quit IRC | 05:54 | |
| *** nadeem has quit IRC | 06:12 | |
| *** janonymous has quit IRC | 06:20 | |
| openstackgerrit | Peter Lisák proposed openstack/swift: swift-ring-builder can't remove a device with zero weight https://review.openstack.org/233096 | 06:29 |
| *** jmccarthy has joined #openstack-swift | 06:35 | |
| *** jmccarthy has quit IRC | 06:36 | |
| *** sanchitmalhotra has quit IRC | 06:52 | |
| *** sanchitmalhotra has joined #openstack-swift | 06:57 | |
| openstackgerrit | Peter Lisák proposed openstack/swift: swift-ring-builder can't remove a device with zero weight https://review.openstack.org/233096 | 07:09 |
| *** aix has joined #openstack-swift | 07:10 | |
| *** SkyRocknRoll has quit IRC | 07:15 | |
| *** Venkatesh has joined #openstack-swift | 07:23 | |
| *** SkyRocknRoll has joined #openstack-swift | 07:32 | |
| *** janonymous has joined #openstack-swift | 07:39 | |
| openstackgerrit | Christian Schwede proposed openstack/swift: pep8 fix: assertNotEquals -> assertNotEqual https://review.openstack.org/218937 | 07:40 |
| openstackgerrit | Christian Schwede proposed openstack/swift: Enable H234 check (assertEquals is deprecated, use assertEqual) https://review.openstack.org/233517 | 07:40 |
| openstackgerrit | Ondřej Nový proposed openstack/swift: swift-init: SIGKILL daemon after kill_wait period. https://review.openstack.org/229532 | 07:43 |
| *** haypo has joined #openstack-swift | 07:44 | |
| *** joeljwright has joined #openstack-swift | 07:47 | |
| *** ChanServ sets mode: +v joeljwright | 07:47 | |
| *** rledisez has joined #openstack-swift | 07:47 | |
| *** marekkaleta has joined #openstack-swift | 07:56 | |
| *** StevenK has quit IRC | 08:03 | |
| *** eranrom has quit IRC | 08:05 | |
| *** StevenK has joined #openstack-swift | 08:07 | |
| *** marzif has joined #openstack-swift | 08:13 | |
| *** marzif has quit IRC | 08:14 | |
| *** marzif has joined #openstack-swift | 08:15 | |
| *** marzif_ has joined #openstack-swift | 08:16 | |
| *** marzif has quit IRC | 08:20 | |
| *** jordanP has joined #openstack-swift | 08:21 | |
| *** onovy has quit IRC | 08:22 | |
| *** mac_ified has quit IRC | 08:22 | |
| *** onovy has joined #openstack-swift | 08:23 | |
| onovy | Hi, second +2 some please, before it get's into merge conflict again :) https://review.openstack.org/#/c/209645/ thank you | 08:24 |
| *** geaaru has joined #openstack-swift | 08:25 | |
| *** tsubic has quit IRC | 08:35 | |
| *** chlong has quit IRC | 08:35 | |
| *** eranrom has joined #openstack-swift | 08:41 | |
| *** jistr has joined #openstack-swift | 08:46 | |
| *** haomaiwang has quit IRC | 08:47 | |
| *** haomaiwang has joined #openstack-swift | 08:47 | |
| *** aix has quit IRC | 08:58 | |
| *** eranrom has quit IRC | 09:00 | |
| *** haomaiwang has quit IRC | 09:01 | |
| *** haomaiwang has joined #openstack-swift | 09:01 | |
| *** marzif_ has quit IRC | 09:08 | |
| *** marzif_ has joined #openstack-swift | 09:09 | |
| *** aix has joined #openstack-swift | 09:12 | |
| openstackgerrit | Prashanth Pai proposed openstack/swift: Make object creation more atomic in Linux https://review.openstack.org/162243 | 09:20 |
| *** marzif_ has quit IRC | 09:21 | |
| *** haypo has quit IRC | 09:25 | |
| *** haypo has joined #openstack-swift | 09:26 | |
| *** silor has quit IRC | 09:31 | |
| *** chlong has joined #openstack-swift | 09:49 | |
| *** haomaiwang has quit IRC | 09:50 | |
| *** haomaiwang has joined #openstack-swift | 09:50 | |
| *** marzif has joined #openstack-swift | 09:58 | |
| *** eranrom has joined #openstack-swift | 10:00 | |
| *** haomaiwang has quit IRC | 10:01 | |
| *** haomaiwang has joined #openstack-swift | 10:01 | |
| openstackgerrit | venkatesh proposed openstack/swift: remove unused parameters https://review.openstack.org/231347 | 10:24 |
| *** jmccarthy has joined #openstack-swift | 10:24 | |
| janonymous | cschwede , onovy: Thanks for the rebase and changes | 10:28 |
| *** joeljwright has quit IRC | 10:46 | |
| *** haomaiwang has quit IRC | 10:46 | |
| *** haomaiwa_ has joined #openstack-swift | 10:49 | |
| *** haomaiwa_ has quit IRC | 11:01 | |
| *** haomaiwa_ has joined #openstack-swift | 11:01 | |
| *** acoles_ has joined #openstack-swift | 11:01 | |
| *** ChanServ sets mode: +v acoles_ | 11:01 | |
| *** acoles_ has quit IRC | 11:19 | |
| *** acoles_ has joined #openstack-swift | 11:20 | |
| *** ChanServ sets mode: +v acoles_ | 11:20 | |
| *** silor has joined #openstack-swift | 11:22 | |
| *** haomaiwa_ has quit IRC | 11:28 | |
| *** haomaiwang has joined #openstack-swift | 11:28 | |
| *** silor1 has joined #openstack-swift | 11:32 | |
| *** silor has quit IRC | 11:33 | |
| *** silor1 is now known as silor | 11:33 | |
| *** haomaiwang has quit IRC | 11:47 | |
| *** haomaiwang has joined #openstack-swift | 11:48 | |
| hugokuo | anyone knows how to make proxy wsgi listen on IPv6 ? The AF_INET6 requires (address, port, flow info, scope id) 4-tuple | 11:50 |
| hugokuo | But there's only address and port could be configured in proxy-server.conf | 11:50 |
| *** haomaiwang has quit IRC | 12:01 | |
| *** haomaiwang has joined #openstack-swift | 12:01 | |
| *** ppai has quit IRC | 12:03 | |
| *** SkyRocknRoll has quit IRC | 12:10 | |
| *** Venkatesh has quit IRC | 12:14 | |
| *** ppai has joined #openstack-swift | 12:16 | |
| *** haomaiwang has quit IRC | 12:24 | |
| *** joeljwright has joined #openstack-swift | 12:38 | |
| *** ChanServ sets mode: +v joeljwright | 12:38 | |
| *** rbrady has joined #openstack-swift | 12:40 | |
| *** ppai has quit IRC | 12:43 | |
| *** akle has joined #openstack-swift | 12:46 | |
| *** ppai has joined #openstack-swift | 12:49 | |
| tdasilva | acoles: around? | 12:50 |
| acoles | tdasilva: hi | 12:51 |
| tdasilva | acoles: hi, i'm finishing reviewing patch 209645. do you see any problems merging that? do you have any patches that would create big rebase problem for you? | 12:52 |
| patchbot | tdasilva: https://review.openstack.org/#/c/209645/ - pep8 fix: assertEquals -> assertEqual | 12:52 |
| acoles | tdasilva: you are a brave man! 3117 changes, phew! | 12:53 |
| tdasilva | acoles: yeah, my eyes are starting to bleed | 12:53 |
| tdasilva | :P | 12:53 |
| acoles | tdasilva: heh. i anticipate some rebase but it has to happen at some time i guess. | 12:54 |
| tdasilva | acoles: yeah...that's the problem with these py3 patches I guess...they are just a "tough pill to swallow" | 12:54 |
| acoles | tdasilva: yep. the assertEquals one is soft too - its deprecated but doesn't break py3. but hey, now is as good a time as any to take the hit. | 12:56 |
| openstackgerrit | Merged openstack/swift: minutæ: port ClientException tweaks from swiftclient; dict .pop https://review.openstack.org/232812 | 12:56 |
| openstackgerrit | Peter Lisák proposed openstack/swift: swift-recon fails with socket.timeout exception https://review.openstack.org/233616 | 12:56 |
| acoles | tdasilva: thanks for reviewing it! | 12:57 |
| *** rbrady has quit IRC | 12:57 | |
| tdasilva | acoles: yeah...i'm thinking we should just try to get over with py3 patches. | 12:57 |
| *** hrou has joined #openstack-swift | 12:58 | |
| acoles | tdasilva: did you check that in all 3117 cases the arg order is (expected, actual) :P :P | 12:58 |
| * acoles knows it is not and really is kidding | 12:59 | |
| tdasilva | acoles: of course!!! | 12:59 |
| tdasilva | ;) | 12:59 |
| tdasilva | self.assertEqual(proc.args, expected_args) | 12:59 |
| acoles | i guess that could be a follow up patch then ;) | 12:59 |
| *** lpabon has joined #openstack-swift | 13:01 | |
| *** wbhuber has joined #openstack-swift | 13:05 | |
| *** wbhuber has quit IRC | 13:05 | |
| *** wbhuber has joined #openstack-swift | 13:05 | |
| *** breitz has joined #openstack-swift | 13:07 | |
| *** dustins has joined #openstack-swift | 13:08 | |
| *** dustins has quit IRC | 13:08 | |
| *** akle has quit IRC | 13:13 | |
| *** marekkaleta has quit IRC | 13:16 | |
| *** lcurtis has joined #openstack-swift | 13:25 | |
| *** mac_ified has joined #openstack-swift | 13:25 | |
| *** jrichli has joined #openstack-swift | 13:31 | |
| *** haomaiwang has joined #openstack-swift | 13:31 | |
| *** eranrom has quit IRC | 13:34 | |
| *** eranrom has joined #openstack-swift | 13:34 | |
| *** william has joined #openstack-swift | 13:39 | |
| *** william is now known as Guest4123 | 13:40 | |
| *** peluse has joined #openstack-swift | 13:46 | |
| *** peluse__ has quit IRC | 13:47 | |
| *** sanchitmalhotra1 has joined #openstack-swift | 13:52 | |
| *** sanchitmalhotra has quit IRC | 13:54 | |
| *** minwoob has joined #openstack-swift | 14:00 | |
| *** haomaiwang has quit IRC | 14:01 | |
| janonymous | tdasilva, acoles : m sry for so huge patches | 14:01 |
| *** haomaiwang has joined #openstack-swift | 14:01 | |
| janonymous | acoles: I'll make those changes too :P | 14:02 |
| *** sanchitmalhotra1 has quit IRC | 14:08 | |
| *** dustins has joined #openstack-swift | 14:14 | |
| tdasilva | janonymous: no problem, thanks for working on it | 14:20 |
| onovy | tdasilva, acoles: if you want to get rit of that py3 patches, https://review.openstack.org/#/c/218937/ and https://review.openstack.org/#/c/233517/ :) | 14:28 |
| janonymous | tdasilva: happy to do these changes :) | 14:28 |
| *** garthb has joined #openstack-swift | 14:28 | |
| janonymous | Please review +2 ones in these if you get time: https://review.openstack.org/#/q/owner:janonymous+status:open,n,z | 14:29 |
| *** eranrom has quit IRC | 14:30 | |
| *** haomaiwang has quit IRC | 14:35 | |
| *** SkyRocknRoll has joined #openstack-swift | 14:37 | |
| *** haomaiwang has joined #openstack-swift | 14:38 | |
| openstackgerrit | Prashanth Pai proposed openstack/swift: Make object creation more atomic in Linux https://review.openstack.org/162243 | 14:45 |
| *** ppai has quit IRC | 14:46 | |
| *** nadeem has joined #openstack-swift | 14:58 | |
| *** nadeem has quit IRC | 14:59 | |
| *** nadeem has joined #openstack-swift | 14:59 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Don't ssync data when only a durable is missing https://review.openstack.org/232684 | 15:00 |
| *** haomaiwang has quit IRC | 15:01 | |
| *** haomaiwang has joined #openstack-swift | 15:01 | |
| *** minwoob has quit IRC | 15:01 | |
| *** minwoob has joined #openstack-swift | 15:02 | |
| *** HenryG_ is now known as HenryG | 15:03 | |
| *** jlhinson has joined #openstack-swift | 15:07 | |
| openstackgerrit | Merged openstack/swift: pep8 fix: assertEquals -> assertEqual https://review.openstack.org/209645 | 15:09 |
| *** hrou has quit IRC | 15:15 | |
| notmyname | good morning | 15:15 |
| *** hrou has joined #openstack-swift | 15:15 | |
| *** bill_az has joined #openstack-swift | 15:15 | |
| *** acoles_ has quit IRC | 15:18 | |
| openstackgerrit | Emile Snyder proposed openstack/swift: Fix 'swift-ring-builder write_builder' after you remove a device https://review.openstack.org/233488 | 15:18 |
| acoles | notmyname: good morning | 15:22 |
| *** diogogmt has joined #openstack-swift | 15:24 | |
| notmyname | please continue to add stuff to https://etherpad.openstack.org/p/tokyo-summit-swift for what you want to discuss at the summit | 15:27 |
| acoles | notmyname: i was looking at patch 219991, got confused - you had a -2 on version 4 and an argument that the DLO doc should stay where it is but the SLO doc move to the overview. That didn't happen but now you are +2? So was the consensus in the end to move doc to dlo.py and link to there? | 15:28 |
| patchbot | acoles: https://review.openstack.org/#/c/219991/ - Moving DLO functionality doc to the middleware code | 15:28 |
| *** wbhuber has quit IRC | 15:28 | |
| *** wbhuber has joined #openstack-swift | 15:29 | |
| notmyname | acoles: I think the difference is the UI vs the internal structure | 15:29 |
| notmyname | the previous patch set was putting everything into the middleware docs. the new one keeps the actual words with the code, but it inlines those words to the large objects overview page (which is where IMO they should be) | 15:29 |
| acoles | right | 15:30 |
| notmyname | the difference is the large objects overview vs the huge scrolly middleware page | 15:30 |
| acoles | darn, gate-swift-docs link has expired, i'll have to build it locally | 15:30 |
| notmyname | interesting. I didn't know those expired | 15:31 |
| acoles | think i have seen that before | 15:31 |
| notmyname | one really nice thing about doing hand-curated release notes for every release is that it's really easy to put together the release notes for the openstack release cycle | 15:36 |
| notmyname | them: "you haven't done this yet! you have 24 hours!!" me: <copy/paste> "done." | 15:37 |
| *** jlhinson has quit IRC | 15:41 | |
| *** BAKfr has quit IRC | 15:43 | |
| *** BAKfr has joined #openstack-swift | 15:46 | |
| *** zaitcev has joined #openstack-swift | 15:47 | |
| *** ChanServ sets mode: +v zaitcev | 15:47 | |
| acoles | notmyname: there's a couple of typos in that dlo doc patch, you happy for me to correct and the +A? | 15:48 |
| notmyname | yes | 15:48 |
| *** Ailus has joined #openstack-swift | 15:53 | |
| *** Ailus has left #openstack-swift | 15:53 | |
| *** Ailus has joined #openstack-swift | 15:54 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Moving DLO functionality doc to the middleware code https://review.openstack.org/219991 | 15:55 |
| *** BAKfr has quit IRC | 15:55 | |
| *** BAKfr has joined #openstack-swift | 15:58 | |
| *** joeljwright has quit IRC | 15:59 | |
| *** haomaiwang has quit IRC | 16:01 | |
| *** haomaiwang has joined #openstack-swift | 16:01 | |
| *** jlhinson has joined #openstack-swift | 16:01 | |
| notmyname | openstack liberty release notes https://wiki.openstack.org/wiki/ReleaseNotes/Liberty | 16:05 |
| notmyname | (so it's a little more than copy/paste. also have to translate formatting) | 16:05 |
| *** haypo has left #openstack-swift | 16:08 | |
| openstackgerrit | Tim Burke proposed openstack/swift: Add delete markers to versioned_writes middleware https://review.openstack.org/214922 | 16:15 |
| *** Fin1te has joined #openstack-swift | 16:16 | |
| *** haomaiwang has quit IRC | 16:17 | |
| *** haomai___ has joined #openstack-swift | 16:17 | |
| *** gyee has joined #openstack-swift | 16:22 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Re-organise ssync tests https://review.openstack.org/220198 | 16:28 |
| *** aix has quit IRC | 16:31 | |
| *** peterlisak has quit IRC | 16:32 | |
| openstackgerrit | Merged openstack/swift: pep8 fix: assertNotEquals -> assertNotEqual https://review.openstack.org/218937 | 16:32 |
| *** diogogmt has quit IRC | 16:36 | |
| *** willsama has joined #openstack-swift | 16:37 | |
| *** rledisez has quit IRC | 16:39 | |
| *** Guest4123 has quit IRC | 16:40 | |
| *** janonymous_ has joined #openstack-swift | 16:41 | |
| *** jordanP has quit IRC | 16:41 | |
| *** jistr has quit IRC | 16:42 | |
| *** diogogmt has joined #openstack-swift | 16:42 | |
| *** amit213 has quit IRC | 16:45 | |
| *** amit213 has joined #openstack-swift | 16:45 | |
| *** ozialien has quit IRC | 16:56 | |
| openstackgerrit | Merged openstack/swift: Enable H234 check (assertEquals is deprecated, use assertEqual) https://review.openstack.org/233517 | 16:57 |
| *** Fin1te has quit IRC | 17:00 | |
| *** haomai___ has quit IRC | 17:01 | |
| *** haomaiwang has joined #openstack-swift | 17:01 | |
| *** klrmn1 has joined #openstack-swift | 17:05 | |
| *** willsama has quit IRC | 17:06 | |
| *** hrou has quit IRC | 17:13 | |
| *** zhill has joined #openstack-swift | 17:18 | |
| *** peterlisak has joined #openstack-swift | 17:21 | |
| ajiang | Has anyone seen a container out of sync issue, that the last update (PUT to the DB) was about several days ago and the HEAD to the container still shows 3 different numbers by now? I have dumped the container db and comparing it manually and they are all different. 2 copies are close, 1 copy has more deltas. | 17:27 |
| *** diogogmt_ has joined #openstack-swift | 17:29 | |
| *** diogogmt has quit IRC | 17:29 | |
| *** diogogmt_ is now known as diogogmt | 17:29 | |
| *** acoles is now known as acoles_ | 17:33 | |
| openstackgerrit | Greg Lange proposed openstack/swift: Adds read only middleware. https://review.openstack.org/233749 | 17:34 |
| *** mmedvede has joined #openstack-swift | 17:35 | |
| *** peterlisak has quit IRC | 17:36 | |
| *** peterlisak has joined #openstack-swift | 17:44 | |
| *** SkyRocknRoll has quit IRC | 17:48 | |
| notmyname | glange: interesting :-) | 17:49 |
| glange | sometimes you have to give customers a timeout :) | 17:50 |
| notmyname | glange: like a "you! go sit in a corner!" ;-) | 17:51 |
| *** peterlisak has quit IRC | 17:51 | |
| glange | simmer down | 17:52 |
| notmyname | it's something I've heard mentioned at least a few times. I think it's a good idea | 17:54 |
| glange | we need it | 17:55 |
| redbo | it's more like read/delete only | 17:57 |
| *** jmccarthy has quit IRC | 17:57 | |
| notmyname | at minimum it looks like something useful when an account is suspended. | 17:57 |
| glange | don't add more bytes middleware | 17:57 |
| notmyname | yeah :-) | 17:58 |
| notmyname | if I had some image editor software, I'd update http://1.bp.blogspot.com/-In5kpT5ziAk/UR675R1WnGI/AAAAAAAACP8/7xGZteHhVag/s1600/19101819-5431-28.jpg | 18:00 |
| *** haomaiwang has quit IRC | 18:01 | |
| *** haomaiwang has joined #openstack-swift | 18:01 | |
| openstackgerrit | janonymous proposed openstack/swift: Python 3 deprecated the logger.warn method in favor of warning https://review.openstack.org/208222 | 18:03 |
| notmyname | glange: I totally get the use case you've got for this. it's definitely part of a "cluster full" scenario. I've also heard from people that want a "don't change anything" sort of read-only. ie that would also disallow deletes. so when you call this one "read only", it gets confusing | 18:06 |
| *** geaaru has quit IRC | 18:07 | |
| glange | yeah, I see what you mean | 18:07 |
| glange | what if you could configure if deletes were included or not | 18:07 |
| tdasilva | glange: the part I didn't understand is how to set the account header. You mentioned: "and can only be set using a direct client to the account nodes" | 18:07 |
| notmyname | glange: yup. I'd be all over that :-) | 18:08 |
| redbo | The name_check middleware still doesn't know about copy destination. Can we just admit it's never done anything good and get rid of it? | 18:08 |
| glange | tdasilva: you could use swiftly to do that | 18:08 |
| notmyname | glange: are you thinking something like "list of methods to disallow"? or something else? | 18:08 |
| glange | swiftly -D /v1/AUTH_test post -H "X-Account-Sysmeta-Read-Only:t" | 18:08 |
| notmyname | tdasilva: yeah, can't go through the proxy to set it. basically the same as any other sysmeta today | 18:09 |
| glange | notmyname: maybe, I don't know | 18:09 |
| glange | or maybe just allow delete to be turned on or off depending if you are responding to a capacity problem or not | 18:09 |
| onovy | hi, somebody to look to https://review.openstack.org/#/c/233096/ please? | 18:11 |
| notmyname | glange: yeah, I could see that working. maybe like a "read_only=true" and a "but_still_allow_delete=false" as the defaults? | 18:12 |
| ahale | a read/write but disallow delete situation isnt impossible to imagine either | 18:12 |
| onovy | clayg, hi, can we talk about https://review.openstack.org/#/c/230352/ please? | 18:12 |
| *** Fin1te has joined #openstack-swift | 18:13 | |
| notmyname | ahale: I haven't come across that one yet | 18:15 |
| glange | ahale: but a write is kind of like a delete if you write an empty file | 18:15 |
| ahale | my old long dead company many years ago used to make WORM drives/media - write once read many | 18:16 |
| glange | I also have a write only middleware I'm making | 18:16 |
| *** haomaiwang has quit IRC | 18:17 | |
| *** joeljwright has joined #openstack-swift | 18:17 | |
| *** ChanServ sets mode: +v joeljwright | 18:17 | |
| notmyname | ahale: oh, yeah. I've definitely seen that. Very Large Customer, so I don't know why I didn't connect that :-) | 18:17 |
| ahale | lol | 18:17 |
| notmyname | glange: yeah, the trick with WORM is the overwrite situation. very tricky | 18:18 |
| redbo | I invented this really good compression algorithm for write-only storage | 18:18 |
| notmyname | S4? | 18:18 |
| notmyname | http://www.supersimplestorageservice.com | 18:18 |
| redbo | /dev/null is webscale | 18:19 |
| ajiang | @notmyname | 18:21 |
| ajiang | is this some known issue that has been fixed? I am trying to understand the scope of problem and have to address that | 18:21 |
| *** joeljwright has quit IRC | 18:21 | |
| notmyname | ajiang: containers not getting in sync? | 18:21 |
| ajiang | yeah, sorry, I may have mixed your reply with someone else | 18:21 |
| ajiang | but, like to get your opinion on the question asked | 18:22 |
| notmyname | ajiang: I'd guess you're seeing this with large containers (containers with lots of objects) | 18:22 |
| ajiang | not very large, only ~200 objects | 18:23 |
| tdasilva | notmyname: great product, I tried to make a payment but then I get his url: s4/is/a/joke/stop/sending/us/money/readthefineprint.aspx | 18:25 |
| tdasilva | :D | 18:25 |
| *** zhill has quit IRC | 18:26 | |
| notmyname | ajiang: that's not what I'd expect. if it's only got ~200 objects in it, then I'd also guess it doesn't get a lot of activity (req/sec). unless someone is using it like a queue (ie rapidly putting and deleting objects in that container and only ~200 are actually there) | 18:30 |
| *** zhill has joined #openstack-swift | 18:32 | |
| openstackgerrit | janonymous proposed openstack/swift: Use six.moves.map/filter/zip inplace of itertools.izip/imap/ifilter https://review.openstack.org/203414 | 18:34 |
| ajiang | notmyname: do you expect to see the created_at in container DB increment chronically if order by row_id? | 18:35 |
| notmyname | ajiang: in the objects table or the container_info table? | 18:36 |
| ajiang | notmyname: objects table | 18:36 |
| notmyname | ajiang: yes. new requests end up appending a row to the objects table. they don't update an old row (even deletes). so I'd expect all later rows to be newer than previous rows | 18:37 |
| ajiang | notmyname: that is what I expect too. But I do see we have ~40 old entries, then some latest entries and some old entries again. | 18:39 |
| ajiang | don't understand how we can get into that situation | 18:39 |
| notmyname | I haven't looked deeply at container replication in a while, so I don't have an answer for that right now | 18:40 |
| ajiang | notmyname: regarding the scenario that you mentioned "rapidly putting and deleting objects", how do you define rapid? I have seen created_at timestamp are within ~200 seconds and they are not continuous(which should be deleted mixed together) | 18:41 |
| notmyname | ajiang: rapid == tens->hundreds per second | 18:43 |
| ajiang | Thanks. I definitely don't see that high put/delete rate to this container | 18:47 |
| ajiang | notmyname: do you think container ring rebalance could play a role in this problem here? the inconsistency seems start from the time we push out container/account ring after adding more meta data nodes. | 18:48 |
| notmyname | ajiang: it wouldn't have anything to do with the object ring operations you're doing. it probably won't help anything if you're pushing out container rings faster than your container replication time. | 18:52 |
| notmyname | ajiang: it's possible to run the replication process for a certain drive or partition on a drive. you could explicitly run that for the drives or partition that this container is on to see if you can force it to get up to date | 18:54 |
| * notmyname needs to go feed lunch to some kids now | 18:56 | |
| clayg | onovy: i'm not adamently aginst patch 230352 or anything - one time I made swift-init do that and someone pointed out to me the current behavior having the property it did so I abandonded the idea - I didn't put any kind of vote on the change I was just pointing out that it doesn't work that way as some sort of oversight; so I wasn't sure if you had a use-case to change it or if you just thought i'm be more pure/correct from | 19:06 |
| patchbot | clayg: https://review.openstack.org/#/c/230352/ - swift-init return codes | 19:06 |
| *** zhill has quit IRC | 19:15 | |
| *** DericHorn-HP has joined #openstack-swift | 19:16 | |
| *** zhill has joined #openstack-swift | 19:20 | |
| clayg | notmyname: you fixed swift.openstack.org | 19:25 |
| *** silor has quit IRC | 19:27 | |
| ajiang | notmyname: Brian Cline has reported a similar container inconsistent issue last year. https://bugs.launchpad.net/swift/+bug/1390472 , seems no one has looked it since 11/7/2014 | 19:29 |
| openstack | Launchpad bug 1390472 in OpenStack Object Storage (swift) "3-way split brain during NIC outage for dedicated replication network" [Undecided,Incomplete] - Assigned to Brian Cline (briancline) | 19:29 |
| *** lpabon has quit IRC | 19:32 | |
| notmyname | clayg: yeah, looks like fungi restored it. no redirect now, though | 19:34 |
| clayg | ajiang: i think people looked at it - but didn't have any good ideas - maybe related lp bug #1332439 | 19:36 |
| openstack | Launchpad bug 1332439 in OpenStack Object Storage (swift) "Deleted container in deleted account gets unreclaimable after rebalance" [Undecided,In progress] https://launchpad.net/bugs/1332439 - Assigned to Takashi Kajinami (kajinamit) | 19:36 |
| ajiang | clayg: thanks. I will look into that. | 19:38 |
| *** hrou has joined #openstack-swift | 19:42 | |
| openstackgerrit | Merged openstack/swift: py3: Use six.reraise() to reraise an exception https://review.openstack.org/232537 | 19:45 |
| openstackgerrit | Merged openstack/swift: py3: Replace gen.next() with next(gen) https://review.openstack.org/232538 | 19:45 |
| notmyname | hrou: after some digging I found all my SMR test results and graphs! (I was worried I had lost them) | 19:51 |
| *** jmccarthy has joined #openstack-swift | 19:51 | |
| hrou | notmyname, That's great, really glad to hear ! I'm curious if they sync up with the results we have (I wager they will) | 19:53 |
| notmyname | hrou: with the general "don't use SMR for swift", yeah ;-) | 19:53 |
| notmyname | I've also got the microbenchmarks | 19:53 |
| notmyname | I'm trying to make all that into a coherent thing today | 19:54 |
| hrou | notmyname: hrou adds "For the time being lacking any mature SMR suited file systems" yep ! : ) | 19:54 |
| hrou | notmyname: Hey if I can help in any way just let me know ! | 19:54 |
| hrou | notmyname, but yea glad we're on the same page regarding the analysis (at least regarding what's available on the market today for swift operators). | 19:55 |
| notmyname | right | 19:56 |
| hrou | notmyname, how's your other EC talk coming along ? | 19:57 |
| notmyname | hrou: haven't started! ;-) | 20:00 |
| notmyname | or rather, I've got some results, and peluse has some results, and I'm sure it will be awesome | 20:00 |
| hrou | notmyname: Its Ok : ) In a similar situation for our other talk (with peluse too !) Though I think Eran is putting slides together as we speak. | 20:00 |
| hrou | notmyname, yea I think that'll be a great talk that should garner a ton of interest ! (The EC one that is) | 20:00 |
| glange | I'm interested in this talk: http://sched.co/49yL :) | 20:05 |
| *** Fin1te has quit IRC | 20:06 | |
| *** wbhuber has quit IRC | 20:07 | |
| *** wbhuber has joined #openstack-swift | 20:08 | |
| notmyname | yeah, as long as it's not "Swift doesn't scale", it should be a great talk ;-) | 20:11 |
| *** Fin1te has joined #openstack-swift | 20:12 | |
| *** haypo has joined #openstack-swift | 20:22 | |
| openstackgerrit | Ondřej Nový proposed openstack/swift: Autodetect systemctl in SAIO and use it on systemd distros https://review.openstack.org/233388 | 20:41 |
| onovy | clayg, ^ just fixed your command, thanks for it | 20:42 |
| *** dustins has quit IRC | 20:42 | |
| clayg | onovy: yeah idk, i'm sure either would have worked well enough - I never acctually used anything besides `which` for that kind of test before honestly... | 20:48 |
| onovy | i'm not bash guru too :) | 20:48 |
| *** haypo has left #openstack-swift | 20:48 | |
| clayg | onovy: it's really not a great language - i try to avoid it when I can :P | 20:49 |
| onovy | me too! | 20:49 |
| onovy | exactly i'm not sure if x"`aaa`" == x"file" should be used | 20:49 |
| *** zhill has quit IRC | 20:50 | |
| onovy | and why there needs to be " and somewhere else should not be, and so on | 20:50 |
| *** janonymous_ has quit IRC | 20:52 | |
| clayg | onovy: yeah idk, which has the nice property of providing non-zero exit code for command not found so you don't have to mess with the equality shiz - I'm really not sure how much not wanting to rely on the core util is a problem... | 20:53 |
| onovy | type is buildin | 20:54 |
| onovy | and because "#!/bin/bash | 20:54 |
| onovy | type will always be there :) | 20:54 |
| clayg | i 'spose | 20:54 |
| *** zhill has joined #openstack-swift | 20:55 | |
| onovy | 11pm here, good night | 20:59 |
| clayg | wait... so do we use assertEqual or assertEquals now? | 21:07 |
| notmyname | clayg: didn't I see your name on the email about enforcing one of those? | 21:09 |
| clayg | doens't sound like m | 21:10 |
| clayg | *me | 21:10 |
| clayg | somewhere along the way I probably said something like 'wtf are we doing is there even a rule to enforce this' - but then I think there like *was* a rule to enforce it - so yay! | 21:10 |
| clayg | but I can't remember which one :'( | 21:10 |
| clayg | oh well - pep8 will tell me i 'spose | 21:10 |
| notmyname | clayg: you approved https://review.openstack.org/#/c/233517/ this very morning | 21:11 |
| clayg | notmyname: well - tests passed | 21:11 |
| clayg | notmyname: oh - and acoles_ put a +2 on it - I'll +A anything with a +2 from acoles | 21:11 |
| notmyname | ship it! | 21:11 |
| clayg | ok, so no s | 21:12 |
| clayg | no s | 21:12 |
| clayg | no s | 21:12 |
| clayg | I think i've got it | 21:12 |
| clayg | it's a new world! | 21:12 |
| clayg | i can barely contain my excitment for having a code base that *finally* has some consistency about the stuff that *matters*! | 21:12 |
| notmyname | clayg: meh, first couple of weeks after a major openstack release seems to be a good time to get stuff like that reviewed | 21:16 |
| clayg | notmyname: hell yeah! | 21:17 |
| *** zhill has quit IRC | 21:17 | |
| *** zhill_ has joined #openstack-swift | 21:17 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 21:18 |
| *** zhiyan has quit IRC | 21:20 | |
| *** nottrobin has quit IRC | 21:21 | |
| *** serverascode has quit IRC | 21:21 | |
| *** briancurtin has quit IRC | 21:22 | |
| *** Fin1te has quit IRC | 21:24 | |
| mattoliverau | Morning | 21:25 |
| notmyname | mattoliverau: hello! | 21:26 |
| *** nadeem has quit IRC | 21:31 | |
| aerwin3_ | morning mattoliverau: | 21:31 |
| *** nadeem has joined #openstack-swift | 21:32 | |
| *** jrichli has quit IRC | 21:47 | |
| *** nadeem has quit IRC | 21:51 | |
| openstackgerrit | Merged openstack/swift: Add __bool__ for classes that implement __nonzero__ https://review.openstack.org/205545 | 21:53 |
| *** earlephilhower has quit IRC | 21:58 | |
| *** jlhinson has quit IRC | 22:04 | |
| *** 32NAAJR1K has joined #openstack-swift | 22:10 | |
| *** nottrobin has joined #openstack-swift | 22:14 | |
| *** zhill_ has quit IRC | 22:18 | |
| *** zhill has joined #openstack-swift | 22:19 | |
| *** serverascode has joined #openstack-swift | 22:21 | |
| *** Ailus1 has joined #openstack-swift | 22:22 | |
| *** Ailus1 has quit IRC | 22:22 | |
| *** Ailus has quit IRC | 22:24 | |
| *** zhiyan has joined #openstack-swift | 22:25 | |
| torgomatic | hooray, everything's got merge conflicts! | 22:34 |
| mattoliverau | \o/ | 22:34 |
| *** wbhuber_ has joined #openstack-swift | 22:36 | |
| *** wbhuber_ has quit IRC | 22:36 | |
| timburke | it seemed to have been because of that last one, but i don't understand why. i didn't touch swift/common/utils.py at all; i doubt the other patches did either :( | 22:37 |
| torgomatic | could just have been delays in the merge-check gate thingy | 22:37 |
| mattoliverau | A bunch of py3 patches merged over the weekend.. so could be related to that | 22:39 |
| *** wbhuber has quit IRC | 22:40 | |
| *** briancurtin has joined #openstack-swift | 22:41 | |
| timburke | torgomatic: no, don't think so... have a look at patch 214922 if you like. merge conflict earlier this morning, resolved, new patchset uploaded. then it breaks again. but looking at `git diff 1d0d16e 5c24ffc --name-only` (first is patchset 4's parent, second is tip of master) there's no overlap | 22:42 |
| patchbot | timburke: https://review.openstack.org/#/c/214922/ - Add delete markers to versioned_writes middleware | 22:42 |
| torgomatic | huh... must be something weird in the merge-conflict checker then? | 22:43 |
| * torgomatic has no idea | 22:43 | |
| timburke | must be. weird... | 22:43 |
| *** _hrou_ has joined #openstack-swift | 22:43 | |
| *** hrou has quit IRC | 22:47 | |
| *** blmartin has joined #openstack-swift | 22:52 | |
| *** zhill has quit IRC | 22:54 | |
| *** km has joined #openstack-swift | 22:58 | |
| *** km is now known as Guest93107 | 22:58 | |
| *** lpabon has joined #openstack-swift | 23:01 | |
| *** david-ly_ has joined #openstack-swift | 23:01 | |
| *** david-lyle has quit IRC | 23:02 | |
| *** david-lyle has joined #openstack-swift | 23:04 | |
| *** david-ly_ has quit IRC | 23:04 | |
| *** zhill has joined #openstack-swift | 23:06 | |
| *** lpabon has quit IRC | 23:09 | |
| *** diogogmt has quit IRC | 23:13 | |
| *** diogogmt has joined #openstack-swift | 23:15 | |
| *** DericHorn-HP has quit IRC | 23:17 | |
| torgomatic | apparently cinder's tests were broken by the new webob release | 23:17 |
| torgomatic | mostly this makes me happy that swift is no longer saddled with webob | 23:18 |
| torgomatic | #success Swift has been WebOb-free for more than 3 years; thanks redbo and commit 5e3e9a88 | 23:19 |
| openstackstatus | torgomatic: Added success to Success page | 23:19 |
| notmyname | :-) | 23:19 |
| notmyname | torgomatic: and it seems nova is blocked by a recent release of oslo.messaging (and the webob one too). (from the ML just now) | 23:22 |
| torgomatic | hell is other peoples' code | 23:22 |
| torgomatic | (note: your own code from more than a year ago also counts as other peoples' code) | 23:23 |
| redbo | hah.. I'm not necessarily proud of that. But we were at a point where we needed one version of webob for unit tests, another version for functional tests, and neither of those worked with the then-current version of webob. | 23:26 |
| torgomatic | eh, I like it | 23:27 |
| openstackgerrit | Samuel Merritt proposed openstack/swift: Improve SLO PUT error checking https://review.openstack.org/228099 | 23:31 |
| *** proteusguy_ has quit IRC | 23:41 | |
| *** kei_yama has joined #openstack-swift | 23:42 | |
| *** proteusguy_ has joined #openstack-swift | 23:54 | |
| *** diogogmt has quit IRC | 23:57 | |
| *** zhill has quit IRC | 23:58 | |
| *** diogogmt has joined #openstack-swift | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
