| *** itlinux has joined #openstack-swift | 00:14 | |
| openstackgerrit | Samuel Merritt proposed openstack/swift master: Don't make async_pendings during object expiration https://review.openstack.org/532383 | 00:14 |
|---|---|---|
| *** itlinux has quit IRC | 00:15 | |
| *** bigdogstl has joined #openstack-swift | 00:19 | |
| *** s992 has joined #openstack-swift | 00:19 | |
| *** itlinux has joined #openstack-swift | 00:21 | |
| *** s992 has quit IRC | 00:21 | |
| *** bigdogstl has quit IRC | 00:25 | |
| *** bigdogstl has joined #openstack-swift | 00:51 | |
| *** tovin07_ has joined #openstack-swift | 00:57 | |
| *** tovin07 has joined #openstack-swift | 00:57 | |
| *** bigdogstl has quit IRC | 01:07 | |
| *** mvk has quit IRC | 01:10 | |
| *** mvk has joined #openstack-swift | 01:25 | |
| openstackgerrit | yanwei proposed openstack/swift master: Solve the zombie process problem of Auditor https://review.openstack.org/533517 | 01:26 |
| openstackgerrit | yanwei proposed openstack/swift master: Solve the zombie process problem of Auditor https://review.openstack.org/533517 | 01:30 |
| *** kei_yama has quit IRC | 01:33 | |
| *** m_kazuhiro has joined #openstack-swift | 01:35 | |
| m_kazuhiro | good morning | 01:40 |
| mattoliverau | m_kazuhiro: morning | 01:45 |
| m_kazuhiro | mattoliverau: morning! Thank you for your review comments. I want your follow up patch. | 01:49 |
| openstackgerrit | yanwei proposed openstack/swift master: Solve the zombie process problem of Auditor https://review.openstack.org/533517 | 01:56 |
| openstackgerrit | yanwei proposed openstack/swift master: Solve the zombie process problem of Auditor https://review.openstack.org/533517 | 01:59 |
| *** bigdogstl has joined #openstack-swift | 02:00 | |
| *** threestrands_ has joined #openstack-swift | 02:02 | |
| kota_ | good morning | 02:02 |
| *** jappleii__ has quit IRC | 02:04 | |
| *** bigdogstl has quit IRC | 02:05 | |
| *** bigdogstl has joined #openstack-swift | 02:08 | |
| *** bigdogstl has quit IRC | 02:17 | |
| *** kei_yama has joined #openstack-swift | 02:20 | |
| mattoliverau | kota_: morning | 02:29 |
| kota_ | mattoliverau: o/ | 02:29 |
| *** JimCheung has quit IRC | 02:34 | |
| *** JimCheung has joined #openstack-swift | 02:35 | |
| *** JimCheung has quit IRC | 02:39 | |
| *** tovin07__ has joined #openstack-swift | 02:42 | |
| *** tovin07_ has quit IRC | 02:45 | |
| *** bigdogstl has joined #openstack-swift | 02:52 | |
| *** bkopilov has quit IRC | 02:58 | |
| *** bigdogstl has quit IRC | 02:58 | |
| *** bigdogstl has joined #openstack-swift | 03:04 | |
| *** JimCheung has joined #openstack-swift | 03:07 | |
| *** JimCheung has quit IRC | 03:12 | |
| *** bigdogstl has quit IRC | 03:19 | |
| *** kei_yama has quit IRC | 03:44 | |
| *** bigdogstl has joined #openstack-swift | 03:46 | |
| *** bigdogstl has quit IRC | 03:51 | |
| *** kei_yama has joined #openstack-swift | 04:01 | |
| *** tanee has quit IRC | 04:05 | |
| *** links has joined #openstack-swift | 04:06 | |
| *** armaan has quit IRC | 04:18 | |
| *** bkopilov has joined #openstack-swift | 04:25 | |
| *** bigdogstl has joined #openstack-swift | 04:26 | |
| *** m_kazuhiro_ has joined #openstack-swift | 04:34 | |
| *** m_kazuhiro has quit IRC | 04:37 | |
| *** bigdogstl has quit IRC | 04:37 | |
| *** gyee has quit IRC | 04:42 | |
| *** bigdogstl has joined #openstack-swift | 04:46 | |
| *** bigdogstl has quit IRC | 04:51 | |
| *** bigdogstl has joined #openstack-swift | 04:58 | |
| *** bigdogstl has quit IRC | 05:03 | |
| *** bigdogstl has joined #openstack-swift | 05:23 | |
| *** psachin has joined #openstack-swift | 05:26 | |
| *** bigdogstl has quit IRC | 05:30 | |
| *** cshastri has joined #openstack-swift | 05:34 | |
| *** bigdogstl has joined #openstack-swift | 05:42 | |
| *** m_kazuhiro_ has quit IRC | 05:49 | |
| *** cshastri has quit IRC | 05:50 | |
| *** bigdogstl has quit IRC | 05:55 | |
| *** SkyRocknRoll_ has joined #openstack-swift | 05:59 | |
| *** arunman has joined #openstack-swift | 06:02 | |
| *** bigdogstl has joined #openstack-swift | 06:11 | |
| *** bigdogstl has quit IRC | 06:20 | |
| *** bigdogstl has joined #openstack-swift | 06:26 | |
| openstackgerrit | guotao proposed openstack/swift master: Update http with https https://review.openstack.org/534119 | 06:31 |
| *** arunman has quit IRC | 06:39 | |
| *** bigdogstl has quit IRC | 06:41 | |
| *** geaaru has quit IRC | 06:58 | |
| *** cshastri has joined #openstack-swift | 07:04 | |
| *** links has quit IRC | 07:06 | |
| *** rcernin has quit IRC | 07:14 | |
| *** bigdogstl has joined #openstack-swift | 07:15 | |
| *** threestrands_ has quit IRC | 07:18 | |
| *** links has joined #openstack-swift | 07:22 | |
| *** bigdogstl has quit IRC | 07:23 | |
| *** bigdogstl has joined #openstack-swift | 07:26 | |
| *** bigdogstl has quit IRC | 07:31 | |
| *** links has quit IRC | 07:39 | |
| *** hseipp has joined #openstack-swift | 07:42 | |
| *** armaan has joined #openstack-swift | 07:44 | |
| *** cshastri has quit IRC | 07:49 | |
| *** gkadam has joined #openstack-swift | 08:00 | |
| *** neonpastor has quit IRC | 08:00 | |
| *** neonpastor has joined #openstack-swift | 08:02 | |
| *** tesseract has joined #openstack-swift | 08:05 | |
| *** links has joined #openstack-swift | 08:15 | |
| *** bigdogstl has joined #openstack-swift | 08:18 | |
| *** arunman has joined #openstack-swift | 08:24 | |
| *** bigdogstl has quit IRC | 08:29 | |
| arunman | kota_: one of my metadata upload operation to swift is stalled for 27 hours before failing. I don't see any obvious errors in the swift logs related to the same. Any idea why it would have failed? | 08:42 |
| arunman | there were other upload operations that were going on successfully during this time. It just happened with one. | 08:43 |
| kota_ | arunman: what error did you get from the client? timeouts? | 08:43 |
| *** geaaru has joined #openstack-swift | 08:44 | |
| kota_ | and let me correct what operation you did. POST metadata to an object, isn't it? | 08:44 |
| arunman | not really... after 27 hours, I see this in the nova compute logs.. | 08:45 |
| arunman | 2017-12-21 00:52:27.966 2757 INFO swiftclient [-] RESP STATUS: 502 Proxy Error | 08:45 |
| arunman | 2017-12-21 00:52:27.966 2757 INFO swiftclient [-] RESP HEADERS: {u'Date': u'Tue, 19 Dec 2017 20:05:47 GMT', u'Content-Length': u'487', u'Content-Type': u'text/html; charset=iso-8859-1', u'Server': u'Apache'} | 08:45 |
| arunman | 2017-12-21 00:52:27.967 2757 INFO swiftclient [-] RESP BODY: <!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 2.0//EN"> | 08:45 |
| arunman | <html><head> | 08:45 |
| arunman | <title>502 Proxy Error</title> | 08:45 |
| arunman | </head><body> | 08:45 |
| arunman | <h1>Proxy Error</h1> | 08:45 |
| arunman | <p>The proxy server received an invalid^M | 08:45 |
| arunman | response from an upstream server.<br />^M | 08:45 |
| arunman | The proxy server could not handle the request <em><a href="/v1/AUTH_37f6563556844643ac7bd69d14bbc3e7/powervm_nvram">PUT /v1/AUTH_37f6563556844643ac7bd69d14bbc3e7/powervm_nvram</a></em>.<p> | 08:45 |
| arunman | Reason: <strong>Error reading from remote server</strong></p></p> | 08:45 |
| arunman | </body></html> | 08:45 |
| arunman | this is the error it logged when it failed after 27 hours... | 08:46 |
| *** cshastri has joined #openstack-swift | 08:46 | |
| arunman | so wondering if I should have a LP bug to have a timeout to prevent such cases. | 08:47 |
| kota_ | it looks like not a POST metadata to an object. | 08:48 |
| kota_ | seems like PUT container (create container or add metadata to a container) | 08:48 |
| arunman | yep.. it was a add metadata/PUT call | 08:49 |
| kota_ | then, actually, 502 won't happen at swift proxy server. | 08:49 |
| *** bigdogstl has joined #openstack-swift | 08:51 | |
| kota_ | i did grep with the code but 502 Bad Gateway cannot be occurred except only one case in bulk that should not happen container operation | 08:51 |
| kota_ | it looks like the body shouldn't be generated from Swift so i'm now doubting network config between the client to swift... | 08:53 |
| kota_ | I had an experience to get odd http response when I was playing under http proxy environment. | 08:54 |
| kota_ | if you could try to trace the request which network it was going through, it would be a hint to resolve the issue. | 08:55 |
| arunman | Ok.. like I mentioned there were no obvious errors in swift logs. So bit clueless on what could have caused this. In general, do you think it's better to have a timeout to fail metadata upload if it's taking longer than usual? | 08:56 |
| *** bigdogstl has quit IRC | 08:56 | |
| arunman | I can try to check if it's something down to the netowkr | 08:56 |
| *** cbartz has joined #openstack-swift | 09:02 | |
| kota_ | arunman: so thinking of the request couldn't reach out to your swift, it's under in the correct logic for the why your swift didn't log the obvious error. | 09:14 |
| kota_ | exactly, timeout would mitigate the time span to find the problem though. | 09:15 |
| *** bigdogstl has joined #openstack-swift | 09:20 | |
| *** mvk has quit IRC | 09:20 | |
| arunman | kota_, not sure if I missed mentioning this. I meant, there were no errors in swift logs. However, I do see logs related to the "PUT" operation for the metadata upload in question as soon as it was triggered. So I believe it did hit swift... but, something went wrong from there. | 09:22 |
| *** armaan has quit IRC | 09:22 | |
| *** bigdogstl has quit IRC | 09:31 | |
| openstackgerrit | Kazuhiro MIYAHARA proposed openstack/swift master: WIP: Update object expirer to use general task queue system https://review.openstack.org/517389 | 09:35 |
| *** zaitcev has quit IRC | 09:39 | |
| *** zaitcev has joined #openstack-swift | 09:40 | |
| *** ChanServ sets mode: +v zaitcev | 09:40 | |
| acoles | good morning | 09:42 |
| *** armaan has joined #openstack-swift | 09:51 | |
| *** mvk has joined #openstack-swift | 09:53 | |
| *** bigdogstl has joined #openstack-swift | 10:00 | |
| *** tovin07__ has quit IRC | 10:04 | |
| openstackgerrit | Matthew Oliver proposed openstack/swift master: Doc follow up: expirer using general task queue https://review.openstack.org/534219 | 10:09 |
| mattoliverau | acoles: morning | 10:10 |
| acoles | hi mattoliverau | 10:10 |
| *** bigdogstl has quit IRC | 10:11 | |
| *** cbartz has quit IRC | 10:36 | |
| *** psachin has quit IRC | 10:37 | |
| *** cbartz has joined #openstack-swift | 10:37 | |
| *** kei_yama has quit IRC | 10:39 | |
| *** SkyRocknRoll_ has quit IRC | 10:41 | |
| *** psachin has joined #openstack-swift | 10:43 | |
| *** arunman has quit IRC | 10:51 | |
| *** SkyRocknRoll_ has joined #openstack-swift | 10:54 | |
| *** mvk has quit IRC | 10:58 | |
| openstackgerrit | Christopher Bartz proposed openstack/swift master: Send correct number of X-Delete-At-* headers https://review.openstack.org/524548 | 11:02 |
| *** psachin has quit IRC | 11:04 | |
| *** bkopilov has quit IRC | 11:09 | |
| *** bigdogstl has joined #openstack-swift | 11:12 | |
| *** mvk has joined #openstack-swift | 11:13 | |
| *** psachin has joined #openstack-swift | 11:13 | |
| *** nakagawamsa has joined #openstack-swift | 11:17 | |
| *** bigdogstl has quit IRC | 11:19 | |
| *** HW-Peter has joined #openstack-swift | 12:12 | |
| *** armaan has quit IRC | 12:14 | |
| *** bigdogstl has joined #openstack-swift | 12:16 | |
| *** HW-Peter has quit IRC | 12:16 | |
| *** HW-Peter has joined #openstack-swift | 12:17 | |
| *** HWPeter has joined #openstack-swift | 12:17 | |
| *** Peter-HW has joined #openstack-swift | 12:18 | |
| *** Peter-HW has quit IRC | 12:21 | |
| *** Peter-HW has joined #openstack-swift | 12:21 | |
| *** Peter-HW has quit IRC | 12:22 | |
| *** bigdogstl has quit IRC | 12:23 | |
| *** Peter-HW has joined #openstack-swift | 12:23 | |
| *** Peter-HW has quit IRC | 12:25 | |
| *** Peter-HW has joined #openstack-swift | 12:25 | |
| *** Peter-HW has quit IRC | 12:26 | |
| *** Peter-HW has joined #openstack-swift | 12:27 | |
| *** Peter-HW has quit IRC | 12:29 | |
| *** pcaruana has joined #openstack-swift | 12:33 | |
| *** Peter-HW has joined #openstack-swift | 12:36 | |
| *** bigdogstl has joined #openstack-swift | 12:38 | |
| *** HW_Peter has joined #openstack-swift | 12:39 | |
| *** bigdogstl has quit IRC | 12:43 | |
| *** bigdogstl has joined #openstack-swift | 12:52 | |
| *** SkyRocknRoll_ has quit IRC | 12:54 | |
| *** bigdogstl has quit IRC | 12:58 | |
| *** Peter-HW has quit IRC | 12:59 | |
| *** HW_Peter has quit IRC | 12:59 | |
| *** HWPeter has quit IRC | 12:59 | |
| *** HW-Peter has quit IRC | 12:59 | |
| *** HW-Peter has joined #openstack-swift | 12:59 | |
| *** bigdogstl has joined #openstack-swift | 13:23 | |
| *** bigdogstl has quit IRC | 13:28 | |
| *** bigdogstl has joined #openstack-swift | 13:33 | |
| *** bigdogstl has quit IRC | 13:43 | |
| openstackgerrit | Merged openstack/swift feature/deep: Pass scanner_batch_size to find_shard_ranges https://review.openstack.org/531144 | 13:48 |
| *** arunman has joined #openstack-swift | 13:51 | |
| *** links has quit IRC | 14:05 | |
| *** psachin has quit IRC | 14:12 | |
| *** arunman has quit IRC | 14:37 | |
| *** arunman has joined #openstack-swift | 14:48 | |
| *** arunman has quit IRC | 14:55 | |
| *** tovin07_ has joined #openstack-swift | 15:00 | |
| *** cshastri has quit IRC | 15:05 | |
| *** arunman has joined #openstack-swift | 15:07 | |
| *** bigdogstl has joined #openstack-swift | 15:13 | |
| *** arunman has quit IRC | 15:14 | |
| *** ukaynar has joined #openstack-swift | 15:19 | |
| *** bigdogstl has quit IRC | 15:22 | |
| *** links has joined #openstack-swift | 15:29 | |
| *** itlinux has quit IRC | 15:42 | |
| *** arunman has joined #openstack-swift | 15:47 | |
| *** bigdogstl has joined #openstack-swift | 15:50 | |
| *** links has quit IRC | 15:51 | |
| *** arunman has quit IRC | 15:55 | |
| *** bigdogstl has quit IRC | 16:02 | |
| *** bkopilov has joined #openstack-swift | 16:08 | |
| *** arunman has joined #openstack-swift | 16:21 | |
| *** links has joined #openstack-swift | 16:25 | |
| *** gyee has joined #openstack-swift | 16:41 | |
| *** itlinux has joined #openstack-swift | 16:44 | |
| *** bigdogstl has joined #openstack-swift | 16:47 | |
| *** ukaynar has quit IRC | 16:48 | |
| *** ukaynar has joined #openstack-swift | 16:48 | |
| *** itlinux has quit IRC | 16:49 | |
| *** bigdogstl has quit IRC | 16:51 | |
| *** arunman has quit IRC | 16:52 | |
| *** hseipp has quit IRC | 16:52 | |
| notmyname | good morning | 16:53 |
| *** bigdogstl has joined #openstack-swift | 16:54 | |
| *** mvk has quit IRC | 16:54 | |
| *** links has quit IRC | 16:54 | |
| *** chsc has joined #openstack-swift | 16:55 | |
| *** chsc has joined #openstack-swift | 16:55 | |
| *** tovin07_ has quit IRC | 16:57 | |
| *** bigdogstl has quit IRC | 16:59 | |
| *** fishbone_ has quit IRC | 17:09 | |
| *** tesseract has quit IRC | 17:21 | |
| *** cbartz has quit IRC | 17:23 | |
| *** JimCheung has joined #openstack-swift | 17:26 | |
| *** bigdogstl has joined #openstack-swift | 17:28 | |
| openstackgerrit | Monty Taylor proposed openstack/swift master: Add a note about the cost of COPY for setting metadata https://review.openstack.org/534389 | 17:33 |
| *** zigo has quit IRC | 17:33 | |
| *** openstackgerrit has quit IRC | 17:33 | |
| *** geaaru has quit IRC | 17:34 | |
| *** ukaynar has quit IRC | 17:35 | |
| *** ukaynar has joined #openstack-swift | 17:36 | |
| *** zigo has joined #openstack-swift | 17:37 | |
| *** HW_Peter has joined #openstack-swift | 17:38 | |
| *** gkadam has quit IRC | 17:38 | |
| *** bigdogstl has quit IRC | 17:38 | |
| acoles | when 'remaining time' has been 0 min for over 6 hours you start to question it | 17:39 |
| *** openstackgerrit has joined #openstack-swift | 17:51 | |
| openstackgerrit | Merged openstack/swift master: Add request_tries option to object-expirer.conf-sample https://review.openstack.org/533708 | 17:51 |
| openstackgerrit | Tim Burke proposed openstack/swift master: expirer: don't retry 404, 409, or 412 failures https://review.openstack.org/534399 | 18:02 |
| timburke | acoles: torgomatic: ^^^ | 18:02 |
| acoles | timburke: ack | 18:02 |
| timburke | maybe that should go further up, in internal client? | 18:02 |
| timburke | not sure. | 18:02 |
| *** bigdogstl has joined #openstack-swift | 18:06 | |
| timburke | also worth noting? https://github.com/openstack/swift/blob/2.16.0/swift/common/utils.py#L1636-L1651 | 18:07 |
| *** bigdogstl has quit IRC | 18:11 | |
| *** links has joined #openstack-swift | 18:16 | |
| *** bigdogstl has joined #openstack-swift | 18:19 | |
| *** bigdogstl has quit IRC | 18:24 | |
| *** bigdogstl has joined #openstack-swift | 18:26 | |
| *** links has quit IRC | 18:34 | |
| *** bigdogstl has quit IRC | 18:35 | |
| *** links has joined #openstack-swift | 18:36 | |
| acoles | mattoliverau: timburke FYI I am currently stuck waiting for p 532549 to make it through zuul before I can update the chain of patches I have on feature/deep. my updates are based on it but also other stuff that has merged so I can't rebase my chain yet :/ | 18:44 |
| patchbot | https://review.openstack.org/#/c/532549/ - swift (feature/deep) - Test that all object methods get shard container u... | 18:44 |
| timburke | ack. i'll keep an eye on the sharding queue today, try to get that chain shorter for you | 18:45 |
| acoles | timburke: thanks, if you could look for any more rechecks required on p 532549, and if you have any cycle to review p 531486 that would be great | 18:48 |
| patchbot | https://review.openstack.org/#/c/532549/ - swift (feature/deep) - Test that all object methods get shard container u... | 18:48 |
| patchbot | https://review.openstack.org/#/c/531486/ - swift (feature/deep) - Add copy method to ShardRange | 18:48 |
| *** pdardeau has joined #openstack-swift | 19:02 | |
| *** links has quit IRC | 19:07 | |
| *** itlinux has joined #openstack-swift | 19:20 | |
| clayg | man, there is a ton of good patches to look at - nice work @timburke @acoles @torgomatic | 19:38 |
| *** bigdogstl has joined #openstack-swift | 19:38 | |
| timburke | clayg: i'm debating about adding p 496534 and p 534399 to priority reviews... i feel like torgomatic's p 532383 is *so much better* with those | 19:39 |
| patchbot | https://review.openstack.org/#/c/496534/ - swift - expirer: unexpected responses don't warrant traceb... | 19:39 |
| patchbot | https://review.openstack.org/#/c/534399/ - swift - expirer: don't retry 404, 409, or 412 failures | 19:39 |
| patchbot | https://review.openstack.org/#/c/532383/ - swift - Don't make async_pendings during object expiration | 19:39 |
| openstackgerrit | Merged openstack/swift feature/deep: Test that all object methods get shard container update headers https://review.openstack.org/532549 | 19:43 |
| timburke | whooo! | 19:44 |
| clayg | reading the comments on getting rid of asyncs from the expirier - i'm curious what can be done about the retry unsuccessfully until the end of a recalim | 19:44 |
| timburke | yeah, i think the two i linked make it tolerable, but it'd be even better to take 409 as success, then have the object-server start returning *that* if you try to delete with the same x-timestamp as the ts it already has | 19:48 |
| *** bigdogstl has quit IRC | 19:48 | |
| timburke | ..and i should look at how timestamps & 409s interact with x-if-delete-ats and 412s... | 19:49 |
| *** itlinux has quit IRC | 19:50 | |
| *** pcaruana has quit IRC | 19:50 | |
| *** mvk has joined #openstack-swift | 19:59 | |
| openstackgerrit | Samuel Merritt proposed openstack/swift master: Minor cleanup in monitoring doc. https://review.openstack.org/534426 | 20:02 |
| openstackgerrit | Samuel Merritt proposed openstack/swift master: Improve object-updater's stats logging https://review.openstack.org/533207 | 20:03 |
| *** itlinux has joined #openstack-swift | 20:05 | |
| *** bigdogstl has joined #openstack-swift | 20:09 | |
| *** bigdogstl has quit IRC | 20:14 | |
| *** SkyRocknRoll has quit IRC | 20:16 | |
| *** mvk has quit IRC | 20:19 | |
| *** bigdogstl has joined #openstack-swift | 20:28 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: expirer: unexpected responses don't warrant tracebacks https://review.openstack.org/496534 | 20:30 |
| timburke | clayg: yeah, i'm really tempted to reference https://github.com/openstack/swift/blob/2.16.0/swift/common/utils.py#L1636-L1651 in internal_client -- i erred on the side of being pretty conservative in the change, though maybe i should re-think that | 20:32 |
| clayg | I mean... if we don't want to do the work to audit everywhere we use internal client we could make plumb it to be optional on instances of the client of even per-request | 20:36 |
| clayg | but I think probably we should just do the work... retry on 4XX is non-sense - which is really what this change is about | 20:36 |
| clayg | fix it instead of work around it is a preferred option when the stupid is in YOUR code base | 20:36 |
| *** ukaynar has quit IRC | 20:37 | |
| *** supamatt has quit IRC | 20:37 | |
| clayg | you can lean on the reviewers too try and audit for you as well - it does make it a little more scary change - but hopefully we'll do a good job | 20:37 |
| *** supamatt has joined #openstack-swift | 20:37 | |
| *** ukaynar has joined #openstack-swift | 20:38 | |
| *** bigdogstl has quit IRC | 20:38 | |
| *** silor has joined #openstack-swift | 20:49 | |
| *** silor has quit IRC | 21:08 | |
| *** bigdogstl has joined #openstack-swift | 21:15 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: internal_client: Don't retry when we expect the same reponse https://review.openstack.org/534399 | 21:16 |
| *** threestrands_ has joined #openstack-swift | 21:18 | |
| timburke | oh, yeah! and there's p 530175! so many great patches we ought to land! | 21:34 |
| patchbot | https://review.openstack.org/#/c/530175/ - swift - Fix InternalClient to drain response body if the r... | 21:34 |
| *** HW-Peter has quit IRC | 21:38 | |
| *** ukaynar has quit IRC | 21:43 | |
| *** bigdogstl has quit IRC | 21:45 | |
| *** bigdogstl has joined #openstack-swift | 21:51 | |
| clayg | mattoliverau: looking at patch 532696 made me think it's obvious this code should be extracted under /cli | 21:52 |
| patchbot | https://review.openstack.org/#/c/532696/ - swift - Let recon-cron work with conf.d | 21:52 |
| clayg | I'd like to file a wishlist bug for that? that's a thing right? | 21:52 |
| clayg | meh, i did it lp bug # | 21:54 |
| clayg | lp bug # | 21:54 |
| clayg | O. M. G. | 21:54 |
| clayg | lp bug #1743656 | 21:55 |
| openstack | Launchpad bug 1743656 in OpenStack Object Storage (swift) "bin/swift-recon-cron is not tested" [Wishlist,New] https://launchpad.net/bugs/1743656 | 21:55 |
| *** itlinux has quit IRC | 21:58 | |
| *** bigdogstl has quit IRC | 22:03 | |
| *** bigdogstl has joined #openstack-swift | 22:05 | |
| openstackgerrit | Tim Burke proposed openstack/swift feature/deep: Define entries of shard range db row in one place https://review.openstack.org/531487 | 22:21 |
| openstackgerrit | Tim Burke proposed openstack/swift feature/deep: annotate_failure: preserve tracebacks https://review.openstack.org/534454 | 22:21 |
| *** bigdogstl has quit IRC | 22:22 | |
| *** rcernin has joined #openstack-swift | 22:31 | |
| *** itlinux has joined #openstack-swift | 22:39 | |
| *** itlinux has quit IRC | 22:41 | |
| *** itlinux has joined #openstack-swift | 22:42 | |
| *** itlinux has quit IRC | 22:47 | |
| mattoliverau | clayg: yeah, I thought the same thing when reviewing. and yes we can tag things as wishlist and maybe even low hanging fruit | 22:56 |
| *** fishbone_ has joined #openstack-swift | 22:57 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: Fix InternalClient to drain response body if the request fails https://review.openstack.org/530175 | 22:57 |
| clayg | kota_: ^ I'm so sorry to push over, I hope it is helpful for you! | 22:57 |
| clayg | timburke: torgomatic: am I on the right track with patch 530175 | 22:57 |
| patchbot | https://review.openstack.org/#/c/530175/ - swift - Fix InternalClient to drain response body if the r... | 22:57 |
| *** bigdogstl has joined #openstack-swift | 23:00 | |
| torgomatic | clayg: I think you're headed in the right direction, but you need to also close resp.app_iter if it is a 2xx and you're not going to return it | 23:04 |
| torgomatic | so you've got "if resp.status_int // 100 != 2:" and that branch closes the iter, but I think you also need "else: close_if_possible(resp.app_iter)" | 23:04 |
| torgomatic | otherwise you get memory and socket leaks | 23:04 |
| clayg | timburke: patch 533028 is insane... is nose really happy to ignore the bash and javascript under test/functional? Is it really a blackbox functional test - you can point it any swift endpoint? | 23:05 |
| patchbot | https://review.openstack.org/#/c/533028/ - swift - Add some functional CORS tests | 23:05 |
| timburke | does it end in .py? :P | 23:05 |
| clayg | timburke: flake8 has figured out some way to check scripts in bin that don't end in .py - I didn't mean for the question to be rhetorical - i really do have reservations about bucking convention and trusting our tooling not to get confused at some point | 23:07 |
| timburke | with appropriate tweaks around https://review.openstack.org/#/c/533028/2/test/functional/cors/run.sh@5 and https://review.openstack.org/#/c/533028/2/test/functional/cors/harness.js@3 , yeah, i totally expect that should work against an arbitrary swift endpoint | 23:07 |
| patchbot | patch 533028 - swift - Add some functional CORS tests | 23:07 |
| patchbot | patch 533028 - swift - Add some functional CORS tests | 23:07 |
| timburke | if you want to make that respect test.conf, more power to you :-) | 23:07 |
| clayg | torgomatic: I think i get what your saying... let me try again | 23:07 |
| *** bigdogstl has quit IRC | 23:08 | |
| clayg | timburke: I think you might just should have pushed it up earlier and tried to get some feedback on a PoC - maybe you did that and I missed it - for me it's gunna take a significant investment to get enough context to understand the value relative to the cost... | 23:11 |
| clayg | timburke: I wonder if maybe it's something we develop stand-alone and consider bringing under community maintenance after the fact as it proves out it's usefulness? | 23:12 |
| clayg | idk, I think i'm mostly just dealing with the shock of how expensive it is to maintain this feature that isn't well used/understood | 23:13 |
| clayg | Kudos for diving in with both feet! I'm sure swift's CORS support will only get better from here on out! | 23:14 |
| timburke | sure? w/e -- i just want something more than https://github.com/openstack/swift/blob/2.16.0/test/functional/test_object.py#L1482-L1586 | 23:16 |
| *** bigdogstl has joined #openstack-swift | 23:16 | |
| timburke | the experiment is *already* revealing bugs / gaps, hence https://review.openstack.org/#/c/533027/1 https://review.openstack.org/#/c/533029/2 https://review.openstack.org/#/c/533030/2 https://review.openstack.org/#/c/533321/1 | 23:17 |
| patchbot | patch 533027 - swift - Fix up some Content-Type handling in account/conta... | 23:17 |
| patchbot | patch 533029 - swift - /info: allow more CORS requests | 23:17 |
| patchbot | patch 533030 - swift - Expose a few more headers in CORS responses | 23:17 |
| patchbot | patch 533321 - swift - slo: Send ETag header in 206 responses | 23:17 |
| timburke | the decision as i see it is either: drop support for this feature we mostly ignore, or add sufficient testing that we actually have confidence that the feature works | 23:20 |
| notmyname | timburke: if the second option is a process (eg we add testing over time and keep improving), then yes, I agree with you | 23:21 |
| *** bigdogstl has quit IRC | 23:22 | |
| timburke | it needs to be. otherwise, we continue being in this rut of kinda-but-not-really supporting CORS | 23:22 |
| notmyname | I just mean that it's not a "drop support" vs "add 100% testing coverage" dichotomy | 23:23 |
| timburke | it's a spectrum, sure. but there's currently next to nothing keeping us from drifting further and further toward the "drop support" end | 23:24 |
| timburke | and trust me, what i've got so far is *nowhere near* 100% coverage :-) | 23:24 |
| notmyname | well, that's why "Tim will save us" with the patches to add some functional testing! | 23:24 |
| timburke | part of all of this will be making sure that (1) people exercise the func tests manually until we have it in a gate job and (2) people think, "hey, can/should i add a CORS func test for this new feature?" | 23:26 |
| timburke | so i'll be super interested in having people try out my starting point, maybe even add to it, and offer suggestions for improvements :D | 23:27 |
| *** kei_yama has joined #openstack-swift | 23:29 | |
| *** bigdogstl has joined #openstack-swift | 23:33 | |
| *** chsc has quit IRC | 23:35 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: Fix InternalClient to drain response body if the request fails https://review.openstack.org/530175 | 23:44 |
| *** bigdogstl has quit IRC | 23:45 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!