15:29:20 #startmeeting Pulp Triage 2016-12-16 15:29:20 #info bizhang has joined triage 15:29:20 Meeting started Fri Dec 16 15:29:20 2016 UTC and is due to finish in 60 minutes. The chair is bizhang. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:29:20 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:29:20 The meeting name has been set to 'pulp_triage_2016_12_16' 15:29:21 bizhang has joined triage 15:29:21 f-in unfortunate to find that out when I do the big migration in a few hours 15:29:30 !here 15:29:30 #info mhrivnak has joined triage 15:29:31 mhrivnak has joined triage 15:29:35 !here 15:29:35 #info bmbouter has joined triage 15:29:35 bmbouter has joined triage 15:29:40 !here 15:29:40 #info dkliban has joined triage 15:29:40 dkliban has joined triage 15:29:49 !next 15:29:50 5 issues left to triage: 2437, 2491, 2492, 2493, 2494 15:29:50 #topic Upgrade from 2.8.0 to 2.10.2: Post upgrade publish creates invalid links for pre-upgrade files - http://pulp.plan.io/issues/2437 15:29:51 Pulp Issue #2437 [NEW] (unassigned) - Priority: Normal | Severity: High 15:29:52 Upgrade from 2.8.0 to 2.10.2: Post upgrade publish creates invalid links for pre-upgrade files - http://pulp.plan.io/issues/2437 15:29:59 !here 15:29:59 #info pcreech has joined triage 15:30:00 pcreech has joined triage 15:30:31 mark as triaged 15:30:40 jortel is already helping the user figure it out 15:30:57 #info daviddavis has joined triage 15:30:57 !here 15:30:57 daviddavis has joined triage 15:31:09 !here 15:31:09 #info ttereshc has joined triage 15:31:10 ttereshc has joined triage 15:31:17 !propose accept 15:31:17 #idea Proposed for #2437: Leave the issue as-is, accepting its current state. 15:31:18 Proposed for #2437: Leave the issue as-is, accepting its current state. 15:31:24 +1 15:31:29 +1 15:32:04 !accept 15:32:04 #agreed Leave the issue as-is, accepting its current state. 15:32:05 Current proposal accepted: Leave the issue as-is, accepting its current state. 15:32:05 #topic When stopping pulp_workers, pulp_celerybeat, and pulp_resource_manager gracefully, the status API still shows them as running - http://pulp.plan.io/issues/2491 15:32:06 4 issues left to triage: 2491, 2492, 2493, 2494 15:32:07 Pulp Issue #2491 [NEW] (unassigned) - Priority: Normal | Severity: Medium 15:32:08 When stopping pulp_workers, pulp_celerybeat, and pulp_resource_manager gracefully, the status API still shows them as running - http://pulp.plan.io/issues/2491 15:32:24 This is very unfortunate. 15:32:28 indeed 15:32:31 !propose triage high high 15:32:31 #idea Proposed for #2491: Priority: High, Severity: High 15:32:31 Proposed for #2491: Priority: High, Severity: High 15:32:37 +1 15:32:39 also we should put this on a sprint 15:32:47 agreed. 15:32:48 s/a /the / 15:33:02 !propose other put on sprint 13 15:33:02 #idea Proposed for #2491: put on sprint 13 15:33:03 Proposed for #2491: put on sprint 13 15:33:19 +1 15:33:30 #agreed put on sprint 13 15:33:30 !accept 15:33:30 Current proposal accepted: put on sprint 13 15:33:32 3 issues left to triage: 2492, 2493, 2494 15:33:32 #topic Small memory leak in httpd when dispatching Pulp tasks - http://pulp.plan.io/issues/2492 15:33:33 Pulp Issue #2492 [NEW] (unassigned) - Priority: Normal | Severity: Low 15:33:34 Small memory leak in httpd when dispatching Pulp tasks - http://pulp.plan.io/issues/2492 15:34:34 I propose accept and it has a BZ attached 15:34:42 I'm not sure if you want this on the sprint or not 15:34:54 +1 add to sprint 15:34:57 +1 15:35:01 !propose accept 15:35:01 #idea Proposed for #2492: Leave the issue as-is, accepting its current state. 15:35:02 Proposed for #2492: Leave the issue as-is, accepting its current state. 15:35:07 !propose other add to sprint 15:35:07 #idea Proposed for #2492: add to sprint 15:35:08 Proposed for #2492: add to sprint 15:35:14 !accept 15:35:14 #agreed add to sprint 15:35:14 Current proposal accepted: add to sprint 15:35:15 2 issues left to triage: 2493, 2494 15:35:16 #topic http://pulpproject.org/docs/ missing links to 2.10 and 2.11 stable docs - http://pulp.plan.io/issues/2493 15:35:16 Pulp Issue #2493 [NEW] (unassigned) - Priority: Normal | Severity: Medium 15:35:17 http://pulpproject.org/docs/ missing links to 2.10 and 2.11 stable docs - http://pulp.plan.io/issues/2493 15:35:57 maybe this is not important but we use the github issue tracker for pulpproject.org issues 15:36:07 https://github.com/pulp/pulpproject.org/issues/ 15:36:08 Title: Issues · pulp/pulpproject.org · GitHub (at github.com) 15:36:19 bmbouter: i thought about this when filing, but since this is about docs i put it here so we would triage it 15:36:32 !here 15:36:32 #info misa_ has joined triage 15:36:33 misa_ has joined triage 15:36:43 dkliban: that makes sense 15:36:52 !propose accept 15:36:52 #idea Proposed for #2493: Leave the issue as-is, accepting its current state. 15:36:53 Proposed for #2493: Leave the issue as-is, accepting its current state. 15:37:11 +1 15:37:45 +1 15:38:18 !accept 15:38:18 #agreed Leave the issue as-is, accepting its current state. 15:38:18 Current proposal accepted: Leave the issue as-is, accepting its current state. 15:38:19 1 issues left to triage: 2494 15:38:20 #topic rename of id to package_group_id did not also populate a 'package_group_id' field in a unit's metadata section - http://pulp.plan.io/issues/2494 15:38:20 Pulp Issue #2494 [NEW] (unassigned) - Priority: Normal | Severity: Medium 15:38:21 rename of id to package_group_id did not also populate a 'package_group_id' field in a unit's metadata section - http://pulp.plan.io/issues/2494 15:40:05 context: I thought all fields listed as unit key in the api would be returned by the api when fetching a single unit, and that is no longer true in 2.10 15:40:05 or 2.8+ for that matter 15:40:08 !here 15:40:08 #info jortel has joined triage 15:40:08 jortel has joined triage 15:41:21 misa_, when we moved modelling to use mongoengine, we had to rename some "id" fields as you see here, but did our best to add serializers that "fix" them on the way in and out of the API. 15:41:31 So the previous behavior would be preserved, and the API user wouldn't be impacted. 15:41:42 yes, you did that :-) 15:41:53 I think you've just found a place in the API we missed. 15:41:57 Does that sound right? 15:42:01 Or is there more to it? 15:42:15 I *think* you achieved your goal 15:42:26 My point is that you now broke what I thought was an invariant 15:42:56 package group units still have an 'id', to preserve backwards compat. That's great 15:43:18 They should also have had a package_group_id (identical) because you claim there is a unit key field named package_group_id 15:44:28 Gotcha. 15:44:52 So it seems like this is a consistency issue we need to fix. Anyone else have thoughts? 15:45:58 * mhrivnak is juggling a separate IRC discussion and could use help getting this one wrapped up 15:46:05 !propose accept 15:46:05 #idea Proposed for #2494: Leave the issue as-is, accepting its current state. 15:46:06 Proposed for #2494: Leave the issue as-is, accepting its current state. 15:46:09 it seems like a valid point to me to expect unit key fields in response 15:46:16 all of them 15:47:23 ichimonji10++ 15:47:23 ichimonji10's karma is now 81 15:47:30 :) 15:48:00 If no one has anymore suggestions I will accept this issue as-is, since I think we do need to fix it 15:48:06 +1 15:48:10 +1 for accept 15:48:30 !accept 15:48:30 #agreed Leave the issue as-is, accepting its current state. 15:48:30 Current proposal accepted: Leave the issue as-is, accepting its current state. 15:48:32 No issues to triage. 15:48:34 !end 15:48:34 #endmeeting