14:30:22 #startmeeting Pulp Triage 2016-05-27 14:30:22 #info bmbouter has joined triage 14:30:23 Meeting started Fri May 27 14:30:22 2016 UTC and is due to finish in 60 minutes. The chair is bmbouter. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:30:23 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:30:23 The meeting name has been set to 'pulp_triage_2016_05_27' 14:30:23 bmbouter has joined triage 14:31:16 !here 14:31:16 #info preethi has joined triage 14:31:16 preethi has joined triage 14:31:19 !here 14:31:19 #info ttereshc has joined triage 14:31:20 ttereshc has joined triage 14:31:32 !here 14:31:32 #info ipanova has joined triage 14:31:33 ipanova has joined triage 14:32:17 we have a quorum so we'll begin 14:32:26 we're going to pick a few high prio issues first 14:32:49 !issue 1950 14:32:49 #topic module upload fails with IOError: [Errno 2] No such file or directory - http://pulp.plan.io/issues/1950 14:32:50 Puppet Support Issue #1950 [ASSIGNED] (amacdona@redhat.com) - Priority: Normal | Severity: High 14:32:51 module upload fails with IOError: [Errno 2] No such file or directory - http://pulp.plan.io/issues/1950 14:33:12 dkliban had suggested this be Urgent 14:33:30 smyers: will an urgent bug block the 2.8.4 release also? 14:35:09 well it's important so we should probably make it urgent for now 14:35:13 If it's urgent with a target release of 2.8.4, then it will prompt a meeting to dig deeper and decide if it should block the release, as I understand it. 14:35:23 If can be urgent and not pinned to 2.8.4 14:35:36 I was going to put it on 2.8.5 w/ urgent 14:36:26 bmbouter: that might related to the bug i was working on w/r to puppet-module's name 14:36:30 I agree 14:37:00 ipanova: I think urgent / high 14:37:06 agree 14:37:10 it would be great if you could look at this 14:37:14 to see if its related 14:37:20 #idea Proposed for #1950: Priority: Urgent, Severity: High, Target Platform Release: 2.8.5 14:37:20 !propose triage u h 2.8.5 14:37:20 #info smyers has joined triage 14:37:21 smyers has joined triage 14:37:22 Proposed for #1950: Priority: Urgent, Severity: High, Target Platform Release: 2.8.5 14:37:42 !accept 14:37:42 #agreed Priority: Urgent, Severity: High, Target Platform Release: 2.8.5 14:37:42 Current proposal accepted: Priority: Urgent, Severity: High, Target Platform Release: 2.8.5 14:37:43 12 issues left to triage: 1932, 1933, 1934, 1936, 1937, 1938, 1943, 1945, 1947, 1948, 1949, 1952 14:37:44 #topic KeyError: '\xff\xff\xff\xff\xd8\xec2\x05' - http://pulp.plan.io/issues/1932 14:37:44 RPM Support Issue #1932 [NEW] (unassigned) - Priority: Normal | Severity: High 14:37:45 KeyError: '\xff\xff\xff\xff\xd8\xec2\x05' - http://pulp.plan.io/issues/1932 14:39:32 The going theory, iirc, is that this is a CDN error. As always, pulp could probably have better error handling, but... 14:39:36 !propose needinfo 14:39:36 #idea Proposed for #1932: This issue cannot be triaged without more info. 14:39:37 Proposed for #1932: This issue cannot be triaged without more info. 14:39:48 ...because we don't know if it's a CDN issue or not yet. 14:39:59 I agree, should I write on the issue? 14:40:03 and we move on? 14:40:07 I can do that. 14:40:13 ok great, we'll move on 14:40:28 ichimonji10: thanks for the PRs you submitted yesterday 14:40:30 ichimonji10++ 14:40:30 ichimonji10's karma is now 7 14:40:32 !issue 1945 14:40:33 #topic pulp 2.8.3 migration fails on unit 0028 - http://pulp.plan.io/issues/1945 14:40:33 Pulp Issue #1945 [POST] (jortel@redhat.com) - Priority: Urgent | Severity: High 14:40:34 pulp 2.8.3 migration fails on unit 0028 - http://pulp.plan.io/issues/1945 14:40:40 this is urgent and already at POST 14:40:57 elyezer: Thank you. :) 14:41:04 thank *you* 14:41:07 #idea Proposed for #1945: Priority: Urgent, Severity: High 14:41:07 !propose triage u h 14:41:07 Proposed for #1945: Priority: Urgent, Severity: High 14:41:20 elyezer: I triaged the Pulp 2.9 test failure with ipanova. That should be taken care of as of the next nightly. 14:41:46 bmbouter, +1 14:42:04 ichimonji10: great! 14:42:10 ipanova++ 14:42:10 ipanova's karma is now 6 14:42:13 ok last call for 1945? 14:42:26 #agreed Priority: Urgent, Severity: High 14:42:26 !accept 14:42:26 Current proposal accepted: Priority: Urgent, Severity: High 14:42:28 #topic KeyError: '\xff\xff\xff\xff\xd8\xec2\x05' - http://pulp.plan.io/issues/1932 14:42:28 11 issues left to triage: 1932, 1933, 1934, 1936, 1937, 1938, 1943, 1947, 1948, 1949, 1952 14:42:29 RPM Support Issue #1932 [NEW] (unassigned) - Priority: Normal | Severity: High 14:42:31 KeyError: '\xff\xff\xff\xff\xd8\xec2\x05' - http://pulp.plan.io/issues/1932 14:42:40 !next 14:42:42 #topic Restricting fields on user search does not work as expected - http://pulp.plan.io/issues/1933 14:42:42 10 issues left to triage: 1933, 1934, 1936, 1937, 1938, 1943, 1947, 1948, 1949, 1952 14:42:43 Pulp Issue #1933 [NEW] (unassigned) - Priority: Normal | Severity: Low 14:42:44 Restricting fields on user search does not work as expected - http://pulp.plan.io/issues/1933 14:43:58 * smyers writes an issue since #1932 shouldn't have reappeared 14:44:55 I think triaging 1933 as is would be good 14:45:05 normal/low 14:45:24 !propose triage n l 14:45:24 #idea Proposed for #1933: Priority: Normal, Severity: Low 14:45:24 Proposed for #1933: Priority: Normal, Severity: Low 14:46:22 +1 14:46:28 !accept 14:46:28 #agreed Priority: Normal, Severity: Low 14:46:29 Current proposal accepted: Priority: Normal, Severity: Low 14:46:29 #topic OSTree syncs are broken - http://pulp.plan.io/issues/1934 14:46:30 9 issues left to triage: 1934, 1936, 1937, 1938, 1943, 1947, 1948, 1949, 1952 14:46:31 OSTree Support Issue #1934 [NEW] (unassigned) - Priority: Normal | Severity: Medium 14:46:32 OSTree syncs are broken - http://pulp.plan.io/issues/1934 14:46:44 smyers, #1932 proposal was not accepted, !next was used 14:47:24 1932 was seen before and should not have come up a second time 14:48:06 this looks serious 14:48:29 high severity if you can't sync OSTrees 14:48:43 and we should fix it either as High or Urgent 14:49:07 Is it a regression in 2.8 or 2.8.3? 14:49:19 (I'm assuming that we could previously sync ostree shiz) 14:49:54 AIUI, yes we could 14:49:56 so yes 14:49:56 smyers: we could 14:50:28 I think urgent and 2.8.5 14:50:44 +1 14:50:47 !propose u h 2.8.5 14:50:47 Error: "propose" is not a valid command. 14:50:57 #idea Proposed for #1934: Priority: Urgent, Severity: High, Target Platform Release: 2.8.5 14:50:57 !propose triage u h 2.8.5 14:50:57 Proposed for #1934: Priority: Urgent, Severity: High, Target Platform Release: 2.8.5 14:51:01 I'll fix that, I promise :) 14:51:05 :)) 14:52:08 +1 we should stop breaking stuff :( 14:52:25 !accept 14:52:25 #agreed Priority: Urgent, Severity: High, Target Platform Release: 2.8.5 14:52:25 Current proposal accepted: Priority: Urgent, Severity: High, Target Platform Release: 2.8.5 14:52:27 8 issues left to triage: 1936, 1937, 1938, 1943, 1947, 1948, 1949, 1952 14:52:28 #topic as user, I can get unit_id from task_result of import action - http://pulp.plan.io/issues/1936 14:52:28 Pulp Issue #1936 [NEW] (unassigned) - Priority: Normal | Severity: Medium 14:52:29 as user, I can get unit_id from task_result of import action - http://pulp.plan.io/issues/1936 14:53:02 Story. 14:53:15 !propose other convert to story, do not triage 14:53:15 #idea Proposed for #1936: convert to story, do not triage 14:53:16 Proposed for #1936: convert to story, do not triage 14:53:39 agreed 14:53:43 !accept 14:53:43 #agreed convert to story, do not triage 14:53:43 Current proposal accepted: convert to story, do not triage 14:53:44 7 issues left to triage: 1937, 1938, 1943, 1947, 1948, 1949, 1952 14:53:44 #topic Syncing a puppet module with the same content as a different repo results in no content - http://pulp.plan.io/issues/1937 14:53:45 Pulp Issue #1937 [POST] (jcline@redhat.com) - Priority: Normal | Severity: Medium 14:53:46 Syncing a puppet module with the same content as a different repo results in no content - http://pulp.plan.io/issues/1937 14:54:11 POST 14:54:14 I forgot to make that one triaged since it's in post 14:54:16 sorry 14:54:26 !propose triage 14:54:27 (propose triage [target_release]) -- Propose triage values including priority, severity, and an optional target release. 14:54:34 !propose accept 14:54:34 #idea Proposed for #1937: Leave the issue as-is, accepting its current state. 14:54:35 Proposed for #1937: Leave the issue as-is, accepting its current state. 14:54:48 (which will soon become propose traige) :) 14:55:23 niiice 14:55:26 +1 14:55:30 And, to be clear, that's accepting its current state *in Redmine*, which really means "Just tick the triaged box" 14:55:38 yes 14:55:48 ok next 14:55:51 !next 14:55:52 6 issues left to triage: 1938, 1943, 1947, 1948, 1949, 1952 14:55:52 #topic force_full is not supported by export_distributor - http://pulp.plan.io/issues/1938 14:55:53 RPM Support Issue #1938 [NEW] (unassigned) - Priority: Normal | Severity: Medium 14:55:54 force_full is not supported by export_distributor - http://pulp.plan.io/issues/1938 14:56:04 hopefully that was the right thing for me to do 14:56:09 !here 14:56:09 #info jcline has joined triage 14:56:09 jcline has joined triage 14:57:05 I think this should be High priority 14:57:15 and not on any specific release 14:57:20 #idea Proposed for #1938: Priority: High, Severity: High 14:57:20 !propose triage h h 14:57:20 Proposed for #1938: Priority: High, Severity: High 14:57:31 this should be fixed before 2.9 release since force_full is a new option 14:57:31 that's good w/ me. last call 14:57:39 ttereshc, +1 14:57:42 ok let's put it on 2.9 14:57:48 is pulp-python available on Pulp 2.7? 14:57:50 !propose triage h h 2.9 14:57:50 #idea Proposed for #1938: Priority: High, Severity: High, Target Platform Release: 2.9 14:57:52 Proposed for #1938: Priority: High, Severity: High, Target Platform Release: 2.9 14:57:57 +1 14:58:01 +1 14:58:03 #agreed Priority: High, Severity: High, Target Platform Release: 2.9 14:58:03 !accept 14:58:03 Current proposal accepted: Priority: High, Severity: High, Target Platform Release: 2.9 14:58:04 5 issues left to triage: 1943, 1947, 1948, 1949, 1952 14:58:04 #topic forge importer downloads modules that already exist in Pulp - http://pulp.plan.io/issues/1943 14:58:05 Puppet Support Issue #1943 [NEW] (unassigned) - Priority: Normal | Severity: Medium 14:58:06 forge importer downloads modules that already exist in Pulp - http://pulp.plan.io/issues/1943 14:59:08 Sounds...annoying, but not breaking anything? 14:59:13 Yup 14:59:17 !propose accept 14:59:17 #idea Proposed for #1943: Leave the issue as-is, accepting its current state. 14:59:18 Proposed for #1943: Leave the issue as-is, accepting its current state. 14:59:23 +1 14:59:23 #agreed Leave the issue as-is, accepting its current state. 14:59:23 !accept 14:59:23 Current proposal accepted: Leave the issue as-is, accepting its current state. 14:59:24 4 issues left to triage: 1947, 1948, 1949, 1952 14:59:25 #topic Concurrent sync of repos is running slow - http://pulp.plan.io/issues/1947 14:59:25 Pulp Issue #1947 [NEW] (unassigned) - Priority: Normal | Severity: Medium 14:59:26 Concurrent sync of repos is running slow - http://pulp.plan.io/issues/1947 15:00:15 this issue is still under investigation by the reporter 15:00:36 I think deferring triage with a note that a more clear reproducer should be put onto it would be the best 15:01:09 My bet is it's the errata thing. I can sit down with him and get the tasks profiled, though. 15:01:23 that would be great 15:01:29 !propose other Need a reproducer to triage this issue, and this is possible a duplicate of syncs being slow due to errata NEVRA filtering. 15:01:29 #idea Proposed for #1947: Need a reproducer to triage this issue, and this is possible a duplicate of syncs being slow due to errata NEVRA filtering. 15:01:30 Proposed for #1947: Need a reproducer to triage this issue, and this is possible a duplicate of syncs being slow due to errata NEVRA filtering. 15:01:42 he is not sure which sync is the slow one though so there is still more info needed 15:02:00 It's probably any of those repos that have errata, and thus all of them. :) 15:02:10 s/:)/:(/ 15:02:21 there was also question about it being the initial or resync 15:02:28 All those repos will have errata so 15:02:39 anyway 15:02:41 #agreed Need a reproducer to triage this issue, and this is possible a duplicate of syncs being slow due to errata NEVRA filtering. 15:02:41 !accept 15:02:41 Current proposal accepted: Need a reproducer to triage this issue, and this is possible a duplicate of syncs being slow due to errata NEVRA filtering. 15:02:43 3 issues left to triage: 1948, 1949, 1952 15:02:43 #topic Upgrading RPMs on EL6 sometimes fails during pre script - http://pulp.plan.io/issues/1948 15:02:44 Pulp Issue #1948 [NEW] (unassigned) - Priority: Normal | Severity: Medium 15:02:45 Upgrading RPMs on EL6 sometimes fails during pre script - http://pulp.plan.io/issues/1948 15:03:16 This is specific to el6 and how we stop the service during the upgrade scriptlet 15:03:38 jcline, can you add more info to the issue to make it triagable? 15:03:51 (triagable is a word, right? :P) 15:03:53 yeah 15:03:53 jcline, that's correct, we hit that on our rcm-pulp 15:03:57 #idea Proposed for #1948: Defer this issue until later in triage. 15:03:57 !propose defer 15:03:57 elyezer: yeah pulp-python is available on 2.7 15:03:57 Proposed for #1948: Defer this issue until later in triage. 15:04:25 jcline, Can you add that info in the next minute or so? :D 15:04:38 * jcline types furiously 15:04:45 lol 15:05:47 I don't think we've ever deferred before. I remain optimistic that it will work if !accept is called. 15:06:56 Done 15:07:43 bmbouter is being helpful in #pulp :) 15:07:48 jcline, I think problem is that prescript tries to stop services and if they are not running it fails 15:07:55 Mind you, I just saw this mentioned a few weeks ago and this is an off-the-cuff analysis 15:08:13 jluza, cool, that was my hypothesis :) 15:08:27 that's pretty funny 15:08:30 we should make this higher priority 15:08:37 I know the upgrade stuff katello uses stops everything prior to upgrade 15:08:55 "Hey, services, I need you to be stopped." "We're already stopped, bro!" "UNACCEPTABLE." 15:09:07 Yeah, haven't scriptlets fail is bad and scary, and it's an easy fix 15:09:12 like high / medium 15:09:27 !propose triage h m 15:09:27 #idea Proposed for #1948: Priority: High, Severity: Medium 15:09:28 Proposed for #1948: Priority: High, Severity: Medium 15:09:41 Yeah, with the added background, +1 to that 15:10:15 ok accepting 15:10:17 !accept 15:10:17 #agreed Priority: High, Severity: Medium 15:10:18 Current proposal accepted: Priority: High, Severity: Medium 15:10:18 #topic re-publish takes longer than expected - http://pulp.plan.io/issues/1949 15:10:19 2 issues left to triage: 1949, 1952 15:10:20 RPM Support Issue #1949 [NEW] (unassigned) - Priority: High | Severity: High 15:10:21 re-publish takes longer than expected - http://pulp.plan.io/issues/1949 15:10:47 This is the same thing as #1947 15:11:40 ...probably 15:12:05 True, I guess we should confirm #1947 is the same thing first 15:12:17 I agree, let's keep them both open and relate them 15:12:26 jcline can you comment on these things so we know 15:13:04 I'm not sure what to do for severity here. It's a performance regression that comes with actually having data integrity in errata publishes. 15:13:17 Can't have one without the other unless we refactor errata (imo) 15:13:34 Yeah. My impression is that the fix is remodel things. Is that correct? 15:13:47 I'm not sure but we should triage it 15:14:44 I agree that we should triage it, but I'm just not sure what's appropriate. Like 2.9 is getting pretty solid, so this is probably pretty important for something like 2.10 15:14:58 (for example) 15:15:25 I think katello is going to want it in the 2.8 line 15:15:34 that may not be possible 15:16:34 Yeah. So I have no idea how to triage it. 15:16:38 It's High/High, so... 15:16:41 #idea Proposed for #1949: Leave the issue as-is, accepting its current state. 15:16:41 !propose accept 15:16:41 Proposed for #1949: Leave the issue as-is, accepting its current state. 15:16:43 ? 15:16:57 !accept 15:16:57 #agreed Leave the issue as-is, accepting its current state. 15:16:57 Current proposal accepted: Leave the issue as-is, accepting its current state. 15:16:59 #topic Repo sync fails after upgrade to 2.8.3. - http://pulp.plan.io/issues/1952 15:16:59 1 issues left to triage: 1952 15:17:00 Pulp Issue #1952 [NEW] (unassigned) - Priority: Normal | Severity: Medium 15:17:01 Repo sync fails after upgrade to 2.8.3. - http://pulp.plan.io/issues/1952 15:17:38 This looks familiar to me 15:18:32 this is another data quality problem with the 2.8 strong validation 15:18:42 "The field 'filelist' does not exist on the document 'RPM'" 15:18:51 Is 2.6 to 2.8 supported, or is the upgrade path 2.6 -> 2.7 -> 2.8? Sounds like some migrations got skipped or something. 15:19:02 2.6 to 2.8 is supported 15:19:20 IIRC, it goes as far back as 2.4 -> 2.8 15:19:28 but not 2.3.z -> 2.8 15:19:33 you have to go through 2.4 15:19:44 jcline, similar to this one? https://pulp.plan.io/issues/1754 15:19:46 Title: Issue #1754: Repo sync with fails on an upgraded pulp server - Pulp (at pulp.plan.io) 15:20:36 I propose urgent / high and 2.8.5 15:20:41 No, I remember something specifically about filelist on the RPM model 15:21:21 Well, I don't actually remember. That's the problem. 15:21:23 ttereshc: looks similar 15:21:27 filelist isn't used in the model definition 15:21:29 https://github.com/pulp/pulp_rpm/blob/master/plugins/pulp_rpm/plugins/db/models.py 15:21:30 Title: pulp_rpm/models.py at master · pulp/pulp_rpm · GitHub (at github.com) 15:21:44 bmbouter ++ 15:21:48 Yeah, I think it's fair to treat this as a regression, since we're complaining about the DB we made. 15:22:03 it's similar to this: https://pulp.plan.io/issues/1931 15:22:04 bmbouter, yeah, but at some point it seems it was and mongoengine doesn't like the extra field 15:22:05 Title: Issue #1931: FieldDoesNotExist: The field '_rpm_references' does not exist on the document 'Errata' - RPM Support - Pulp (at pulp.plan.io) 15:22:34 yeah we're getting reports of a field here or there that people have in their DB 15:22:41 they are easy to fix 15:22:47 * smyers has a nice pile of known issues to add to the 2.8.4 release notes, it seems 15:22:48 Does it make sense to have a sweeping migration that drops any fields that aren't part of our defined models? 15:22:49 see the last comment on 1931 15:23:10 bmbouter: so since these a same issues let's make same u p 15:23:16 jcline: that would be really great I think 15:23:25 we cannot make one blocker and another not if they have same root cause 15:23:27 that fix these once and for all 15:23:31 or we can fix them one by one 15:23:45 I can say fixing them as they come up is easy 15:24:01 let's triage this one that way and we can talk more about a comprehensive fix 15:25:39 bmbouter: i just propose to treat them with same u/p and not put diffent ones on each case 15:26:07 so triage h h ? 15:26:38 bmbouter: yeah probably 15:27:04 #idea Proposed for #1952: Priority: High, Severity: High, Target Platform Release: 2.8.5 15:27:04 !propose triage h h 2.8.5 15:27:04 Proposed for #1952: Priority: High, Severity: High, Target Platform Release: 2.8.5 15:27:28 !accept 15:27:28 #agreed Priority: High, Severity: High, Target Platform Release: 2.8.5 15:27:28 Current proposal accepted: Priority: High, Severity: High, Target Platform Release: 2.8.5 15:27:30 No issues to triage. 15:27:36 great job everyone! 15:27:53 !end 15:27:53 #endmeeting