14:30:11 #startmeeting Pulp Triage 2016-05-06 14:30:11 #info dkliban has joined triage 14:30:12 Meeting started Fri May 6 14:30:11 2016 UTC and is due to finish in 60 minutes. The chair is dkliban. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:30:12 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:30:12 The meeting name has been set to 'pulp_triage_2016_05_06' 14:30:12 dkliban has joined triage 14:30:20 !here 14:30:20 #info ttereshc has joined triage 14:30:21 ttereshc has joined triage 14:30:25 !here 14:30:25 #info bowlofeggs has joined triage 14:30:26 bowlofeggs has joined triage 14:30:28 #info preethi has joined triage 14:30:28 !here 14:30:29 preethi has joined triage 14:30:32 !here 14:30:32 #info smyers has joined triage 14:30:32 smyers has joined triage 14:30:34 #info ipanova has joined triage 14:30:34 !here 14:30:34 ipanova has joined triage 14:30:39 #info jcline has joined triage 14:30:39 !here 14:30:39 jcline has joined triage 14:30:55 we have a quorum 14:31:01 !care 1899 14:31:09 Let's hope that works^ :) 14:31:20 !next 14:31:22 16 issues left to triage: 1773, 1865, 1866, 1881, 1883, 1884, 1885, 1886, 1890, 1893, 1894, 1895, 1897, 1898, 1899, 1900 14:31:22 #topic No "unprotected/http" option available for ostree repos - http://pulp.plan.io/issues/1773 14:31:23 OSTree Support Issue #1773 [NEW] (unassigned) - Priority: Normal | Severity: Medium 14:31:23 No "unprotected/http" option available for ostree repos - http://pulp.plan.io/issues/1773 14:31:30 !care 1890 14:31:37 !care 1893 14:31:45 !care 1900 14:32:00 We needinfo'd this last week, and it doesn't appear to have changed. 14:32:01 Was someone supposed to take a look at 1773? 14:32:05 !here 14:32:05 #info pcreech has joined triage 14:32:06 pcreech has joined triage 14:32:13 bowlofeggs, you are so careful) 14:32:13 i propose we skip it 14:32:19 #idea Proposed for #1773: Skip this issue for this triage session. 14:32:19 !propose skip 14:32:19 Proposed for #1773: Skip this issue for this triage session. 14:32:28 !agreed 14:32:28 Error: "agreed" is not a valid command. 14:32:33 dkliban, you can !accept to...er...accept :) 14:32:37 !accept 14:32:37 #agreed Skip this issue for this triage session. 14:32:38 Current proposal accepted: Skip this issue for this triage session. 14:32:39 !accept 14:32:39 15 issues left to triage: 1865, 1866, 1881, 1883, 1884, 1885, 1886, 1890, 1893, 1894, 1895, 1897, 1898, 1899, 1900 14:32:39 #topic Error searching rpm repositories by fields using dot notation - http://pulp.plan.io/issues/1865 14:32:40 Pulp Issue #1865 [NEW] (unassigned) - Priority: Normal | Severity: Medium 14:32:40 Error searching rpm repositories by fields using dot notation - http://pulp.plan.io/issues/1865 14:33:08 asmacdo: are you working on 1865? 14:33:21 is this a regression? 14:33:39 This is likely a regression. I think asmacdo is still out 14:33:59 !propose high, high 14:33:59 Error: "propose" is not a valid command. 14:34:09 "This pulp-admin command works with pulp/pulp-admin 2.5.3 and MongoDB 2.4.9 installed on RHEL 6." Definitely a regression 14:34:18 Unexpected error, try !propose triage high high" 14:34:25 !propose triage high high 14:34:25 #idea Proposed for #1865: Priority: High, Severity: High 14:34:26 Proposed for #1865: Priority: High, Severity: High 14:34:37 everyone agree? 14:34:45 I need to support +1, I think, because +1 14:34:53 well. pulpbot does. Not me. :) 14:34:57 lol 14:35:02 Yeah 14:35:10 do we want to set a target release of 2.8.4? 14:35:16 yes 14:35:18 !propose h h 2.8.4 14:35:18 Error: "propose" is not a valid command. 14:35:22 hah 14:35:25 !propose triage h h 2.8.4 14:35:25 #idea Proposed for #1865: Priority: High, Severity: High, Target Platform Release: 2.8.4 14:35:25 Proposed for #1865: Priority: High, Severity: High, Target Platform Release: 2.8.4 14:35:28 sure 14:35:54 I like it 14:35:56 hm, apparently I didn't get any of these messages since triage started. 14:35:58 * mhrivnak catches up 14:36:00 !next 14:36:02 14 issues left to triage: 1866, 1881, 1883, 1884, 1885, 1886, 1890, 1893, 1894, 1895, 1897, 1898, 1899, 1900 14:36:03 #topic The Pulp Python bindings don't use requests and fail to retry - http://pulp.plan.io/issues/1866 14:36:03 Pulp Issue #1866 [NEW] (unassigned) - Priority: Normal | Severity: Medium 14:36:04 The Pulp Python bindings don't use requests and fail to retry - http://pulp.plan.io/issues/1866 14:36:08 !here 14:36:08 #info mhrivnak has joined triage 14:36:09 mhrivnak has joined triage 14:36:25 I updated this a little bit I think, but not enough (perhaps) 14:36:37 Another one I think we skipped last week. jcline do you still think it's NEEDINFO? 14:36:58 RHEL 5 might be a concern here, unless it has software collections too? 14:37:18 the story i'm workign on is to switch us to SCLs on EL 6, but maybe i could do it for RHEL 5 too to enable requests? 14:37:19 Yeah, I don't think we could switch to requests soon 14:37:20 * smyers kicks rhel5 14:37:30 i'm not sure whether EL 5 has SCLs or not though 14:37:36 Or maybe only for pulp-admin if we want to only support pulp-admin on rhel6+ 14:37:39 I think we can just wait for pulp 3 on that, when we won't need to support el5 anymore. 14:38:02 !propose triage wait 14:38:02 (propose triage [target_release]) -- Propose triage values including priority, severity, and an optional target release. 14:38:07 whoops 14:38:17 well i propose we hold off 14:38:23 Fine with me 14:38:30 but what do we do with the issue? skip it? 14:38:32 yeah 14:38:38 !skip 14:38:40 13 issues left to triage: 1881, 1883, 1884, 1885, 1886, 1890, 1893, 1894, 1895, 1897, 1898, 1899, 1900 14:38:41 #topic pulp-admin fails to display the erratum if not all metadata is present - http://pulp.plan.io/issues/1881 14:38:41 RPM Support Issue #1881 [NEW] (unassigned) - Priority: Normal | Severity: Medium 14:38:42 pulp-admin fails to display the erratum if not all metadata is present - http://pulp.plan.io/issues/1881 14:39:28 seems bad - it's a crash 14:39:35 !propose medium high 2.8.4 14:39:35 Error: "propose" is not a valid command. 14:39:42 !propose triage normal high 2.8.4 14:39:42 #idea Proposed for #1881: Priority: Normal, Severity: High, Target Platform Release: 2.8.4 14:39:43 Proposed for #1881: Priority: Normal, Severity: High, Target Platform Release: 2.8.4 14:39:43 and apparently a regression. 14:39:44 !propose triage medium high 2.8.4 14:39:48 heh 14:39:51 I think we agree. 14:39:55 agree 14:39:58 +1 14:39:58 +1 14:40:04 +1 14:40:12 Also it's easy to fix 14:40:31 remember to !accept, not !next 14:40:46 !accept 14:40:46 #agreed Priority: Normal, Severity: High, Target Platform Release: 2.8.4 14:40:47 Current proposal accepted: Priority: Normal, Severity: High, Target Platform Release: 2.8.4 14:40:48 #topic As a user, I can sync and publish all package types - http://pulp.plan.io/issues/1883 14:40:48 12 issues left to triage: 1883, 1884, 1885, 1886, 1890, 1893, 1894, 1895, 1897, 1898, 1899, 1900 14:40:48 Python Support Issue #1883 [ASSIGNED] (amacdona@redhat.com) - Priority: Normal | Severity: Medium 14:40:48 * smyers makes a note to have !next implicitly !accept, because it's silly to have two commands for that 14:40:48 As a user, I can sync and publish all package types - http://pulp.plan.io/issues/1883 14:41:00 I think this should be a story. 14:41:05 +1 14:41:06 And I left a comment to that effect. 14:41:10 this is also already in progress 14:41:13 i propose we move on 14:41:26 !propose accept 14:41:26 #idea Proposed for #1883: Leave the issue as-is, accepting its current state. 14:41:27 Proposed for #1883: Leave the issue as-is, accepting its current state. 14:41:38 does it need to be a story? 14:41:48 yeah this is a story 14:41:55 im going to update that one 14:41:59 ok 14:42:00 you can move on 14:42:05 !next 14:42:07 11 issues left to triage: 1884, 1885, 1886, 1890, 1893, 1894, 1895, 1897, 1898, 1899, 1900 14:42:07 #topic Lazy integration - http://pulp.plan.io/issues/1884 14:42:08 Python Support Issue #1884 [NEW] (unassigned) - Priority: Normal | Severity: Medium 14:42:08 Lazy integration - http://pulp.plan.io/issues/1884 14:42:09 this one is also a story 14:42:11 same 14:42:15 !propose accept 14:42:15 #idea Proposed for #1884: Leave the issue as-is, accepting its current state. 14:42:16 Proposed for #1884: Leave the issue as-is, accepting its current state. 14:42:34 Could also !propose other Convert to story 14:42:40 does "accept" mark it as triaged? 14:42:59 let's move on 14:43:02 That's the idea, but since it doesn't touch redmine it's still up to the triager 14:43:03 er, does proposing "accept" mean proposing marking it as triaged? 14:43:12 ok 14:43:20 ok. I think we should not mark it triaged. 14:43:25 !propose skip 14:43:25 #idea Proposed for #1884: Skip this issue for this triage session. 14:43:26 Proposed for #1884: Skip this issue for this triage session. 14:43:28 and move on. :) 14:43:32 +1 14:43:32 #agreed Skip this issue for this triage session. 14:43:32 !accept 14:43:32 Current proposal accepted: Skip this issue for this triage session. 14:43:34 10 issues left to triage: 1885, 1886, 1890, 1893, 1894, 1895, 1897, 1898, 1899, 1900 14:43:34 #topic Pulp uploads are failing with "too many open files" error - http://pulp.plan.io/issues/1885 14:43:35 Pulp Issue #1885 [NEW] (unassigned) - Priority: Normal | Severity: Medium 14:43:35 Pulp uploads are failing with "too many open files" error - http://pulp.plan.io/issues/1885 14:43:42 this one does sound bad 14:43:48 !propose triage medium high 2.8.4 14:43:48 #idea Proposed for #1885: Priority: Normal, Severity: High, Target Platform Release: 2.8.4 14:43:49 ooof 14:43:49 Proposed for #1885: Priority: Normal, Severity: High, Target Platform Release: 2.8.4 14:43:53 I don't think we should accept this bug at this time 14:43:57 why not? 14:43:58 mhrivnak: i did mark it as triaged but i'll change that now 14:44:02 bmbouter: ? 14:44:12 it's very likely an environmtal problem 14:44:30 bmbouter: what if we accept it just to try to reproduce? 14:44:47 agree with bowlofeggs 14:45:02 that sounds fine 14:45:06 bmbouter: we could ask them for reproduction steps with a clean pulp install 14:45:11 #idea Proposed for #1885: Leave the issue as-is, accepting its current state. 14:45:11 !propose accept 14:45:11 Proposed for #1885: Leave the issue as-is, accepting its current state. 14:45:18 Yeah. If it turns out to be an environmental problem we can close it with a note 14:45:32 +1 14:45:33 cool sounds good 14:45:34 +1 14:45:37 I think they are hitting a scale limitation 14:45:39 we should make sure to put in a note that it may be an environmental problem if it isn't already there 14:45:45 we could document that better perhaps 14:45:48 so whoever picks it up will know 14:45:57 I'll put a note on it 14:46:01 bmbouter, thank you 14:46:03 bmbouter: that could be the fix for the issue 14:46:08 mark it as triaged? 14:46:14 (making a docs pr, that is) 14:46:20 let's not mark it triaged for this time 14:46:27 #action bmbouter Update #1885 to note that this might be an environmental problem, not a pulp-specific problem. 14:46:27 !action bmbouter Update #1885 to note that this might be an environmental problem, not a pulp-specific problem. 14:46:38 but let's suggest it might be environmental, and ask for reproduction steps 14:46:47 can we skip now? 14:46:49 !action bmbouter ask for reproduction steps 14:46:49 #action bmbouter ask for reproduction steps 14:46:53 so !propose skip? 14:46:54 yeah 14:46:57 !propose skip[ 14:46:57 Error: Missing "]". You may want to quote your arguments with double quotes in order to prevent extra brackets from being evaluated as nested commands. 14:47:02 hah 14:47:04 !propose skip 14:47:04 #idea Proposed for #1885: Skip this issue for this triage session. 14:47:04 Proposed for #1885: Skip this issue for this triage session. 14:47:07 you make pulpbot so mad 14:47:10 haha 14:47:11 #agreed Skip this issue for this triage session. 14:47:11 !accept 14:47:11 Current proposal accepted: Skip this issue for this triage session. 14:47:13 #topic Removal of the unit fails if criteria contains unknown fields - http://pulp.plan.io/issues/1886 14:47:13 9 issues left to triage: 1886, 1890, 1893, 1894, 1895, 1897, 1898, 1899, 1900 14:47:14 Pulp Issue #1886 [NEW] (unassigned) - Priority: Normal | Severity: Medium 14:47:15 Removal of the unit fails if criteria contains unknown fields - http://pulp.plan.io/issues/1886 14:47:15 are you getting tracebacks? 14:47:21 not at all 14:47:27 nice 14:47:47 this is likely a regression 14:47:51 !propose medium medium 2.8.4 14:47:51 Error: "propose" is not a valid command. 14:47:59 !propose triage medium medium 2.8.4 14:47:59 #idea Proposed for #1886: Priority: Normal, Severity: Medium, Target Platform Release: 2.8.4 14:48:00 Proposed for #1886: Priority: Normal, Severity: Medium, Target Platform Release: 2.8.4 14:48:03 I'm not so sure 14:48:10 Is this a bug? :) 14:48:29 well, nothing a user can do should be able to produce a traceback 14:48:31 Maybe this should return a 400? 14:48:33 so it's at least a bug in that sense 14:48:38 Yeah 14:48:38 yep 14:48:42 if anything, we could return a nice error 14:48:44 Right, but not a regression 14:48:48 smyers: seems like just for docker plugin, other plugins can handle that 14:48:55 smyers: yeah maybe you are right 14:49:06 Yeah, I'd be fine with not 2.8.4 14:49:15 if it's not a regression, I'd say not 2.84 14:49:20 agreed. 14:49:22 what jcline said :) 14:49:23 !propose med med 14:49:23 Error: "propose" is not a valid command. 14:49:24 this is weird 14:49:28 !propose triage med med 14:49:28 #idea Proposed for #1886: Priority: Normal, Severity: Medium 14:49:29 Proposed for #1886: Priority: Normal, Severity: Medium 14:49:33 you queried about docker 14:49:39 but the traceback is about rpm? 14:49:40 obviously need to support the naked !propose command. 14:49:51 bowlofeggs, both 14:49:57 !accept 14:49:57 #agreed Priority: Normal, Severity: Medium 14:49:57 Current proposal accepted: Priority: Normal, Severity: Medium 14:49:59 8 issues left to triage: 1890, 1893, 1894, 1895, 1897, 1898, 1899, 1900 14:50:00 #topic pulp-qpid-ssl-cfg echoes the NSS DB password - http://pulp.plan.io/issues/1890 14:50:00 Pulp Issue #1890 [NEW] (unassigned) - Priority: Normal | Severity: Medium 14:50:00 pulp-qpid-ssl-cfg echoes the NSS DB password - http://pulp.plan.io/issues/1890 14:50:01 bowlofeggs: Issue 1890 is currently being discussed. 14:50:08 ^ BAM 14:50:38 hey nice! 14:50:47 this one is a security hardening bug 14:50:55 Need a CVE for it? 14:50:57 it's pretty lame - it echoes a password onto your screeen 14:50:59 Nope 14:51:07 smyers: no i talked to PS and they said not a CVE but we should do it 14:51:09 (to the CVE need) 14:51:13 it's also a 1-liner 14:51:16 ah 14:51:17 just remove one line 14:51:20 easy peasey 14:51:20 * smyers finished reading the bug 14:51:28 !propose triage med med 2.8.4 14:51:28 #idea Proposed for #1890: Priority: Normal, Severity: Medium, Target Platform Release: 2.8.4 14:51:29 Proposed for #1890: Priority: Normal, Severity: Medium, Target Platform Release: 2.8.4 14:51:33 nice friday task 14:51:40 +1 14:51:44 !suggest also add the easy fix tag 14:51:44 #idea also add the easy fix tag 14:52:02 #agreed Priority: Normal, Severity: Medium, Target Platform Release: 2.8.4 14:52:02 !accept 14:52:02 Current proposal accepted: Priority: Normal, Severity: Medium, Target Platform Release: 2.8.4 14:52:03 !action bowlofeggs add the easy tag 14:52:03 7 issues left to triage: 1893, 1894, 1895, 1897, 1898, 1899, 1900 14:52:04 #topic pulp_docker does not support Docker < 1.10 with manifests that were published from Docker >= 1.10 - http://pulp.plan.io/issues/1893 14:52:04 #action bowlofeggs add the easy tag 14:52:04 Docker Support Issue #1893 [NEW] (unassigned) - Priority: Normal | Severity: Medium 14:52:04 pulp_docker does not support Docker < 1.10 with manifests that were published from Docker >= 1.10 - http://pulp.plan.io/issues/1893 14:52:05 bowlofeggs: Issue 1893 is currently being discussed. 14:52:22 so this issue is theoretical at the moment 14:52:38 i asked RCM to do some testing to find out if they are immediately impacted or not 14:52:47 they haven't responded to that yet 14:52:54 How was it discovered? 14:52:54 we will have to do something eventually 14:53:03 !propose skip 14:53:03 #idea Proposed for #1893: Skip this issue for this triage session. 14:53:04 Proposed for #1893: Skip this issue for this triage session. 14:53:05 i think they were just reading the 1.10 release notes 14:53:17 and realized there was a potential problem 14:53:20 let's wait on RCM's input 14:53:21 really, there is a real problem 14:53:35 but it's unknown how urgent our attention needs to be 14:53:36 +1 14:53:46 #agreed Skip this issue for this triage session. 14:53:46 !accept 14:53:46 Current proposal accepted: Skip this issue for this triage session. 14:53:48 #topic unassociate rpms API request using unit_ids removes all rpm from repo - http://pulp.plan.io/issues/1894 14:53:48 6 issues left to triage: 1894, 1895, 1897, 1898, 1899, 1900 14:53:49 Pulp Issue #1894 [NEW] (unassigned) - Priority: Normal | Severity: Medium 14:53:50 unassociate rpms API request using unit_ids removes all rpm from repo - http://pulp.plan.io/issues/1894 14:54:10 lol 14:54:33 FWIW, I wouldn't mind slowing this triage down just a touch. 14:54:40 still loling 14:54:42 ok 14:54:45 whoah 14:54:52 this bug is crazy 14:54:57 !propose high high 2.8.4 14:54:57 Error: "propose" is not a valid command. 14:54:57 Yeah, haha 14:55:02 !propose triage high high 2.8.4 14:55:02 #idea Proposed for #1894: Priority: High, Severity: High, Target Platform Release: 2.8.4 14:55:03 Proposed for #1894: Priority: High, Severity: High, Target Platform Release: 2.8.4 14:55:07 Unless I'm missing something? 14:55:09 +1 14:55:10 Because wow. 14:55:17 +1 14:55:25 +1 14:55:31 This one is quite bizarre. +1 14:55:41 !action pulp unbreak pulp 14:55:41 #action pulp unbreak pulp 14:55:48 Maybe might even bump the priority to urgent 14:56:04 Because while the client should never return tracebacks it should *really* never empty your repos? 14:56:15 hahaha 14:56:19 I think that urgent's s a stretch, though 14:56:25 there was an error so I DELETED EVERYTHING 14:56:39 I could see it either way. 14:56:49 i marked it high/high 2.8.4 14:57:19 next ? 14:57:21 !propose move along 14:57:21 Error: "propose" is not a valid command. 14:57:24 !propose next 14:57:24 Error: "propose" is not a valid command. 14:57:27 !propose accept 14:57:27 #idea Proposed for #1894: Leave the issue as-is, accepting its current state. 14:57:28 Proposed for #1894: Leave the issue as-is, accepting its current state. 14:57:32 #agreed Leave the issue as-is, accepting its current state. 14:57:32 !accept 14:57:32 Current proposal accepted: Leave the issue as-is, accepting its current state. 14:57:32 whoops 14:57:34 noooo 14:57:34 5 issues left to triage: 1895, 1897, 1898, 1899, 1900 14:57:34 #topic Recursive RPM unit copies are not recursive - http://pulp.plan.io/issues/1895 14:57:35 RPM Support Issue #1895 [MODIFIED] (mhrivnak) - Priority: Normal | Severity: Medium | Target Release: 2.8.4 14:57:35 Recursive RPM unit copies are not recursive - http://pulp.plan.io/issues/1895 14:57:39 hahah i ruined it 14:57:46 !propose triage high high 2.8.4 14:57:46 #idea Proposed for #1895: Priority: High, Severity: High, Target Platform Release: 2.8.4 14:57:47 Proposed for #1895: Priority: High, Severity: High, Target Platform Release: 2.8.4 14:57:52 already fixed. 14:58:01 +1 14:58:08 +1 14:58:11 recursive copy was completely broken. 14:58:13 +1 14:58:15 fwiw, i think i actually caused the last one to not be high/high 14:58:21 so let's make sure we get that one right 14:58:43 i was thinking +1 so i typed !propose accept, as in "i propose that dennis type !accept" 14:58:44 #agreed Priority: High, Severity: High, Target Platform Release: 2.8.4 14:58:44 !accept 14:58:44 Current proposal accepted: Priority: High, Severity: High, Target Platform Release: 2.8.4 14:58:45 #topic catalog entries not created for pre-existing units - http://pulp.plan.io/issues/1897 14:58:46 4 issues left to triage: 1897, 1898, 1899, 1900 14:58:46 1894 is set high/high 14:58:46 RPM Support Issue #1897 [NEW] (unassigned) - Priority: Normal | Severity: Medium 14:58:47 catalog entries not created for pre-existing units - http://pulp.plan.io/issues/1897 14:58:52 but i changed the proposal ☹ 14:59:10 * smyers punches bowlofeggs 14:59:23 haha 14:59:32 does this one cause any practical error? 14:59:36 Yeah 14:59:36 !propose high high 2.8.4 14:59:36 Error: "propose" is not a valid command. 14:59:48 as in, what problem is there if the catalog entries are missing? 14:59:48 It'll make fetching lazy content fail 14:59:50 bowlofeggs: please stop messing around with the propose command 14:59:52 ah 14:59:55 dkliban: lol 15:00:00 404s. Thousands of them. 15:00:01 !propose triage high high 2.8.4 15:00:01 #idea Proposed for #1897: Priority: High, Severity: High, Target Platform Release: 2.8.4 15:00:01 Proposed for #1897: Priority: High, Severity: High, Target Platform Release: 2.8.4 15:00:01 +1 15:00:05 +1 15:00:10 +1 15:00:17 +1 15:00:26 * jcline jumps on the +1 train 15:00:34 The problem is content cannot be accessed on-demand through the streamer. 15:00:45 !action jcline needs to paint a picture of the 404 devastation that occurs as a result of #1897 15:00:45 #action jcline needs to paint a picture of the 404 devastation that occurs as a result of #1897 15:00:52 We had a katello dev get into that situation. 15:00:57 jcline: water colours please 15:01:22 are we ready to move on? 15:01:23 let's move on 15:01:25 +1 15:01:26 yes 15:01:29 !accept 15:01:29 #agreed Priority: High, Severity: High, Target Platform Release: 2.8.4 15:01:30 Current proposal accepted: Priority: High, Severity: High, Target Platform Release: 2.8.4 15:01:31 3 issues left to triage: 1898, 1899, 1900 15:01:31 #topic Errata search does not filter packages that are not in repo - http://pulp.plan.io/issues/1898 15:01:32 RPM Support Issue #1898 [NEW] (unassigned) - Priority: Normal | Severity: Medium 15:01:32 Errata search does not filter packages that are not in repo - http://pulp.plan.io/issues/1898 15:01:43 haha right 15:01:48 oh boy 15:01:49 Because we do that filtering on publish :( 15:01:55 exactly 15:01:57 * smyers punches errata 15:02:10 * ttereshc joins smyers 15:02:13 smyers: is so violent today 15:02:19 yeah i'm not sure what i think about this one 15:02:20 * smyers punches violence 15:02:30 smyers: lol 15:02:30 lol 15:02:46 Given recent changes, it would actually be pretty easy to do the filtering. 15:02:52 I think that it's possible that with ttereshc's work in rm #858 we can potentially refactor errata to be less junky 15:02:52 RPM Support Issue #858 [MODIFIED] (ttereshc) - Priority: High | Severity: Medium | Target Release: 2.8.4 15:02:53 As a user, I would like to receive updated errata metadata - http://pulp.plan.io/issues/858 15:03:01 !propose triage medium med 15:03:01 #idea Proposed for #1898: Priority: Normal, Severity: Medium 15:03:01 Proposed for #1898: Priority: Normal, Severity: Medium 15:03:07 +1 15:03:15 smyers, I am not sure but we can discuss it later 15:03:19 +1 15:03:21 +1 15:03:21 cool 15:03:43 I think we're good then 15:03:48 +1 15:04:05 move along? 15:04:10 let's move on 15:04:15 these are not the droids we're looking for. 15:04:21 haha they really aren't 15:04:30 !accept 15:04:30 #agreed Priority: Normal, Severity: Medium 15:04:30 Current proposal accepted: Priority: Normal, Severity: Medium 15:04:31 #topic Repo sync failure with "time data '' does not match format '%Y-%m-%d %H:%M:%S'" - http://pulp.plan.io/issues/1899 15:04:32 2 issues left to triage: 1899, 1900 15:04:32 RPM Support Issue #1899 [NEW] (unassigned) - Priority: Normal | Severity: Medium 15:04:33 Repo sync failure with "time data '' does not match format '%Y-%m-%d %H:%M:%S'" - http://pulp.plan.io/issues/1899 15:04:34 ttereshc: Issue 1899 is currently being discussed. 15:05:21 ttereshc, sorry I haven't elaborated on our last emails about this one. 15:05:22 is it a pulp bug? 15:05:25 I think it is not a bug in pulp, and I made a PR to fix fixtures repo 15:05:56 mhrivnak, we can discuss after triage if it is a proper pulp behavior 15:05:57 i propose we skip until we know if it's a fixture problem or not 15:06:03 !propose skip 15:06:03 #idea Proposed for #1899: Skip this issue for this triage session. 15:06:03 Is it possible that users will upgrade to pulp 2.8.4 and start getting this failure? 15:06:03 Proposed for #1899: Skip this issue for this triage session. 15:06:04 i'd close it as it is not a bug in pulp 15:06:11 It really looks like a fixture problem to me 15:06:12 when before it didn't fail? 15:06:19 smyers: +1 15:06:24 Yeah, it looks like data validation is rightfully failing 15:06:40 mhrivnak, first sync won't fail 15:06:44 data validation shouldn't traceback though 15:06:52 so we *should* at least fix that 15:07:07 it should give a helpful error message to the user instead 15:07:08 smyers: is there some !propose lest's close it? 15:07:09 My concern is that I think we're getting more strict about validation in a Z release. 15:07:14 "the remote repo is out of its mind" 15:07:23 #idea Proposed for #1899: Close as NOTABUG 15:07:23 !propose other Close as NOTABUG 15:07:23 Proposed for #1899: Close as NOTABUG 15:07:27 ^ipanova 15:07:36 smyers: sweet 15:07:39 With !propose other you can do whatever you want 15:07:41 i propose we keep it as a bad error message bug 15:07:49 #idea Proposed for #1899: keep it as a bad error message bug 15:07:49 !propose other keep it as a bad error message bug 15:07:49 Proposed for #1899: keep it as a bad error message bug 15:08:05 I think that could/should be a new issue, rather than mangling this one 15:08:13 yeah works for me 15:08:23 bowlofeggs: this kind of error we have in multiple situations 15:08:27 !propose other Close as NOTABUG, open a new issue to track the validation error 15:08:27 #idea Proposed for #1899: Close as NOTABUG, open a new issue to track the validation error 15:08:28 Proposed for #1899: Close as NOTABUG, open a new issue to track the validation error 15:08:28 we really do ☹ 15:08:32 ttereshc, are we sure there is no data like this on CDN? 15:08:32 +1 15:08:43 I agree it shouldn't traceback, and I'm indifferent about whether this issue transforms or a new one is made. 15:09:00 as in, we're not going to have users try to sync repos they care about and suddenly start hitting this failure? 15:09:27 mhrivnak, no we are not 15:09:28 seems like a problem to me 15:09:50 I'm generally against completely rewriting issues into new ones, since we lose some paper trail in the event that this really *is* a bug and we should reopen this. 15:09:53 Well if there is, isn't a malformed erratum that should get fixed there? 15:09:58 mhrivnak, but if we will silently skip, then they won't be aware of the problem 15:10:12 That said, I agree with ttereshc, this is a malformed errata, not a pulp bug. 15:10:48 Ok. If we think there's a chance the stricter validation will affect real-world repos, I think we need to carefully consider *some change* in 2.8.4. 15:11:12 And also ttereshc++ for making the pulp-fixture PR, which has already been merged. 15:11:26 FYI: I changed the errata with the first commit to the fixtures 15:11:30 malformed or not, if pulp was happily syncing a repo with 2.8.3, and starts failing with 2.8.4, that's a bad user experience. 15:11:49 I was told the zoo repo errata was bad and that the one from jsherrill's zoo repo was the one to use 15:12:04 mhrivnak: so the repository chagned 15:12:08 that likely introduced this issue 15:12:10 Maybe have a task to check the errata on the CDN? 15:12:25 if we're talking about the zoo repo here 15:12:26 Ok. If we're certain it's isolated to this repo, then that's fine. 15:12:28 I think the current triage proposal is still good, and this is a discussion for after this meeting. 15:12:30 looking at pulp-fixtures updateinfo.xml it missed the updated field, but afaik errata should always have updated field. that said --> updateinfo.xml was not correct 15:12:35 ttereshc: am i right?^ 15:12:50 ^ that is my understanding 15:12:51 I think so 15:13:06 then 15:13:07 !propose other Close as NOTABUG 15:13:07 #idea Proposed for #1899: Close as NOTABUG 15:13:08 Proposed for #1899: Close as NOTABUG 15:13:23 we need to fix the fixture repo 15:13:27 so the bug could be about tht 15:13:32 but mhrivnak's concern is also valid 15:13:33 dkliban: ttereshcsubmitted PR already 15:13:39 I see. 15:13:49 ipanova: PR to fix the fixture? 15:13:54 dkliban, yes 15:13:54 dkliban: yep 15:13:56 ttereshc, if this came up only because of an error in the recent changes to the zoo repo, then I feel good about it. 15:13:59 and seems like already merged 15:14:16 mhrivnak, let's discuss it. 15:14:21 i think we should still have the last proposal, to file a new bug too 15:14:28 yerp 15:14:28 i am not closing it yeet ... i am just skipping it. 15:14:34 !propose other Close as NOTABUG, open a new issue to track the validation error 15:14:34 #idea Proposed for #1899: Close as NOTABUG, open a new issue to track the validation error 15:14:34 mhrivnak, I think the zoo repo was not good but you are also right 15:14:34 Proposed for #1899: Close as NOTABUG, open a new issue to track the validation error 15:14:52 mhrivnak, after triage 15:14:55 It really should be closed, imo (as proposed) 15:15:07 I agree with the proposal. 15:15:16 #idea Proposed for #1899: Leave the issue as-is, accepting its current state. 15:15:16 !propose accept 15:15:17 Proposed for #1899: Leave the issue as-is, accepting its current state. 15:15:21 ugh 15:15:24 #idea Proposed for #1899: Close as NOTABUG, open a new issue to track the validation error 15:15:24 !propose other Close as NOTABUG, open a new issue to track the validation error 15:15:24 Proposed for #1899: Close as NOTABUG, open a new issue to track the validation error 15:15:29 i keep forgetting what that means 15:15:45 ttereshc, I think a new bug to improve the error handling would be great, and we can discuss details there? Does that line up with your thinking? 15:15:55 !skip 15:15:58 mhrivnak, ok 15:15:59 1 issues left to triage: 1900 15:16:00 #topic RFE: Please enable/support Redis as a celery broker transport in pulp - http://pulp.plan.io/issues/1900 15:16:01 Pulp Issue #1900 [NEW] (unassigned) - Priority: Normal | Severity: Medium 15:16:01 RFE: Please enable/support Redis as a celery broker transport in pulp - http://pulp.plan.io/issues/1900 15:16:02 bowlofeggs: Issue 1900 is currently being discussed. 15:16:04 :( it should have been accepted 15:16:21 yeah i thought we had agreement on that last one 15:16:41 ok so this one is really a feature request, but i'm not sure how much they need it yet 15:16:41 !suggest resolution for #1899 should have been "Close as NOTABUG, open a new issue to track the validation error" 15:16:41 #idea resolution for #1899 should have been Close as NOTABUG, open a new issue to track the validation error 15:16:43 bowlofeggs: to close it as not a bug? i think that we need to test the change ttereshc made and then close it 15:16:58 i've been talking with maxamillion about this one 15:17:09 as fedora wants to use pulp, they do not want to use qpidd 15:17:16 but not rabbit? 15:17:17 they might use rabbit 15:17:19 :) 15:17:23 but they already have redis in a few places 15:17:26 and they'd prefer to use that 15:17:31 we are anyway planning to get rid of qpid 15:17:32 Are we doing anything currently to make redis not work? 15:17:35 since celery can do it, i think it won't be much work for us 15:17:43 i think we may need ot test it 15:17:48 i'd bet it doesn't "just work" 15:17:52 just because of our settings 15:17:59 i've not tried it though 15:18:23 !propose other Convert to task: Test pulp with redis as a kombu backend 15:18:23 #idea Proposed for #1900: Convert to task: Test pulp with redis as a kombu backend 15:18:24 Proposed for #1900: Convert to task: Test pulp with redis as a kombu backend 15:18:28 +1 15:18:28 +1 15:18:31 +1 15:18:39 !accept 15:18:39 #agreed Convert to task: Test pulp with redis as a kombu backend 15:18:39 +1 15:18:39 Current proposal accepted: Convert to task: Test pulp with redis as a kombu backend 15:18:40 No issues to triage. 15:18:44 yay 15:18:48 nice 15:18:50 good job pulpbot 15:18:55 do we want to revisit any issues i skipped? 15:19:03 on #1900, who is going to do the converting? 15:19:05 maybe that last one 15:19:15 bowlofeggs, you mean 1899? 15:19:37 smyers: yeah 15:20:11 Yeah, I'd like to revisit that one as well 15:20:26 !issue 1899 15:20:26 #topic Repo sync failure with "time data '' does not match format '%Y-%m-%d %H:%M:%S'" - http://pulp.plan.io/issues/1899 15:20:27 RPM Support Issue #1899 [NEW] (unassigned) - Priority: Normal | Severity: Medium 15:20:27 Repo sync failure with "time data '' does not match format '%Y-%m-%d %H:%M:%S'" - http://pulp.plan.io/issues/1899 15:20:28 ttereshc: Issue 1899 is currently being discussed. 15:20:54 bowlofeggs, got that prosal handy? 15:21:00 proposal 15:21:41 #idea Proposed for #1899: Close as NOTABUG, open a new issue to track the validation error 15:21:41 !propose other Close as NOTABUG, open a new issue to track the validation error 15:21:41 Proposed for #1899: Close as NOTABUG, open a new issue to track the validation error 15:21:43 found it 15:21:57 oh thanks 15:22:17 So why not close it? 15:22:32 2 things that I noticed with that bug 15:23:06 syncing with same feed was successful in 2.8.3 15:23:39 also the zoo repo in demo worked 15:23:47 and the fixture one failed 15:23:52 in 2.9 15:24:04 Sorry, computer froze, had to reboot. :( 15:24:07 preethi: yeah ... those repos aredifference 15:24:12 they are different 15:24:52 and in 2.8.3 and before errata were not updated 15:25:08 preethi: zoo repo in demo_repos is different from the fixtures repo. 15:25:27 and the fixtures repo has now been fixed. so you should try it again 15:26:11 dkliban: that is no problem. Just mentioed it since we were discussing the issue. 15:26:45 preethi: gotcha 15:26:50 I guess the question is whether or not we introduced a regression by rejecting malformed errata that we used to accept, is that right? 15:27:12 And then, is that really an issue, since the only malformed errata we've seen is the one we made for zoo? 15:27:19 did we used to accept it though? 15:27:22 smyers, I think that's exactly it. 15:27:38 dkliban, yes we did 15:27:41 I think we did get stricter in a Z release. 15:27:50 updated field is just a string 15:27:55 And also that the only place we've seen a problem is our own recent change to zoo. 15:28:05 And that's the annoying part... 15:28:12 ttereshc: gotcha 15:28:20 mhrivnak, we still accept malformed errata for the first time 15:28:27 Ya, it's kind of bad luck. 15:28:37 ttereshc, ah yes. 15:28:37 but we can't update if it is malformed 15:29:00 so that's potentially the new issue about how to handle the validation error 15:29:02 in this case we fail 15:29:14 smyers, agree 15:29:15 But #1899 is not a bug, imo. 15:29:59 And if we break backward compatibility in 2.8.3 for this, it's perfectly reasonable to fix it in 2.8.4. 15:30:33 But while I think we did technically break backward compatibility, we only did it when updating the broken errata the we made for zoo. 15:30:51 It's not even an edge case, it's a completely contrived unreal case. 15:31:11 +1 15:32:08 so whlol 15:32:11 lol 15:32:37 smyers, we do not know what data can be in the `updated` field in our customers' databases.. 15:32:42 anything) 15:33:30 Don't we, though? It would be whatever was in the errata they got from their feed, which (so far) we have no reason to think has been made "wrong" 15:33:44 smyers, they can upload 15:33:56 not only sync 15:34:01 But if they upload something that isn't an errata... 15:34:04 It's not pulp's fault 15:34:14 They can "upload" an errata. 15:34:38 But I suspect nobody does except RCM. 15:34:39 Right, but if they upload something as an errata that isn't a valid errata, that isn't pulp's problem. 15:35:00 smyers: +1 15:35:04 Except for the part where we need to improve the feedback on out validation, and explode with a good reason, not a traceback. 15:35:12 +1 15:35:18 +1 15:35:22 (which is the current proposal) 15:35:25 * mhrivnak notes that we are 5 minutes over time. 15:35:48 smyers, I agree it isn't Pulp's problem, but it could be perceived that it was our problem, which is why I think we resolve that by improving the feedback 15:35:58 ...which is the current proposal 15:36:06 smyers, yep - I'm agreeing :) 15:36:17 sweet 15:36:17 can i so can we end triage? 15:36:29 I think !accept then !end, yeah 15:36:34 #agreed Close as NOTABUG, open a new issue to track the validation error 15:36:34 !accept 15:36:34 Current proposal accepted: Close as NOTABUG, open a new issue to track the validation error 15:36:35 #endmeeting 15:36:35 !end 15:36:35 No issues to triage.