15:31:08 <fao89> #startmeeting Pulp Triage 2019-12-03
15:31:08 <fao89> #info fao89 has joined triage
15:31:08 <fao89> !start
15:31:08 <pulpbot> Meeting started Tue Dec  3 15:31:08 2019 UTC.  The chair is fao89. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:31:08 <pulpbot> Useful Commands: #action #agreed #help #info #idea #link #topic.
15:31:08 <pulpbot> The meeting name has been set to 'pulp_triage_2019-12-03'
15:31:08 <pulpbot> fao89: fao89 has joined triage
15:31:12 <ppicka> #info ppicka has joined triage
15:31:12 <ppicka> !here
15:31:12 <pulpbot> ppicka: ppicka has joined triage
15:31:12 <bmbouter> yes because at the db level it's fine https://github.com/pulp/pulpcore/blob/master/pulpcore/app/models/repository.py#L44
15:31:14 <ggainey> #info ggainey has joined triage
15:31:14 <ggainey> !here
15:31:14 <pulpbot> ggainey: ggainey has joined triage
15:31:14 <bmbouter> ok more on this later
15:31:15 <ttereshc> gmbnomis, yeah, I see now what you mean
15:31:16 <bmbouter> #info bmbouter has joined triage
15:31:16 <bmbouter> !here
15:31:16 <pulpbot> bmbouter: bmbouter has joined triage
15:31:22 <fao89> !next
15:31:23 <pulpbot> fao89: 7 issues left to triage: 5826, 5825, 5822, 5820, 5818, 5813, 4343
15:31:24 <fao89> #topic https://pulp.plan.io/issues/5826
15:31:24 <pulpbot> RM 5826 - jsherril@redhat.com - NEW - bindings ruby and python libraries are missing license or declare (somewhat) incorrect one
15:31:25 <dawalker> #info dawalker has joined triage
15:31:25 <dawalker> !here
15:31:25 <ttereshc> bmbouter, yup
15:31:26 <pulpbot> https://pulp.plan.io/issues/5826
15:31:27 <pulpbot> dawalker: dawalker has joined triage
15:31:30 <ttereshc> #info ttereshc has joined triage
15:31:30 <ttereshc> !here
15:31:30 <pulpbot> ttereshc: ttereshc has joined triage
15:31:58 <dkliban> #info dkliban has joined triage
15:31:58 <dkliban> !here
15:31:58 <pulpbot> dkliban: dkliban has joined triage
15:31:59 <ipanova> #info ipanova has joined triage
15:31:59 <ipanova> !here
15:31:59 <pulpbot> ipanova: ipanova has joined triage
15:32:01 <fao89> #idea Proposed for #5826: Leave the issue as-is, accepting its current state.
15:32:01 <fao89> !propose accept
15:32:01 <pulpbot> fao89: Proposed for #5826: Leave the issue as-is, accepting its current state.
15:32:07 <ggainey> +1
15:32:28 <bmbouter> add to 3.0 blockers and add documentation
15:32:30 <bmbouter> and add ot sprint
15:32:36 <dawalker> +1
15:32:38 <ipanova> i've seen a pr somewhere
15:32:40 <dkliban> yeah ... let's fix this week
15:32:49 <ttereshc> +1
15:32:49 <ipanova> +1
15:32:52 <dawalker> there's already an issue on the blockers to clarify licensing
15:33:03 <fao89> #idea Proposed for #5826: accept, add to 3.0 blockers and add to sprint
15:33:03 <fao89> !propose other accept, add to 3.0 blockers and add to sprint
15:33:03 <pulpbot> fao89: Proposed for #5826: accept, add to 3.0 blockers and add to sprint
15:33:07 <fao89> #agreed accept, add to 3.0 blockers and add to sprint
15:33:07 <fao89> !accept
15:33:07 <pulpbot> fao89: Current proposal accepted: accept, add to 3.0 blockers and add to sprint
15:33:07 <dkliban> this is for the bindings package that we generate
15:33:08 <fao89> #topic https://pulp.plan.io/issues/5825
15:33:08 <pulpbot> fao89: 6 issues left to triage: 5825, 5822, 5820, 5818, 5813, 4343
15:33:09 <pulpbot> RM 5825 - daviddavis - NEW - Pulp doesn't work with the newest django 3.0 release
15:33:10 <pulpbot> https://pulp.plan.io/issues/5825
15:33:15 <dawalker> ok
15:33:27 <dkliban> let's accept and not add to sprint
15:33:30 <bmbouter> +1
15:33:42 <dawalker> +1
15:33:43 <bmbouter> we wouldn't fix until 3.1 at the easiest anyway
15:33:47 <dalley> +1
15:33:48 <ppicka> +1
15:33:49 <ipanova> +1
15:34:08 <fao89> #idea Proposed for #5825: Leave the issue as-is, accepting its current state.
15:34:08 <fao89> !propose accept
15:34:08 <pulpbot> fao89: Proposed for #5825: Leave the issue as-is, accepting its current state.
15:34:11 <fao89> #agreed Leave the issue as-is, accepting its current state.
15:34:11 <fao89> !accept
15:34:11 <pulpbot> fao89: Current proposal accepted: Leave the issue as-is, accepting its current state.
15:34:12 <mikedep333> !here
15:34:12 <fao89> #topic https://pulp.plan.io/issues/5822
15:34:12 <mikedep333> #info mikedep333 has joined triage
15:34:13 <pulpbot> fao89: 5 issues left to triage: 5822, 5820, 5818, 5813, 4343
15:34:14 <pulpbot> RM 5822 - flamarion - NEW - Debian Repos - Not possible sync multiple releases
15:34:15 <pulpbot> https://pulp.plan.io/issues/5822
15:34:16 <pulpbot> mikedep333: mikedep333 has joined triage
15:34:50 <dkliban> let's move to Debian support
15:34:56 <ttereshc> yeah
15:35:05 <bmbouter> +1
15:35:06 <ttereshc> #idea Proposed for #5822: move to debian project
15:35:06 <ttereshc> !propose other move to debian project
15:35:06 <pulpbot> ttereshc: Proposed for #5822: move to debian project
15:35:09 <ipanova> +1
15:35:15 <dawalker> +1
15:35:27 <mikedep333> +1
15:35:30 <fao89> #agreed move to debian project
15:35:30 <fao89> !accept
15:35:30 <pulpbot> fao89: Current proposal accepted: move to debian project
15:35:31 <pulpbot> fao89: 4 issues left to triage: 5820, 5818, 5813, 4343
15:35:31 <fao89> #topic https://pulp.plan.io/issues/5820
15:35:32 <pulpbot> RM 5820 - ttereshc - NEW - pulp-2to3-migration requires all supported Pulp 3 plugins to be installed
15:35:33 <pulpbot> https://pulp.plan.io/issues/5820
15:35:54 <dkliban> accept and add to sprint
15:36:04 <ipanova> +1
15:36:05 <fao89> #idea Proposed for #5820: accept and add to sprint
15:36:05 <fao89> !propose other accept and add to sprint
15:36:05 <pulpbot> fao89: Proposed for #5820: accept and add to sprint
15:36:06 <ttereshc> !propose other accept and add to sprint
15:36:13 <bmbouter> +1
15:36:14 <ttereshc> :)
15:36:18 <fao89> #agreed accept and add to sprint
15:36:18 <fao89> !accept
15:36:18 <pulpbot> fao89: Current proposal accepted: accept and add to sprint
15:36:19 <fao89> #topic https://pulp.plan.io/issues/5818
15:36:19 <pulpbot> fao89: 3 issues left to triage: 5818, 5813, 4343
15:36:20 <pulpbot> RM 5818 - mdepaulo@redhat.com - NEW - pulp_user gets created with wrong settings when developer_user is defined and different
15:36:21 <pulpbot> https://pulp.plan.io/issues/5818
15:36:33 <dkliban> #idea Proposed for #5818: accept and add to sprint
15:36:33 <dkliban> !propose other accept and add to sprint
15:36:34 <pulpbot> dkliban: Proposed for #5818: accept and add to sprint
15:36:38 <bmbouter> +1
15:36:39 <dkliban> there is a PR already
15:36:42 <ttereshc> +1
15:36:44 <dawalker> +1
15:36:44 <ipanova> +1
15:36:46 <fao89> #agreed accept and add to sprint
15:36:46 <fao89> !accept
15:36:46 <pulpbot> fao89: Current proposal accepted: accept and add to sprint
15:36:47 <fao89> #topic https://pulp.plan.io/issues/5813
15:36:47 <pulpbot> fao89: 2 issues left to triage: 5813, 4343
15:36:48 <pulpbot> RM 5813 - jsherril@redhat.com - ASSIGNED - syncing a very large file repository takes a VERY long time
15:36:49 <pulpbot> https://pulp.plan.io/issues/5813
15:36:59 <ttereshc> accept
15:37:00 <dkliban> bmbouter: have you been able to reproduce yet?
15:37:04 <ttereshc> already on the sprint
15:37:07 <dkliban> cool
15:37:09 <dkliban> +1
15:37:11 <ipanova> +1
15:37:14 <bmbouter> not yet the 100k local test ran pretty fast
15:37:14 <fao89> #idea Proposed for #5813: Leave the issue as-is, accepting its current state.
15:37:14 <fao89> !propose accept
15:37:14 <pulpbot> fao89: Proposed for #5813: Leave the issue as-is, accepting its current state.
15:37:27 <fao89> #agreed Leave the issue as-is, accepting its current state.
15:37:27 <fao89> !accept
15:37:27 <pulpbot> fao89: Current proposal accepted: Leave the issue as-is, accepting its current state.
15:37:28 <fao89> #topic https://pulp.plan.io/issues/5701
15:37:28 <pulpbot> fao89: 2 issues left to triage: 5701, 4343
15:37:29 <pulpbot> RM 5701 - dalley - NEW - Performance improvement in remote duplicates
15:37:30 <pulpbot> https://pulp.plan.io/issues/5701
15:37:53 <dalley> originally a refactor, bumping up now since we're running into issues
15:38:13 <ipanova> add to the sprint?
15:38:16 <ttereshc> fao89, did you comment out the remove_duplicates in your experiments?
15:38:19 <dawalker> +1
15:38:37 <ttereshc> fao89, , I believe we discussed that
15:38:41 <fao89> it did not reach that point ttereshc
15:38:45 <ttereshc> right
15:38:51 <ttereshc> so it's crashing earlier
15:38:55 <bmbouter> I'd like to investigate more before we start refactoring
15:38:57 <fao89> yep
15:38:57 <ttereshc> now I remember
15:39:06 <ttereshc> +1 for investigation first
15:39:37 <dalley> +1. it showed up high in the tracemalloc report I thought, though
15:39:39 <ttereshc> can we relate this issue to the performance one?
15:39:56 <ttereshc> to the one fao89 is working on
15:40:04 <ttereshc> and move it back to refactor
15:40:11 <ttereshc> just so it's not lost
15:40:13 <dalley> I'm fine with that
15:40:24 <bmbouter> gathering them up with a tag or epic would be great
15:40:27 <ttereshc> and after investigation it would be clear how much it is affecting the sync
15:41:09 <ipanova> sounds good
15:41:17 <fao89> I'm collecting some good data today, I think by the end of the day I will see how remove_duplicates is behaving
15:41:51 <dalley> fao89, did the tracemalloc report come from a smaller, successful sync?
15:42:27 <fao89> tracemalloc came from a successful sync with 23gb of memory
15:43:24 <dalley> ah, I understand now. +1
15:43:46 <fao89> from what I've seen till now, my mistake was to think it would be a root cause for memory problem, now I realize there are several root causes
15:43:51 <dalley> to moving to refactor, relating them
15:44:42 <fao89> I can say that till now, all the guesses were right, some of them reduced a little some reduced a lot
15:45:56 <fao89> #idea Proposed for #5701: change to refactor and relate to 5688
15:45:56 <fao89> !propose other change to refactor and relate to 5688
15:45:56 <pulpbot> fao89: Proposed for #5701: change to refactor and relate to 5688
15:46:06 <fao89> #agreed change to refactor and relate to 5688
15:46:06 <fao89> !accept
15:46:06 <pulpbot> fao89: Current proposal accepted: change to refactor and relate to 5688
15:46:08 <pulpbot> fao89: 1 issues left to triage: 4343
15:46:08 <fao89> #topic https://pulp.plan.io/issues/4343
15:46:09 <pulpbot> RM 4343 - dalley - NEW - As a user, I can see what parameters were involved in a task
15:46:10 <pulpbot> https://pulp.plan.io/issues/4343
15:46:35 <bmbouter> convert ot story
15:46:47 <fao89> #idea Proposed for #4343: convert to story
15:46:47 <fao89> !propose  other convert to story
15:46:47 <pulpbot> fao89: Proposed for #4343: convert to story
15:46:55 <dkliban> +1
15:47:27 <dawalker> yes, but as it was converted to get it brought up here, what else do we think?
15:47:41 <dalley> I was thinking it's a good 3.1 candidate
15:48:00 <dawalker> +1
15:48:01 <dalley> the later we do it the more lost history there will be
15:49:50 <bmbouter> I'm not totally convinced until a feature is saying they need it also
15:50:04 <bmbouter> s/feature/user/
15:51:07 <bmbouter> also this type of feature has challenges when some params need to not be displayed for some tasks
15:51:17 <bmbouter> for security reasons
15:52:14 <dkliban> yeah
15:52:17 <dalley> +1 to convert to story and wait
15:52:25 <dkliban> cool
15:52:30 <dkliban> +1
15:52:54 <bmbouter> to move this forward I think we need a way to mark some tasks as opting out specific params or entirely via a decorator
15:52:56 <fao89> #agreed convert to story
15:52:56 <fao89> !accept
15:52:56 <pulpbot> fao89: Current proposal accepted: convert to story
15:52:57 <pulpbot> fao89: No issues to triage.
15:53:08 <bmbouter> and we should do a 3.1 call here later in Dec like 3rd week
15:53:24 <fao89> Open floor!
15:53:50 <bmbouter> I'm going to prepare the pulpcore and pulp_file RCs today
15:53:53 <dkliban> i've added a #pulp-meeting channel. do we want to move triage to there?
15:54:01 <bmbouter> +1
15:54:14 <dkliban> we need to update the website/docs
15:54:21 <dkliban> and send an email
15:54:28 <dkliban> i can do that
15:54:34 <bmbouter> sweeeeeet
15:54:36 <bmbouter> do that
15:55:06 <dkliban> bmbouter: do we have any outstanding PRs that need to be merged before you can make an RC?
15:55:45 <bmbouter> just these two which I'm waiting on travis for https://github.com/pulp/pulpcore/pull/420 https://github.com/pulp/pulp_file/pull/325
15:56:03 <bmbouter> I also need a decision on the full_clean() discussion
15:58:04 <dawalker> gmbnomis linked to some good discussion on the pulp-dev email thread.
15:58:33 <dkliban> what i gathered from all that is we shold not call full_clean() during save
15:59:21 <dkliban> so i think it's up to the plugin to do validation before saving the model
15:59:30 <dkliban> be it with a serializer or however
15:59:44 <bmbouter> this is also my takeaway
16:00:27 <bmbouter> we are using DRF and it's providing validation for us, and we express this validation in the serializers
16:01:03 <dkliban> yeah ... i am having a deja vu moment right now
16:01:11 <bmbouter> so generally having the model "also provide some portion of validation" weakens the design by spreading it out
16:02:02 <dkliban> +1 ... are we not always able to use a serializer?
16:02:42 <bmbouter> there are codepaths that don't want to involve the serializer atm, e.g. the data exchange between createrepo_c and pulp_rpm's models
16:03:07 <bmbouter> drf architecutrally expects data to flow in from the api
16:04:00 <bmbouter> so in cases where that isn't the case we kind of fall back to the django-based model validation and we should also call full_clean there if plugin devs need it
16:08:13 <fao89> #endmeeting