14:30:35 <asmacdo> #startmeeting Pulp Triage 2019-08-23
14:30:35 <asmacdo> #info asmacdo has joined triage
14:30:35 <asmacdo> !start
14:30:35 <pulpbot> Meeting started Fri Aug 23 14:30:35 2019 UTC.  The chair is asmacdo. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:30:35 <pulpbot> Useful Commands: #action #agreed #help #info #idea #link #topic.
14:30:35 <pulpbot> The meeting name has been set to 'pulp_triage_2019-08-23'
14:30:35 <pulpbot> asmacdo: asmacdo has joined triage
14:31:06 <dawalker> #info dawalker has joined triage
14:31:06 <dawalker> !here
14:31:06 <pulpbot> dawalker: dawalker has joined triage
14:32:00 <ggainey> #info ggainey has joined triage
14:32:00 <ggainey> !here
14:32:03 <asmacdo> !next
14:32:04 <asmacdo> #topic https://pulp.plan.io/issues/5329
14:32:07 <pulpbot> ggainey: ggainey has joined triage
14:32:08 <pulpbot> asmacdo: 6 issues left to triage: 5329, 5327, 5321, 5311, 5308, 5304
14:32:09 <pulpbot> RM 5329 - Poil - NEW - Pulp 2.20 upgrade chown -R on /var/lib/pulp failed due to NFS mountpoint with Netapp snapshot
14:32:10 <pulpbot> https://pulp.plan.io/issues/5329
14:32:26 <ppicka> #info ppicka has joined triage
14:32:26 <ppicka> !here
14:32:26 <pulpbot> ppicka: ppicka has joined triage
14:32:49 <asmacdo> does the suggestion make sense? maybe we can ask for a PR
14:33:10 <ggainey> is this the chown that we decided to revert?
14:33:17 <ggainey> dkliban: ^^
14:33:51 <bmbouter> #info bmbouter has joined triage
14:33:51 <bmbouter> !here
14:33:51 <pulpbot> bmbouter: bmbouter has joined triage
14:34:33 <asmacdo> i guess lets skip and come back to it
14:34:37 <ggainey> k
14:34:40 <asmacdo> !skip
14:34:41 <asmacdo> #topic https://pulp.plan.io/issues/5327
14:34:41 <pulpbot> asmacdo: 5 issues left to triage: 5327, 5321, 5311, 5308, 5304
14:34:42 <pulpbot> RM 5327 - ttereshc - NEW - psql doesn't work on the pulp2-nightly-pulp3-source-centos7 box
14:34:42 <ggainey> I'm digging, will see what i find
14:34:43 <pulpbot> https://pulp.plan.io/issues/5327
14:35:09 <asmacdo> well this must be irritating
14:35:13 <ggainey> heh
14:35:15 <asmacdo> #idea Proposed for #5327: accept and add to sprint
14:35:15 <asmacdo> !propose other accept and add to sprint
14:35:15 <pulpbot> asmacdo: Proposed for #5327: accept and add to sprint
14:35:22 <ggainey> concur
14:35:47 <asmacdo> #agreed accept and add to sprint
14:35:47 <asmacdo> !accept
14:35:47 <pulpbot> asmacdo: Current proposal accepted: accept and add to sprint
14:35:48 <asmacdo> #topic https://pulp.plan.io/issues/5321
14:35:48 <pulpbot> asmacdo: 4 issues left to triage: 5321, 5311, 5308, 5304
14:35:49 <pulpbot> RM 5321 - daviddavis - NEW - Stop having the installer call makemigrations
14:35:50 <pulpbot> https://pulp.plan.io/issues/5321
14:36:45 <asmacdo> #idea Proposed for #5321: accept, add to sprint
14:36:45 <asmacdo> !propose other accept, add to sprint
14:36:45 <pulpbot> asmacdo: Proposed for #5321: accept, add to sprint
14:36:52 <dawalker> +1
14:36:54 <ggainey> ow, painful - concur
14:36:55 <ppicka> +1
14:37:02 <asmacdo> #agreed accept, add to sprint
14:37:02 <asmacdo> !accept
14:37:02 <pulpbot> asmacdo: Current proposal accepted: accept, add to sprint
14:37:03 <asmacdo> #topic https://pulp.plan.io/issues/5339
14:37:03 <pulpbot> asmacdo: 4 issues left to triage: 5339, 5311, 5308, 5304
14:37:05 <pulpbot> RM 5339 - lmjachky - NEW - Add support for role based access control
14:37:06 <pulpbot> https://pulp.plan.io/issues/5339
14:37:37 <ggainey> is this is a story?
14:37:50 <dawalker> yes
14:37:52 <dawalker> i think so
14:37:54 <asmacdo> more of an epic really
14:38:01 <dawalker> also that ^
14:38:12 <asmacdo> #idea Proposed for #5339: convert to story
14:38:12 <asmacdo> !propose other convert to story
14:38:12 <pulpbot> asmacdo: Proposed for #5339: convert to story
14:38:34 <asmacdo> don't think further action can be taken without a decently large effort to plan
14:38:38 <ggainey> concur, both to story and epic - adding role-based ACLs is...big
14:38:42 <ggainey> concur
14:38:59 <asmacdo> #agreed convert to story
14:38:59 <asmacdo> !accept
14:38:59 <pulpbot> asmacdo: Current proposal accepted: convert to story
14:39:00 <asmacdo> #topic https://pulp.plan.io/issues/5311
14:39:00 <pulpbot> asmacdo: 3 issues left to triage: 5311, 5308, 5304
14:39:01 <lmjachky> yes, it was just an idea; to implement ACL
14:39:02 <dawalker> does this need to be taken to the list?
14:39:02 <pulpbot> RM 5311 - daviddavis - NEW - Hitting /pulp/api/v3/ raises a 500 error
14:39:03 <pulpbot> https://pulp.plan.io/issues/5311
14:39:22 <fabricioo0> #info fabricioo0 has joined triage
14:39:22 <fabricioo0> !here
14:39:22 <pulpbot> fabricioo0: fabricioo0 has joined triage
14:39:30 <asmacdo> i just replicated this
14:39:36 <asmacdo> #idea Proposed for #5311: accept and add to sprint
14:39:36 <asmacdo> !propose other accept and add to sprint
14:39:37 <pulpbot> asmacdo: Proposed for #5311: accept and add to sprint
14:39:38 <ggainey> concur
14:39:41 <dawalker> +1
14:40:07 <asmacdo> #agreed accept and add to sprint
14:40:07 <asmacdo> !accept
14:40:07 <pulpbot> asmacdo: Current proposal accepted: accept and add to sprint
14:40:08 <asmacdo> #topic https://pulp.plan.io/issues/5308
14:40:08 <pulpbot> asmacdo: 2 issues left to triage: 5308, 5304
14:40:09 <pulpbot> RM 5308 - dkliban@redhat.com - NEW - pulp can't handle update version as a string in updateinfo.xml
14:40:10 <pulpbot> https://pulp.plan.io/issues/5308
14:40:26 <asmacdo> #idea Proposed for #5308: move to rpm
14:40:26 <asmacdo> !propose other move to rpm
14:40:26 <pulpbot> asmacdo: Proposed for #5308: move to rpm
14:41:04 <dawalker> +1
14:41:19 <ggainey> yeah, that makes sense - and accept? because versions in rpm-land are historically...filthy. Alas.
14:41:21 <dkliban> #info dkliban has joined triage
14:41:21 <dkliban> !here
14:41:21 <pulpbot> dkliban: dkliban has joined triage
14:41:28 <dkliban> that issue is a regression
14:41:33 <dkliban> it needs to be added to the sprint
14:41:35 <ggainey> even more reason to accept :)
14:41:39 <ggainey> concur
14:41:42 <asmacdo> #idea Proposed for #5308: accept, add to sprint, change to rpm
14:41:42 <asmacdo> !propose other accept, add to sprint, change to rpm
14:41:42 <pulpbot> asmacdo: Proposed for #5308: accept, add to sprint, change to rpm
14:41:58 <asmacdo> #agreed accept, add to sprint, change to rpm
14:41:58 <asmacdo> !accept
14:41:58 <pulpbot> asmacdo: Current proposal accepted: accept, add to sprint, change to rpm
14:41:59 <asmacdo> #topic https://pulp.plan.io/issues/5304
14:42:00 <pulpbot> asmacdo: 1 issues left to triage: 5304
14:42:01 <pulpbot> RM 5304 - dkliban@redhat.com - NEW - Pulp 3 can't publish metadata when installed together with Pulp 2
14:42:02 <pulpbot> https://pulp.plan.io/issues/5304
14:42:15 <dkliban> i was supposed to comment on this
14:42:20 <dkliban> and i did not
14:42:31 <dkliban> we will change how publishing works
14:42:34 <dkliban> in pulp 3
14:43:12 <asmacdo> skip or accept?
14:43:20 <dkliban> accept
14:43:27 <asmacdo> #idea Proposed for #5304: Leave the issue as-is, accepting its current state.
14:43:27 <asmacdo> !propose accept
14:43:27 <pulpbot> asmacdo: Proposed for #5304: Leave the issue as-is, accepting its current state.
14:43:30 <ttereshc> +1
14:43:54 <asmacdo> #agreed Leave the issue as-is, accepting its current state.
14:43:54 <asmacdo> !accept
14:43:54 <pulpbot> asmacdo: Current proposal accepted: Leave the issue as-is, accepting its current state.
14:43:55 <pulpbot> asmacdo: No issues to triage.
14:44:01 <asmacdo> open floor
14:44:22 <ggainey> dkliban: question - is https://pulp.plan.io/issues/5329 the result of the chown we/you reverted?
14:46:47 <asmacdo> we can shift back to ^ whenever. in the meantime,  would it be helpful to add a 2to3 migration redmine tag? seems like a large enough effort that might be useful to see on its own.
14:47:15 <asmacdo> ttereshc^
14:47:21 <dawalker> makes sense to me
14:47:52 <dkliban> ggainey: i am not sure
14:47:56 * dkliban looks
14:48:04 <ggainey> dkliban: I can't remember where that chown was
14:48:18 <ggainey> asmacdo: new tag sounds reasonable to me?
14:48:33 <dkliban> ggainey: yeah ... we reverted this
14:48:39 <dkliban> ggainey: but we never released another version
14:48:57 <ggainey> gotcha - so this will be fixed when 2.21 goes out?
14:49:03 <dkliban> i beieve so
14:49:26 <dkliban> i'll comment on the issue
14:49:41 <dkliban> we should have released a 2.20.1
14:51:00 <ggainey> ahhh, interesting, ok
14:51:25 <asmacdo> any other topics?
14:51:40 <dawalker> Not sure if there's enough folks around right now, but I added this to the headlines for next team meeting and could start talking now
14:52:12 <dawalker> Now that we're not supporting Mariadb, what do we think of changing TextField to JSONField where appropriate in pulpcore?
14:52:21 <dawalker> We have a ticket to do so in pulp_rpm now: https://pulp.plan.io/issues/5326
14:52:25 <asmacdo> dawalker: looks like the title is backwards
14:52:32 <dawalker> but wanted to have a wider discussion to address this elsewhere
14:53:01 <asmacdo> oh nevermind, just read it backwards
14:53:24 <dawalker> :)
14:53:53 <dawalker> Thoughts?
14:54:05 <dawalker> Can also table this until Monday when more folks are around
14:54:08 <dkliban> dawalker: we arleady have a story filed to change to JSON fields
14:54:19 <dkliban> dalley: didn't you find one the other day?
14:54:24 <bmbouter> yeah I've had to go to conflicting meeting so I couldn't be here for much
14:54:35 <dalley> dawalker, there's one for RPM at least
14:54:47 <asmacdo> i think this would be more useful as individual issues for specific fields that should hcnage
14:54:47 <dalley> ..oh, you posted it :)
14:55:03 <dalley> dkliban, I don't know if there's one for core
14:55:30 <dkliban> oh ok
14:55:32 <dawalker> dalley, yup :)
14:56:05 <dkliban> dawalker: where is it needed in pulpcore?
14:56:20 <dawalker> asmacdo, I prefer consistency in handling it if possible, but addressing individually is fine too
14:56:49 <dawalker> dkliban, not sure.  I ran into it with comps.xml where we were dumping lists and dictionaries into TextField.
14:57:00 <asmacdo> dawalker: thats fine, then we need to list those fields on the issue. at the moment, its not really actionable
14:57:01 <dawalker> If you don't need to manipulate them further, JSONField would be better
14:57:11 <dawalker> but if you do need to manipulate them further, it's a moot point
14:57:17 <dalley> dawalker, that's what I get for responding to the last 5 lines of conversation instead of reading first :)
14:58:24 <dawalker> asmacdo, we can hunt up each case, but first let's decide if the consistency in changing over is even a goal?
14:58:42 <dawalker> If it's not, no reason to take the time listing them on tickets.
14:59:08 <asmacdo> dawalker: by "consistency" do you mean to change all text fields?
14:59:20 <dawalker> no, not all
14:59:49 <dawalker> more of a consistent strategy in how to handle the ones that had been dumped lists/dictionaries.
15:00:21 <dawalker> models where you see: models.TextField(default='[]')
15:00:59 <dawalker> but again, not all, I'm only just learning about this, but it appears to depend on what we intend to do with the data
15:01:08 <dawalker> if we need to manipulate it further or not
15:02:48 <dalley> +1 to change all "text fields storing json dumps" to json fields
15:03:01 <asmacdo> a reasonable proposition :)
15:03:30 <dalley> we've got some of those in Pulp Python also I think
15:04:08 <asmacdo> so maybe this issue should be a "contributor docs" issue
15:04:15 <ppicka> +1 to change (less and nicer code)
15:04:15 <dawalker> possibly?
15:05:35 <asmacdo> more on that? or other issues?
15:06:33 <dawalker> i'm ready to move on
15:07:06 <asmacdo> :q
15:07:10 <asmacdo> erm no
15:07:12 <asmacdo> #endmeeting
15:07:12 <asmacdo> !end