18:02:24 <lbragstad> #startmeeting keystone
18:02:25 <openstack> Meeting started Tue Sep  1 18:02:24 2015 UTC and is due to finish in 60 minutes.  The chair is lbragstad. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:02:26 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
18:02:28 <openstack> The meeting name has been set to 'keystone'
18:02:33 <bknudson> hi
18:02:42 <lbragstad> #topic Slow unit tests bknudson
18:02:44 <lbragstad> bknudson: o/
18:02:45 <morgan> O/
18:02:52 <bknudson> the unit tests slowed down a lot
18:03:04 <bknudson> so now it takes 15 mins where it used to take 5 mins
18:03:19 <bknudson> and it's because many tests are written incorrectly
18:03:35 <gyee> \o
18:03:37 <bknudson> they're setting up for REST calls even when most of them don't need to
18:03:51 <marekd> bknudson: but why the the time increased 3x just now?
18:03:53 <bknudson> so I've proposed a few patches already to get rid of some REST testing
18:04:02 <ayoung> bknudson, ++
18:04:20 <ayoung> bknudson, what about the sql stuff?  Is that still slowing things down, too?
18:04:21 <lbragstad> I think this is related to us using entry points
18:04:26 <gyee> bknudson, rest testing going to move to func tests right?
18:04:32 <dstanek> marekd: it's because i moved the paste.ini to use entry points
18:04:34 <lbragstad> #link http://git.openstack.org/cgit/openstack/keystone/commit/?id=4bda26a7e391129366fcfdde2fe84258f6b7493a
18:04:42 <morgan> ayoung: SQL is pretty sane arm
18:04:46 <morgan> Atm
18:04:59 <ayoung> morgan, we doing mysql or sqlite?
18:05:10 <lbragstad> there is a fix up to do caching, too
18:05:14 <lbragstad> #link https://review.openstack.org/#/c/219323/
18:05:19 <bknudson> using entry points for paste is a legitimate change. I don't think we should blame that change for it.
18:05:41 <bknudson> the tests were incorrect to begin with. the change to use entry points just exposed that we're doing testing incorrectly
18:05:53 <morgan> ayoung: SQLite in unit
18:05:56 <ayoung> bknudson, we should judge that empirically, though.  If the entry points are slowing things down, it should be easy to determine
18:06:03 <dstanek> the fix makes the tests runs in the same time as they were before
18:06:14 <ayoung> morgan, in memory?
18:06:34 <morgan> ayoung: yes
18:06:37 <ayoung> k
18:06:45 <lbragstad> bknudson: do you have a change branch for all the tests your fixing?
18:07:05 <lbragstad> s/change branch/topic branch/
18:07:07 <bknudson> lbragstad: y, topic:bug/1490160
18:07:16 <lbragstad> bknudson: perfect, thanks
18:07:29 <bknudson> I haven't made a whole lot of progress but that's where I'll make the changes
18:07:52 <lbragstad> #link for anyone who wants to help - https://bugs.launchpad.net/keystone/+bug/1490160
18:07:53 <openstack> Launchpad bug 1490160 in Keystone "Unit tests are super slow" [Undecided,In progress] - Assigned to Brant Knudson (blk-u)
18:08:13 <bknudson> I'm thinking about setting up a hackathon to get the tests under control.
18:08:24 <lbragstad> bknudson: ++ do you have a date in mind?
18:08:26 <bknudson> wouldn't be this week since I'm at the security mid-cycle
18:08:35 <bknudson> maybe next week?
18:08:45 <bknudson> if anybody else is interested...
18:08:55 <lbragstad> bknudson: maybe send out a couple proposals to the ML?
18:09:03 <gyee> what hackathon?
18:09:21 <lbragstad> bknudson: if you can get a "title" for it, I'll have more leverage to work on it
18:09:30 <bknudson> lbragstad: I can send out a proposal to the ML
18:09:56 <bknudson> gyee: a hackathon to work on the tests and get them to do less REST testing
18:10:07 <gyee> I mean you are running it in #openstack-keystone irc?
18:10:14 <bknudson> essentially to pay down the technical debt
18:10:29 <bknudson> gyee: #openstack-keystone or maybe a new channel just for it?
18:10:45 <gyee> bknudson, k, I'll see I can get some folks to join from here as well
18:10:55 <bknudson> either way works for me. I don't want to get disrupted by other chatter on openstack-keystone
18:12:23 <lbragstad> bknudson: anything else on the tests?
18:12:29 <bknudson> nope, that was it.
18:12:33 <lbragstad> #topic Reseller htruta, henrynash
18:12:46 <lbragstad> htruta: o/
18:13:43 <lbragstad> well we can circle back on that
18:13:53 <lbragstad> #topic critical bugs
18:13:55 <lbragstad> dstanek: o/
18:14:01 <morgan> lbragstad: they just want reviews
18:14:13 <lbragstad> morgan: sounds good
18:14:58 <lbragstad> I was working a bit with dstanek on these yesterday;
18:15:00 <dstanek> anyone have objections to closing those bugs out?
18:15:06 <lbragstad> #link https://bugs.launchpad.net/keystone/+bug/1434034
18:15:07 <openstack> Launchpad bug 1434034 in Keystone juno "Disabling users & groups may not invalidate previously-issued tokens" [Critical,In progress] - Assigned to Morgan Fainberg (mdrnstm)
18:15:30 <lbragstad> and #link https://bugs.launchpad.net/keystone/+bug/1459828 and #link https://bugs.launchpad.net/keystone/+bug/1466485
18:15:32 <openstack> Launchpad bug 1459828 in Keystone "keystone-all crashes when ca_certs is not defined in conf" [Critical,Incomplete]
18:15:32 <bknudson> what do you mean by closing out?
18:15:33 <openstack> Launchpad bug 1466485 in grenade "keystone fails with: ArgsAlreadyParsedError: arguments already parsed: cannot register CLI option" [Undecided,In progress] - Assigned to Ihar Hrachyshka (ihar-hrachyshka)
18:15:55 <bknudson> don't fix them?
18:15:58 <gyee> as Working-as-Expected? :)
18:16:19 <dstanek> bknudson: that's what i'm asking :-) some of the (i think) don't have any Keystone work
18:16:30 <dolphm> lbragstad: i think we need to document the behavior
18:16:36 <dolphm> lbragstad: no code fix, iirc
18:17:02 <lbragstad> dolphm: ++
18:17:15 <dolphm> and no OSSN work has been done, which might be sufficient documentation?
18:17:22 <lbragstad> for https://bugs.launchpad.net/keystone/+bug/1434034 for sure
18:17:23 <openstack> Launchpad bug 1434034 in Keystone juno "Disabling users & groups may not invalidate previously-issued tokens" [Critical,In progress] - Assigned to Morgan Fainberg (mdrnstm)
18:17:27 <dstanek> i just wanted to raise awareness since these are marked as critical
18:17:52 <bknudson> we'll set aside some time at the ossg meetup to work on ossns.
18:18:03 <samueldmq> hello all, sorry I am late
18:18:08 <morgan> That one can be downgraded to high
18:18:10 <bknudson> another option is to mark them not critical
18:18:16 <morgan> Or medium
18:18:18 <dolphm> bknudson: cool; is there something we should have done on the keystone side to keep the OSSN ball moving?
18:18:43 <bknudson> dolphm: no, I just think ossg folks haven't had time to work on the ossns.
18:18:54 <dolphm> morgan: Won't Fix in keystone, but Medium OSSN?
18:20:00 <dolphm> (although i'm not 100% sure how OSSN uses priorities)
18:20:39 <morgan> dolphm: yah let ossg do things with prop
18:20:42 <bknudson> we'll go through and triage the ossn bugs... there's 14 of them.
18:20:43 <morgan> Prio
18:21:00 <dolphm> okay, i'm going to mark it as Medium across the board, Won't Fix in keystone, and i'll leave a comment
18:21:06 <morgan> Ok
18:21:18 <htruta> lbragstad: looks like I am too late :( was at another meeting
18:21:28 <lbragstad> htruta: no worries, we'll circle back
18:21:40 <htruta> cool
18:22:13 <lbragstad> dolphm: do we need an action item for that?
18:23:41 <lbragstad> dstanek: anything else for critical bugs?
18:24:23 <dstanek> lbragstad: nope, just wanted to raise awareness here
18:24:34 <lbragstad> dstanek: thanks, moving on
18:24:36 <lbragstad> #topic Reseller (round 2!)
18:24:40 <lbragstad> htruta: o/
18:24:42 <dolphm> lbragstad: already DONE
18:24:47 <htruta> lbragstad: o/
18:24:49 <lbragstad> dolphm: thanks!
18:25:04 <dolphm> lbragstad: morgan: bknudson: https://bugs.launchpad.net/keystone/+bug/1434034/comments/58
18:25:05 <openstack> Launchpad bug 1434034 in OpenStack Security Notes "Disabling users & groups may not invalidate previously-issued tokens" [Medium,Confirmed] - Assigned to Dave Walker (davewalker)
18:25:07 <bknudson> not sure how bug/1459828 got to be critical anyways
18:25:38 <htruta> so, guys... as henrynash may've talked to you before, we're running out of time for it... We need to know if we really are going to put reseller in Liberty
18:26:16 <bknudson> htruta: you need at least 2 cores to agree to review the changes.
18:26:25 <htruta> we had a great help from ayoung recently... but we still need some reviews. the patches are much smaller, as dstanek suggested
18:26:46 <htruta> bknudson: yes, I know that... We had henry at most of time
18:27:00 <htruta> and sometimes, dstanek and ayoung
18:27:59 <lbragstad> htruta: do you have link to the topic branch?
18:28:07 <lbragstad> htruta: or a series of commits, since they are smaller now?
18:28:19 <htruta> lbragstad: just a sec
18:28:40 <bknudson> if you can't get at least 2 cores to sign up to review the changes for liberty then it's not likely that it's going to be in liberty.
18:29:18 <htruta> lbragstad: this is the first patch of the chain now: https://review.openstack.org/#/c/213448/
18:29:45 <lbragstad> htruta: so #link https://review.openstack.org/#/q/status:open+project:openstack/keystone+branch:master+topic:bp/reseller,n,z should be everything related to reseller that needs to be reviewed?
18:30:02 <htruta> bknudson: agreed. That's why I'm talking here. To see if there is any other core interested on that
18:30:12 <htruta> lbragstad: https://review.openstack.org/#/q/status:open+project:openstack/keystone+branch:master+topic:bp/add-isdomain-to-token,n,z
18:30:31 <htruta> these ones are strictly related too. they just need a rebase since yesterday's merge
18:31:10 <lbragstad> htruta: so those will require a rebase before reviewing, i assume
18:31:42 <lbragstad> htruta: anything else you want to request reviews for?
18:31:48 <htruta> lbragstad: sure. will do it today
18:32:01 <htruta> lbragstad: that's it for now
18:32:06 <lbragstad> htruta: cool, thanks
18:32:11 <lbragstad> #topic open discussion
18:32:42 <bknudson> ossg meetup is this week... not sure if we'll talk about anything keystone specific
18:32:56 <bknudson> other than we're planning to work on ossn
18:34:24 <bknudson> we'll be working on some bandit improvements
18:34:29 <bknudson> docs and unit tests and stuff
18:34:43 <dolphm> bknudson: what days?
18:34:56 <bknudson> working on bandit this morning...
18:35:04 <bknudson> they were talking about tomorrow afternoon for ossn
18:35:11 <bknudson> the meetup goes today to friday
18:35:15 <bknudson> through friday
18:35:34 <bknudson> PAX was yesterday so there were a bunch of cosplayers walking around
18:36:07 <lbragstad> so, liberty-3 is suppose to be this week, right?
18:36:25 <marekd> Sep 3rd ?
18:36:29 <lbragstad> https://launchpad.net/keystone/+milestone/liberty-3
18:36:33 <lbragstad> I think so,
18:36:38 <dolphm> lbragstad: yes
18:36:50 <dolphm> marekd: now through sept 3
18:37:50 <marekd> dstanek: oh, btw, you were talking about bugs: There is one looong standing and i thnk there is patch ready for a review: so the bug is https://bugs.launchpad.net/keystone/+bug/1291157 and patch https://review.openstack.org/#/c/210456/
18:37:51 <openstack> Launchpad bug 1291157 in Keystone "idp deletion should trigger token revocation" [High,In progress] - Assigned to Marek Denis (marek-denis)
18:38:21 <marekd> if anyone is willing to take a look, that would be great - dstanek already did in earlier rounds :-)
18:38:48 <lbragstad> marekd: I've been meaning to follow back up on that, I'll add it to my queue
18:38:56 <marekd> lbragstad: thanks.
18:38:58 <dstanek> marekd: i'll take a look after this meeting - i think i have it starred alfready
18:39:05 <marekd> dstanek: great
18:39:35 <marekd> are bugs also constrained to libery-3 milestone or they can be approved/landed later?
18:40:19 <lbragstad> marekd: I think we can land those after bug I'm not 100% sure
18:40:28 <lbragstad> if we *have* to
18:40:38 <marekd> lbragstad: that would be good i think.
18:41:28 <lbragstad> blueprints are obviously where the cutoff is
18:41:49 <lbragstad> anyone have anything else to bring up?
18:41:58 <lbragstad> reviews, bug triaging?
18:42:15 <amakarov> https://review.openstack.org/#/c/198418/16
18:42:33 <amakarov> I've done performance testing
18:42:45 <lbragstad> amakarov: ah! yeah I had a question on that.. do we have numbers on what kind of performance that gets us?
18:43:12 <lbragstad> amakarov: I think gyee asked that before, I was just following it up
18:43:18 <amakarov> lbragstad, I've stated the numbers in commit message
18:43:48 <amakarov> is it what you are looking for?
18:44:13 <morgan> I will be cleaning up the milestones today
18:44:14 <lbragstad> amakarov: you stated numbers in the latest commit message?
18:44:17 <morgan> Btw
18:44:34 <bknudson> dirty milestones
18:44:35 <amakarov> load-test is in 16 CR only - if causes tests to fail (to expose prints :) )
18:44:53 <amakarov> lbragstad, no, in 16 only
18:45:03 <amakarov> https://review.openstack.org/#/c/198418/16
18:45:10 <lbragstad> amakarov: ah, got it
18:45:34 <lbragstad> amakarov: thanks
18:45:59 <lbragstad> if no one has anything else; we can probably end a bit early and give everyone a few minutes to do a review :)
18:46:34 <lbragstad> thanks all!
18:46:39 <lbragstad> #endmeeting