15:03:16 <e0ne> #startmeeting horizon
15:03:17 <openstack> Meeting started Wed Jul 31 15:03:16 2019 UTC and is due to finish in 60 minutes.  The chair is e0ne. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:03:18 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:03:21 <openstack> The meeting name has been set to 'horizon'
15:03:55 <e0ne> hi all. is anybody around for horizon meeting?
15:03:59 <rdopiera> o/
15:04:42 <e0ne> amotoki: are you around?
15:04:48 <amotoki> o/
15:05:38 <e0ne> #topic Notices
15:05:46 <e0ne> nothing special for me today
15:06:05 <e0ne> just a friendly reminder that we passed train-2 milestone
15:06:33 <e0ne> we're about one month before train-3 and feature freeze
15:06:34 <e0ne> #link https://releases.openstack.org/train/schedule.html
15:07:42 <e0ne> #topic Shanghai PTG
15:08:00 <e0ne> is anybody going to attend Summit and PTG this fall?
15:08:14 <e0ne> The PTG will be held Tuesday, November 5th through Friday, November 8th in Shanghai, China
15:08:53 * rdopiera is going
15:08:54 <e0ne> vishalmanchanda: hi
15:09:05 <vishalmanchanda> hi all
15:09:08 <amotoki> precisely, from Tuesday afternoon till Friday
15:09:31 <e0ne> I'm not 100% sure, but I hope that I'll manage to be there
15:09:58 <rdopiera> well, I need to get my visa, so not 100% sure either
15:10:24 <e0ne> at least, I can confirm Foundation team that we need some place for Horizon there
15:10:38 <e0ne> rdopiera: I need to get a new visa too
15:11:47 <amotoki> I think OSF is trying to check which teams would like to have PTG sessions at Shanghai now. the beginning of Aug is one of milestones
15:11:56 <e0ne> I'll create an etherpad for ideas to gather ideas to discuss there
15:12:18 <e0ne> amotoki: yes, it's  August 11th
15:15:26 <e0ne> seems that we can go forward to the next topic
15:16:17 <e0ne> do we have anything specific or it's time for Open Discussion?
15:17:15 <e0ne> I don't have anything more to discuss
15:18:13 <amotoki> we passed Train-2 milestone
15:18:35 <e0ne> I mentioned it earlier
15:18:37 <amotoki> it looks like time to check our priorities and their status.
15:19:00 <e0ne> #link https://etherpad.openstack.org/p/horizon-release-priorities
15:19:20 <e0ne> #topic Project priorities for Train
15:19:43 <rdopiera> I added one blueprint I was working on in there, that I failed to set to 'accepted' before, so it wasn't included in the planning (but we did discuss it on a meeting previously)
15:19:47 <e0ne> I proposed keystone v2 removal today
15:19:49 <e0ne> #link https://review.opendev.org/#/c/673807/
15:19:51 <rdopiera> not sure if that's fine?
15:20:09 <e0ne> the next patch will be removing mox from openstack_auth
15:20:35 <rdopiera> I love the smell of removed code in the morning
15:20:49 <amotoki> :)
15:20:49 <vishalmanchanda> e0ne: +1
15:20:52 <vishalmanchanda> :)
15:20:54 <e0ne> rdopiera: morning? did you relocated to US?
15:21:13 <e0ne> rdopiera: what blueprint do you mean?
15:21:13 <rdopiera> figurative morning
15:21:22 <rdopiera> e0ne: the expired password one
15:21:23 <amotoki> rdopiera: your patch sounds a good one.
15:21:32 <e0ne> rdopiera: :)
15:21:46 <e0ne> amotoki: +1
15:22:02 <e0ne> I didn't test that patch, just put some minor note about keystone api v2
15:22:14 <rdopiera> I will make one more patch in the series with docs, settings and release notyes for it
15:22:29 <amotoki> hopefully we can see topic names in addition to review links in the priority etherpad. it helps us understand which links mean which topic.
15:22:43 <rdopiera> but I don't have any tests for it -- not sure how to even test that
15:22:51 <e0ne> amotoki: I love that idea!
15:24:28 <amotoki> rdopiera: I am not sure how we can test it in our integration tests, but unit tests can cover it.
15:24:46 <e0ne> amotoki: +1
15:26:25 <amotoki> who would like to move django 2.2 work forward in Train?
15:26:25 <e0ne> according to the django 2.2
15:26:46 <e0ne> I'll check if we can have the same code for 1.11 and 2.2
15:27:24 <e0ne> if not, I'll  remove WIP from my patch and leave py2django22 job in non-voting mode for now
15:28:17 <amotoki> Previously we use if clause to check django version if we could not use the same code
15:28:53 <e0ne> in U release we can drop py27 support so django1.11 support won't be needed to
15:29:18 <e0ne> so I prefer to wait one more release and do not introduce a lot of if-else statements in the code
15:30:00 <e0ne> 2.2 is the next LTS after 1.11 so it could be possible to have the same codebase
15:30:01 <amotoki> so, it means we need a release note on deprecating py27 in Train at least
15:30:18 <e0ne> TBH, I didn't check it and didn't read release notes
15:30:29 <e0ne> amotoki: good point
15:32:01 <e0ne> #link https://governance.openstack.org/tc/resolutions/20180529-python2-deprecation-timeline.html
15:33:10 <amotoki> yeah, I know that, but not all folks are aware of that. It is worth mentioned in individual projects too
15:33:31 <e0ne> #link https://governance.openstack.org/tc/goals/train/python3-updates.html#non-goals
15:33:31 <amotoki> you are faster than me to find the link :)
15:33:42 <e0ne> amotoki: :)
15:34:07 <e0ne> amotoki: I'm ok to explicitly deprecate python2 in Train
15:34:20 * e0ne wants to remove six
15:35:05 <amotoki> hehe
15:35:08 <amotoki> regarding deprecation notices, I am planning to deprecate enable_quotas in OPENSTACK_NEUTRON_NETWORK settings in Train (which is related to bug 1838522)
15:35:09 <openstack> bug 1838522 in OpenStack Dashboard (Horizon) "Disabled quotas break floating IPs" [Undecided,New] https://launchpad.net/bugs/1838522
15:35:37 <amotoki> I added "Deprecation notices in Train" section in https://etherpad.openstack.org/p/horizon-release-priorities
15:37:11 <e0ne> amotoki: thanks
15:40:01 <amotoki> rdopiera: regarding your blueprint on password change, is it better we check the detail of bug 1791111 in detail? you reopened keystone status and it was marked as invalid again.
15:40:02 <openstack> bug 1791111 in OpenStack Dashboard (Horizon) "allow change password upon first use as user" [High,Confirmed] https://launchpad.net/bugs/1791111 - Assigned to Radomir Dopieralski (deshipu)
15:41:10 <amotoki> I am not familiar with that, but at a glance it seems we need to check it in the API level.
15:41:27 <rdopiera> amotoki: yeah, but now I just parse their error message to get the user_id
15:41:33 <rdopiera> amotoki: so I don't care
15:42:19 <rdopiera> it's ugly, but what can you do when your API is designed for CLI
15:42:56 <amotoki> rdopiera: hmm..... I don't like to parse error message to retrieve some important information..... don't we need to raise this to keystone folks?
15:43:13 <rdopiera> amotoki: well, I did, they closed it
15:43:40 <amotoki> perhaps they didn't captured the point well....
15:43:46 <rdopiera> perhaps
15:44:07 <rdopiera> maybe I can find them at the ptg and bash their heads against a wall until they do
15:44:37 <rdopiera> but that's too much work
15:44:37 <vishalmanchanda> rdopiera: ::)
15:44:54 <amotoki> sounds a good approach unless we haven't found a good solution till then
15:45:22 <rdopiera> I think the error messages are a part of the API
15:45:42 <rdopiera> so parsing them, aparat from being a bit ugly in the code, is a valid way of getting information
15:46:02 <rdopiera> it's not like they are going to localize them
15:46:28 <amotoki> I am okay to skip localization on it.
15:46:42 <amotoki> it is not a topic specific to this one
15:47:41 <rdopiera> what I meant is that it's not going to break suddenly depending on the settings
15:47:56 <rdopiera> so it's not as fragile as it might seem
15:48:30 <rdopiera> I have to parse the error message anyways, since the error code is generic, so I can't tell that error apart from others
15:48:33 <amotoki> rdopiera: I see. you are talking about is  parsing a message
15:49:45 <rdopiera> so since I'm parsing the error message anyways, I can as well get the user_id from it
15:50:58 <amotoki> i haven't understood the whole picture including why we need used_id or what is a problem on the current keystone API. I will try to understand the whole picture next week.
15:51:50 <e0ne> I'm back
15:53:17 <amotoki> e0ne: rdopiera and I talked about the password expiration blueprint. it is done.
15:53:26 <e0ne> great!
15:53:29 <rdopiera> amotoki: the thing is, when you login with horizon, you provide user name and password, then you get the password expired error, including your user_id, and then you use the API to change your password, but you have to provide user_id, old password and new password
15:53:47 <rdopiera> so you need to get the user_id from the error
15:53:59 <rdopiera> because keystone has not other api for that
15:54:26 <amotoki> ah, got it
15:55:34 <amotoki> e0ne: can we check the status of train community goals?
15:55:41 <e0ne> sure
15:55:50 <amotoki> re PDF one, I signed up on it.
15:55:50 <e0ne> mox => mock migration
15:56:09 <amotoki> re ipv6, we need to check https://review.opendev.org/#/c/671910/
15:56:27 <e0ne> I'll push a patch this week. just removed keystone v2 code, so I no need fix mox tests for keystone v2
15:57:33 <amotoki> re: py3 test runtimes for train, we already use openstack-python3-train-jobs in .zuul.yaml and the jobs succeed, so it looks like completed.
15:57:40 <e0ne> +1
15:59:07 <amotoki> on PDF doc one https://storyboard.openstack.org/#!/story/2006084, I marked horizon related repos except horizon itself as invalid
15:59:20 <amotoki> as we don't provide documents
15:59:41 <e0ne> +1
15:59:44 <e0ne> unfortunately, we're out of time:(
15:59:53 <amotoki> e0ne: can we recheck it?
16:00:12 <e0ne> recheck what?
16:00:16 <amotoki> double check
16:00:47 <e0ne> I read throw all tasks in that story and I'm ok with it
16:00:59 <amotoki> e0ne: okay, thanks
16:01:25 <e0ne> we can continue in #openstack-horizon if there is something more
16:01:27 <e0ne> #endmeeting