10:05:15 <chriadam|windows> #startmeeting CalDAV/CardDAV Test Servers Meeting
10:05:15 <merbot> Meeting started Mon Oct 31 10:05:15 2016 UTC.  The chair is chriadam|windows. Information about MeetBot at http://wiki.merproject.org/wiki/Meetings.
10:05:15 <merbot> Useful Commands: #action #agreed #help #info #idea #link #topic.
10:05:26 <chriadam|windows> #topic Follow up from last meeting
10:06:01 <chriadam|windows> My understanding from last meeting is that a VM had been set up in the Mer infra, and the firewall etc was set up, and community members needed to get in contact with lbt to request access
10:06:14 <chriadam|windows> lbt: what is the status update since then?
10:06:28 <lbt> yep - dr_gogeta86 sent over his details and I setup an account for him
10:06:33 <larstiq> dr_gogeta86 has set up his farm of services
10:06:48 <larstiq> just need to fix exposing them to the world
10:07:09 <chriadam|windows> larstiq: this is dns entries?
10:07:39 <larstiq> chriadam|windows: dns is there, but last bit of firewall config
10:07:44 <chriadam|windows> ok cool
10:07:54 <lbt> however fred.tst.merproject.org works
10:07:54 <chriadam|windows> #topic Next steps
10:08:02 <chriadam|windows> so I guess next steps include:
10:08:06 <lbt> as does any other *.tst.merproject.org - so we have DNA
10:08:08 <lbt> S
10:08:11 <chriadam|windows> 1) larstiq solving the last bit of dns + firewall config
10:08:30 <chriadam|windows> 2) then we need to think about how to expose "reset" switches
10:08:38 <chriadam|windows> 3) then I need to start testing
10:08:40 <chriadam|windows> ?
10:08:47 <chriadam|windows> is that about right, or have I missed something important?
10:09:22 <larstiq> chriadam|windows: 2) I think we can solve as it comes up
10:10:08 <chriadam|windows> larstiq: it's something we'll need to be able to do regularly (e.g., after running a test, if there's a test failure, then the server state might not be left "clean" - so it needs to be reset back to empty database before the next test can be run)
10:10:20 <chriadam|windows> I assume it's something simple (some docker command)?
10:10:35 <larstiq> chriadam|windows: empyting is simple, repopulating maybe a bit more work?
10:10:43 <chriadam|windows> emptying is enough to begin with
10:10:51 <chriadam|windows> repopulating is something which could be done by the unit test itself I guess
10:10:59 <lbt> yep - we probably need some simple UI web service
10:11:05 <larstiq> mayhaps
10:11:21 * larstiq would advocate starting to use it and see what is problematic
10:11:43 <chriadam|windows> maybe to begin with I can just write a script to get all of the data UIDs and send the appropriate DELETE requests to clear them, anyway
10:11:50 <larstiq> sure
10:11:55 <chriadam|windows> shouldn't be tricky, and doesn't require any server/infra stuff
10:11:58 <chriadam|windows> ok, I'll do that
10:12:03 <ljo_> good
10:12:06 <chriadam|windows> #topic Any other discussion
10:12:23 <chriadam|windows> Anything else which needs discussion?
10:12:37 <chriadam|windows> Pity dr_gogeta86 isn't online at the moment, but I can sync up with him later
10:12:47 <larstiq> chriadam|windows: nafaict
10:12:51 <chriadam|windows> with regards to how to add mroe services
10:13:15 <chriadam|windows> e.g. cemoi73 asked about adding COZY instance to the docker compose env
10:13:23 <chriadam|windows> but that's something which can happen later
10:13:33 <larstiq> so
10:13:41 <lbt> do we know how the dockers are setup ?
10:13:44 <larstiq> on the dns level anything under tst.merproject.org will be redirected
10:13:50 <lbt> larstiq: yes
10:14:04 <larstiq> then dr_gogeta86 is running a webserver that decides on the hostname which service to send the request to
10:14:19 <larstiq> so it should be just a matter of adding COZY to the git repo dr_gogeta86 advertised
10:14:31 <chriadam|windows> lbt: dr_gogeta86 pushed the docker config to a git repo, I don't have the link handy but can send it to you.  it just uses images from docker hub, I believe.
10:14:48 <lbt> OK - so my main thought is about how others can add images
10:14:49 <chriadam|windows> but i haven't experimented with it myself, yet- haven't had the time :-(
10:14:54 <chriadam|windows> yes, I agree - that's important
10:14:57 <larstiq> lbt: pull request
10:15:04 <chriadam|windows> hopefully it's just merge requests to a git.merproject.org repo
10:15:07 <lbt> and if we can have multiple instances of any given image running
10:15:11 <larstiq> chriadam|windows: yeah
10:15:27 <larstiq> lbt: to do what, separate db?
10:15:29 <lbt> I'd like to see something like lbt_cozy.tst.merproject.org go to my instance of cozy
10:15:37 <lbt> eventually
10:15:53 <larstiq> that can be done
10:15:54 <lbt> just want to ensure the design doesn't block that too much
10:16:06 <chriadam|windows> I will try to get (or ask dr_gogeta86 for) step by step instructions to add a new service to the docker service, and add those to the wiki
10:16:17 <larstiq> lbt: no, "bottleneck" is the vhost proxying, but if parties agree no problem
10:16:47 <chriadam|windows> great
10:16:54 <chriadam|windows> any other points?
10:17:16 <chriadam|windows> next week's meeting is the full one, not just about test servers, we will include test server topics in it also of course
10:17:32 <chriadam|windows> at 0900 again - does that suit you both, or?
10:17:38 <chriadam|windows> dst causing trouble with that time now?
10:17:52 <lbt> 09:00 onwards is OK for me
10:18:08 <larstiq> chriadam|windows: I'll try :)
10:18:17 <chriadam|windows> cool.  if that changes, let me know and I'll resched
10:18:25 <chriadam|windows> thanks!
10:18:31 <chriadam|windows> if nothing else, will close meeting in 3...
10:18:39 <chriadam|windows> 2...
10:18:49 <chriadam|windows> 1...
10:18:54 <chriadam|windows> #endmeeting