15:00:42 <dmcbride> #startmeeting OPNFV Colorado release daily
15:00:42 <collabot> Meeting started Thu Sep 22 15:00:42 2016 UTC.  The chair is dmcbride. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:42 <collabot> Useful Commands: #action #agreed #help #info #idea #link #topic.
15:00:42 <collabot> The meeting name has been set to 'opnfv_colorado_release_daily'
15:00:55 <dmcbride> #topic roll call
15:01:05 <dmcbride> #info David McBride
15:01:27 <fzhadaev> #info Fedor Zhadaev
15:01:57 <ChrisPriceAB> #info Chris Price
15:02:12 <dmcbride> what do you know?  RELEASE DAY!!!!
15:02:44 <jose_lausuch> https://m4i3.files.wordpress.com/2015/05/cartoon_releaseday.jpg?w=990
15:02:50 <dmcbride> ChrisPriceAB: did we get the scenario documentation from narindergupta for the lxd scenarios?
15:03:11 <narindergupta> dmcbride, yes i did ch3cked in the patch for that
15:03:28 <ChrisPriceAB> Haha jose_lausuch, your night in planned then!
15:03:40 <fdegir> #info Fatih Degirmenci
15:04:09 <jose_lausuch> ChrisPriceAB: :) no more testing after tagging!
15:04:23 <narindergupta> #info Narinder Gupta
15:04:29 <ChrisPriceAB> #info Yes, all Colorado 1.0  documents are in place
15:04:32 <trozet> #info Tim Rozet
15:04:41 <morgan_orange> #info Morgan Richomme
15:04:48 <kubi001> #info kubi
15:04:48 <dmcbride> ChrisPriceAB: coolio!
15:04:51 <bryan_att> #info Bryan Sullivan
15:05:11 <dmcbride> aricg: ping
15:06:22 <narindergupta> dmcbride, http://artifacts.opnfv.org/joid/docs/scenarios_os-nosdn-lxd-ha/index.html
15:06:31 <narindergupta> dmcbride, http://artifacts.opnfv.org/joid/docs/scenarios_os-nosdn-lxd-noha/index.html
15:06:41 <dmcbride> ChrisPriceAB: the only issues I'm aware of are tagging problems reported by mbeierl and Fu Quiao
15:06:56 <mbeierl> #info Mark Beierl
15:06:57 <jmorgan1> #info Jack Morgan
15:07:06 <ChrisPriceAB> Yes, we have a workaround for that so it seems OK.
15:07:09 <mbeierl> dmcbride: Tagging was done by aricg of my behalf
15:07:19 <dmcbride> mbeierl: thanks for the update
15:07:28 <ChrisPriceAB> I assume there may be some projects that fail to tag, and we have aricg for that.  :)
15:07:38 <mbeierl> correct :)
15:07:43 <dmcbride> ChrisPriceAB: yes
15:07:53 <trozet> Apex hasnt tagged yet, will do so soon
15:08:35 <dmcbride> mbeierl: the last time I looked, you still had a number of unresolved JIRA issues assigned to Col 1.0
15:08:49 <ChrisPriceAB> I assume we unleash aricg on unsuspecting projects in an hour or so and claim completion in 2 hours or so?
15:08:54 <mbeierl> Right - I keep forgetting that you track subtasks
15:09:00 <dmcbride> ditto for morgan_orange with functest
15:09:09 <mbeierl> I'm used to tracking Stories and Bugs only
15:09:25 <mbeierl> subtasks are usually assumed to follow the story they belong to
15:10:19 <dmcbride> mbeierl: well... if the subtask is assigned Col 1.0 and Col 1.0 is done, does it really matter whether its a subtask, or not?  Still unresolved and assigned to a finished release.
15:10:54 <dmcbride> ok - so that leaves Fu Quiao
15:10:56 <mbeierl> dmcbride: what I mean is I had to manually assign all my subtasks to Colorado 1.0 earlier because you include them in your report
15:11:12 <mbeierl> so now I have to manually update all of them with a bulk update in a query
15:11:40 <dmcbride> mbeierl: you need something to do with your time, anyway ;-)
15:12:11 <dmcbride> trying to raise aricg to see if he has an update on status of Fu Quiao
15:12:12 <mbeierl> dmcbride: I need to comply with the standards, I just keep forgetting to do so :)
15:13:03 * mbeierl looks and cannot see any issues assigned to Colorado 1.0 anymore ;)
15:13:08 <mbeierl> for StorPerf
15:13:24 <aricg> dmcbride: Im here
15:13:30 <dmcbride> excellent
15:13:48 <dmcbride> aricg: did you see the email from Fu Quiao reporting problems with tagging?
15:14:36 <jmorgan1> dmcbride: aricg was not on the email thread
15:14:38 <aricg> avaliability?
15:14:50 <aricg> he said he tagged alright
15:15:14 <aricg> all the users who miss tag, or some that tagged long ago are being directed to tag as 1.0.1
15:15:19 <jmorgan1> aricg: he tagged master not stable branch
15:15:30 <aricg> I added a note about this on the tagging page
15:15:31 <mbeierl> can't we just delete the tag in git?
15:15:37 <mbeierl> I thought that was allowed
15:15:53 <aricg> you can, but its not a good idea.
15:15:58 <mbeierl> ok, fair enough
15:16:00 <aricg> its covered on the tagging page
15:16:04 <fdegir> that tag on master will cause issues later on
15:16:28 <dmcbride> yeah - agree
15:16:55 <aricg> fdegir: agreed as well, but removing it will casue a conflict in any clone that had pulled in the deleted tag
15:17:02 <dmcbride> aricg: can you do the git-fu to place the tag on the correct branch?
15:17:09 <fdegir> thats right aricg
15:17:19 <aricg> well, you just tag the correct branch 1.0.1
15:17:33 <dmcbride> aricg: works for me
15:17:55 <aricg> and If we have to go through and do some cleaning, we need to see how much cleaning there is and asses the impact or breaking all the clones.
15:17:59 * ChrisPriceAB suggest we drop a "development" tag on master after release to clear up at least the docs version labelling.
15:18:01 <fdegir> danube comes and all the availability docs will have colorado tag in them
15:18:11 <fdegir> until we tag danube
15:18:16 <jmorgan1> i just checked availability project and i don't see tag on master, just colorado.1.0.1
15:18:24 <jmorgan1> on stable/colorado
15:18:28 <aricg> we can tag danube 0.9? at the start?
15:18:41 <aricg> not that people wont miss tag that. and the cycle repeats
15:18:56 <fdegir> aricg: that works
15:19:12 <aricg> perhaps, I should do the tagging in the future.
15:19:33 <fdegir> with a simple script on stated day/time
15:19:54 <aricg> I don't see an ideal answer, unless we can just handle the case where the proper tag is the highest version of 1.0 or 1.0.x
15:20:07 <aricg> fdegir: yeah.
15:20:26 <aricg> I just need to add myself as a tmp committer on all repos
15:20:34 <aricg> (via all-projects)
15:20:40 <fdegir> i see where this discussion might end up so i stop
15:21:23 <dmcbride> yes - I don't want to linger on the git details too much
15:21:37 <dmcbride> any other pressing issues we need to discuss?
15:21:49 <dmcbride> it seems we have documentation in order
15:21:59 <dmcbride> tagging is more or less complete
15:22:30 <dmcbride> aricg: have you checked to see if any projects have not yet tagged?  We might want to keep a running list.
15:22:44 <bryan_att> when do we expect the release and doc links to go live on the opnfv.org site?
15:22:49 <dmcbride> I think I said the deadline was 12 noon PST
15:22:57 <dmcbride> bryan_att: Monday
15:23:23 <bryan_att> I recommend that all PTLs review the doc links in the interim, and I need to see the patch that I submitted merged somehow
15:23:37 <bryan_att> that's in ChrisPriceAB hands I think
15:24:07 <trozet> dmcbride: I have a question about the "intent to release" column on scenario wiki?
15:24:15 <bryan_att> so afaict docs are not complete
15:24:22 <trozet> don't know why i put a ? on the end of that :)
15:25:07 <ChrisPriceAB> bryan_att, can you check th master version.  I believe your patch replicates that.
15:25:22 <bryan_att> can you drop a link to it?
15:25:39 * ChrisPriceAB wonders is trozet double-negatived his question into a statement....
15:26:14 <ChrisPriceAB> http://artifacts.opnfv.org/opnfvdocs/colorado/docs/documentation/index.html
15:26:31 <trozet> dmcbride: I was just wondering if the intent to release means anything for documentation
15:26:42 <trozet> dmcbride: or is it just for your info
15:26:45 <bryan_att> ChrisPriceAB: OK, looks good. I will abandon the patch.
15:27:46 <ChrisPriceAB> ack
15:29:01 <dmcbride> trozet: we agreed that test results are not a definitive gate for scenarios, so that left us with the good faith assessment of the scenario owner as to whether the scenario should be releaased, or not
15:29:11 <aricg> dmcbride: I will make a list of projects that have tagged
15:29:28 <trozet> dmcbride: ok cool
15:29:31 <trozet> is anyone here from Fuel?
15:29:56 <trozet> narindergupta: you around?
15:29:57 <fzhadaev> me
15:29:59 <dmcbride> trozet: so, the "intent to release" is just a recording of scenario owners assessments
15:30:04 <narindergupta> trozet, yes
15:30:21 <trozet> fzhadaev, narindergupta: we are seeing an ONOS bug where sometimes instances do not get DHCP IP
15:30:31 <trozet> fzhadaev, narindergupta: i see it also has happend on Fuel: https://build.opnfv.org/ci/view/functest/job/functest-fuel-baremetal-daily-colorado/173/console
15:30:57 <trozet> fzhadaev, narindergupta: it looks like an ONOS specific bug to me...was wondering if 1) JUJU has seen it, 2) has it been documented in Fuel scenario release notes?
15:31:25 <narindergupta> trozet, in our deployment we are not seeing recently. Earlier in June we saw this issue. But they have known metadata issue though
15:31:38 <dmcbride> trozet: the only definitive technical gate for scenarios is ability to successfully deploy
15:31:38 <trozet> narindergupta: what version of ONOS do you use?
15:32:24 <trozet> fzhadaev, narindergupta: we are using onos-1.6.0-rc2.tar.gz
15:32:25 <narindergupta> trozet, golden eye
15:32:44 <trozet> narindergupta: not sure what the code name maps to, can you tell me?
15:33:51 <narindergupta> trozet, there charm uses onos-1.6.0.tar.gz
15:34:12 <trozet> narindergupta: ok so we are on the same version...hmm
15:34:20 <narindergupta> 1.6.0-rc4
15:34:27 <trozet> narindergupta: oh rc4
15:34:38 <narindergupta> yeah thats what i can see in deployments
15:34:48 <trozet> narindergupta: thanks for the info, we can tyr switching to rc4 for Colorado 2.0 and see if that fixes it
15:34:58 <narindergupta> sure
15:35:07 <trozet> either way dmcbride: we need to update our release notes with this bug
15:35:46 <dmcbride> trozet: is there a JIRA?
15:35:57 <trozet> dmcbride: we are writing one now, then patching the release notes
15:36:14 <trozet> dmcbride: https://jira.opnfv.org/browse/APEX-281
15:36:35 <dmcbride> trozet: ok - good
15:36:58 <trozet> narindergupta, fzhadaev: fyi it seems to be a race condition - how fast instances are created.  I see this traceback in neutron: https://paste.fedoraproject.org/432623/raw/
15:37:20 <narindergupta> trozet, interesting
15:48:17 <dmcbride> ok - team - thanks for your hard work - congratulations on the release - please keep me up to date on any  issues
15:48:43 <jmorgan1> dmcbride: are you going to have a meeting tomorrow or cancel?
15:48:56 <dmcbride> we will meet one more time on Friday, in case there are any lingering issues with Col 1.0
15:49:05 <dmcbride> no daily meeting next week during ODL
15:49:33 <dmcbride> we will resume the Monday, Thurs, Fri meetings the week after in preparation for Col 2.0
15:49:56 <dmcbride> #endmeeting