| *** lucas-dinner has quit IRC | 00:03 | |
| *** lucasagomes_ has quit IRC | 00:04 | |
| *** gfidente has quit IRC | 00:05 | |
| *** Goneri has joined #tripleo | 00:10 | |
| *** Marga_ has quit IRC | 00:17 | |
| *** weshay_xchat has joined #tripleo | 00:18 | |
| *** penick has quit IRC | 00:20 | |
| *** Marga_ has joined #tripleo | 00:25 | |
| *** shivrao has quit IRC | 00:47 | |
| *** slagle has joined #tripleo | 00:47 | |
| *** yamahata has quit IRC | 00:50 | |
| *** shivrao has joined #tripleo | 00:50 | |
| *** shivrao has quit IRC | 00:56 | |
| *** Marga_ has quit IRC | 01:22 | |
| *** stendulker has joined #tripleo | 01:26 | |
| *** Goneri has quit IRC | 01:37 | |
| *** julim has joined #tripleo | 01:49 | |
| *** Marga_ has joined #tripleo | 01:49 | |
| *** cody-somerville has joined #tripleo | 02:09 | |
| *** panda has quit IRC | 02:09 | |
| *** panda has joined #tripleo | 02:10 | |
| *** stendulker has quit IRC | 02:10 | |
| *** rhallisey has quit IRC | 02:22 | |
| *** weshay_xchat has quit IRC | 02:31 | |
| *** sthillma has quit IRC | 02:44 | |
| *** yamahata has joined #tripleo | 02:46 | |
| *** yamahata has quit IRC | 02:48 | |
| *** yamahata has joined #tripleo | 02:48 | |
| *** untriaged-bot has joined #tripleo | 03:00 | |
| untriaged-bot | No untriaged bugs so far! \o/ | 03:00 |
|---|---|---|
| *** untriaged-bot has quit IRC | 03:00 | |
| *** tzumainn has quit IRC | 03:27 | |
| *** mgagne has quit IRC | 03:28 | |
| *** rainya has quit IRC | 03:28 | |
| *** jroll has quit IRC | 03:28 | |
| *** gus has quit IRC | 03:29 | |
| *** persia has quit IRC | 03:29 | |
| *** julim has quit IRC | 03:30 | |
| *** rainya has joined #tripleo | 03:30 | |
| *** persia has joined #tripleo | 03:30 | |
| *** persia has quit IRC | 03:30 | |
| *** persia has joined #tripleo | 03:30 | |
| *** tzumainn has joined #tripleo | 03:30 | |
| *** gus has joined #tripleo | 03:31 | |
| *** mgagne has joined #tripleo | 03:31 | |
| *** jroll has joined #tripleo | 03:31 | |
| openstackgerrit | Nisha Agarwal proposed openstack/diskimage-builder: Add a new element hpdsa https://review.openstack.org/220888 | 03:35 |
| *** jroll has quit IRC | 03:41 | |
| *** mgagne has quit IRC | 03:41 | |
| *** gus has quit IRC | 03:41 | |
| *** tzumainn has quit IRC | 03:41 | |
| *** persia has quit IRC | 03:41 | |
| *** rainya has quit IRC | 03:41 | |
| *** persia has joined #tripleo | 03:42 | |
| *** gus has joined #tripleo | 03:44 | |
| *** adam_g has quit IRC | 03:45 | |
| *** shivrao has joined #tripleo | 03:46 | |
| *** david-lyle has joined #tripleo | 03:50 | |
| *** shivrao has quit IRC | 03:51 | |
| *** adam_g has joined #tripleo | 03:51 | |
| *** adam_g has quit IRC | 03:51 | |
| *** adam_g has joined #tripleo | 03:51 | |
| *** shivrao has joined #tripleo | 03:55 | |
| *** tzumainn has joined #tripleo | 03:56 | |
| *** mgagne has joined #tripleo | 04:03 | |
| *** rainya has joined #tripleo | 04:03 | |
| *** jroll has joined #tripleo | 04:04 | |
| *** michchap_ has quit IRC | 04:09 | |
| *** michchap has joined #tripleo | 04:09 | |
| *** cody-somerville has quit IRC | 04:32 | |
| *** yog_ has joined #tripleo | 04:57 | |
| *** tzumainn has quit IRC | 05:00 | |
| *** jaosorior has joined #tripleo | 05:08 | |
| *** lazy_prince has joined #tripleo | 05:11 | |
| *** masco_ has joined #tripleo | 05:15 | |
| *** Marga_ has quit IRC | 05:24 | |
| *** marios has joined #tripleo | 05:49 | |
| *** shivrao_ has joined #tripleo | 05:51 | |
| *** shivrao has quit IRC | 05:51 | |
| *** shivrao_ is now known as shivrao | 05:51 | |
| *** Marga_ has joined #tripleo | 05:52 | |
| *** Marga_ has quit IRC | 05:52 | |
| *** Marga_ has joined #tripleo | 05:53 | |
| *** yog_ has quit IRC | 06:02 | |
| *** panda has quit IRC | 06:09 | |
| *** panda has joined #tripleo | 06:10 | |
| *** egafford has quit IRC | 06:11 | |
| *** shivrao has quit IRC | 06:14 | |
| *** yog_ has joined #tripleo | 06:19 | |
| *** shardy has joined #tripleo | 06:22 | |
| *** jprovazn has joined #tripleo | 06:41 | |
| jprovazn | shardy: good morning | 06:43 |
| *** shadower has joined #tripleo | 06:46 | |
| *** adam_g has quit IRC | 06:55 | |
| shardy | jprovazn: hi! | 06:55 |
| jprovazn | shardy: hi, here are the times of getting resources and events on 50 nodes - https://gist.github.com/jprovaznik/aee95bf6e0056fd4fcce | 06:58 |
| jprovazn | shardy: the script is bellow times | 06:58 |
| shardy | dtantsur|afk: Hey, when you're around can you check if there's a problem with discoverd pacakaging? | 06:59 |
| shardy | http://logs.openstack.org/40/214740/11/check-tripleo/gate-tripleo-ironic-overcloud-f21-nonha/87c9d98/console.html | 06:59 |
| shardy | ERROR:delorean:Error while building packages for openstack-ironic-discoverd | 07:00 |
| shardy | jprovazn: thanks, looking | 07:00 |
| jprovazn | shardy: IOW getting resources takes 50-77s (depending on buffer cache), then getting events time is linear to number of nodes | 07:00 |
| shardy | jprovazn: Well the time is spent getting events for every single resource isn't it? | 07:02 |
| shardy | jprovazn: It would probably be quicker to get all resources per-stack, instead of per-resource | 07:03 |
| jprovazn | shardy: you mean all events per-stack, don't you? | 07:03 |
| jprovazn | shardy: yes, if it's already possible to get nested events for the whole stack, I can improve this on client side | 07:04 |
| shardy | Sorry yes -ENOCOFFEE | 07:04 |
| jprovazn | :) | 07:04 |
| jprovazn | or | 07:04 |
| shardy | jprovazn: Can you time heat resource-list -n5 overcloud && time heat event-list -n5 overcloud? | 07:04 |
| shardy | I can also try running your reproducer locally but I'm already familiar with the heatclient code | 07:05 |
| shardy | the event-list code takes the per-stack approach already | 07:05 |
| *** ryansb_ has joined #tripleo | 07:05 | |
| *** ryansb_ has joined #tripleo | 07:05 | |
| jprovazn | shardy: once mwagner is online I will ask him (+6 TZ) | 07:05 |
| shardy | as mentioned yesterday it should be possible to optimise that via a new flag to the event API, similar to that already in the resource-list API | 07:05 |
| shardy | we'll have to be (very) quick if that's needed for Liberty tho | 07:05 |
| shardy | it's been on my todo list with about 100 other things ;) | 07:06 |
| jprovazn | shardy: I tried event-list -n5 on my local setup and it was slow like a hell yesterday | 07:06 |
| shardy | jprovazn: Yeah, it is slow unfortunately | 07:06 |
| shardy | jprovazn: I need to spend this morning on the recursive validation stuff for jdob but I'll try to carve out an hour this afternoon to look at adding a nested-depth argument to the event list API | 07:07 |
| jprovazn | shardy: thanks | 07:07 |
| shardy | perhaps you can then test it & we can quantify if it's worth pushing to get it into liberty or not | 07:07 |
| jprovazn | shardy: yes, though | 07:08 |
| jprovazn | shardy: supposing we improve event-list to be resurcive | 07:08 |
| *** ryansb has quit IRC | 07:08 | |
| *** ryansb_ is now known as ryansb | 07:08 | |
| jprovazn | then I can save ~20 secs, but 50s-70s for getting resources remains | 07:08 |
| jprovazn | shardy: https://gist.github.com/jprovaznik/aee95bf6e0056fd4fcce#file-script-L22 | 07:09 |
| jprovazn | ^ this one takes the longest time | 07:09 |
| jprovazn | shardy: then getting events for UpdateDeployment resources takes about nodes_num/2 secs | 07:10 |
| shardy | jprovazn: You don't have any print with times around the all_res = heat.resources.list(stack.id, nested_depth=5) ? | 07:10 |
| jprovazn | shardy: I have - it's the first time in each run | 07:11 |
| jprovazn | xxx0 prefix | 07:11 |
| shardy | Oh right, sorry I see it now | 07:11 |
| * shardy really does need more coffee | 07:11 | |
| jprovazn | haha, no worries - to output is not good, I put quickly for mark at the EOD | 07:12 |
| shardy | jprovazn: my point is if you can get a recursive list of all resources, you may not need the recursive resource list anymore | 07:12 |
| shardy | e.g like in the hook-poll heatclient code | 07:13 |
| shardy | that shouldn't be that slow though, I'll try to figure out where the bottleneck is | 07:13 |
| *** yog_ has quit IRC | 07:13 | |
| shardy | how loaded is the undercloud? | 07:13 |
| jprovazn | shardy: it's true I wouldn't have to do it if I'd like to just find out "is there some hook reached" | 07:13 |
| jprovazn | shardy: but for the CLI workflow | 07:13 |
| jprovazn | shardy: I also show which were already completed, which didn't start | 07:14 |
| *** jtomasek has joined #tripleo | 07:14 | |
| jprovazn | shardy: and I need resource-list for this | 07:14 |
| shardy | jprovazn: Ok, cool, well lets take it as two problems and I'll try to reproduce both issues later | 07:14 |
| jprovazn | shardy: these numbers from the scale lab - I don't have access to it but we can check with Mark in the afternoon | 07:15 |
| jprovazn | shardy: ok, thanks | 07:15 |
| *** ukalifon has joined #tripleo | 07:15 | |
| shardy | jprovazn: In future it'd be cool if we could figure out an easy way to reproduce "heat is slow" issues under a profiler, as it would save a lot of time attempting to reproduce locally | 07:16 |
| jprovazn | shardy: +1 | 07:17 |
| *** ukalifon has quit IRC | 07:20 | |
| *** ifarkas has joined #tripleo | 07:26 | |
| *** aufi has joined #tripleo | 07:27 | |
| d0ugal | jprovazn: Is the StackUpdateManager in tripleo-common used for CREATE and UPDATE? | 07:31 |
| jprovazn | d0ugal: for stack update only | 07:31 |
| d0ugal | jprovazn: Thanks - I am trying to figure out the best way to mover over the tripleocommon deploy code | 07:32 |
| d0ugal | s/tripleocommon/tripleoclient/ | 07:32 |
| jprovazn | d0ugal: cool! | 07:33 |
| *** paramite has joined #tripleo | 07:34 | |
| d0ugal | jprovazn: scale.py is just for scaling the different roles? | 07:35 |
| jprovazn | d0ugal: yes | 07:35 |
| d0ugal | update.py is just for updating packages? | 07:35 |
| jprovazn | d0ugal: yes | 07:35 |
| jprovazn | d0ugal: and stack_update for running a general stack-update with breakpoints | 07:35 |
| d0ugal | jprovazn: Then what is stack_update for? :) | 07:35 |
| d0ugal | aha | 07:35 |
| d0ugal | Thanks | 07:35 |
| jprovazn | ^ this one has quite confusing name :/ | 07:35 |
| d0ugal | Gonna add some comments | 07:35 |
| d0ugal | jprovazn: I thought we didn't support scale down yet? | 07:36 |
| jprovazn | d0ugal: we do | 07:37 |
| d0ugal | oh, cool | 07:37 |
| jprovazn | d0ugal: the scale down - you can scale down just by decreasing number or by removing particular nodes | 07:38 |
| jprovazn | tripleo-common implements removing specific nodes (for scaling down be removing any nodes - there is no job to do - just updating stack) | 07:38 |
| *** yog_ has joined #tripleo | 07:38 | |
| *** stendulker has joined #tripleo | 07:47 | |
| *** paramite is now known as paramite|afk | 07:50 | |
| d0ugal | jprovazn: Neat, I see that now. | 07:52 |
| d0ugal | Sorry got distracted by coffee :) | 07:52 |
| jprovazn | d0ugal: same distraction here ;) | 07:52 |
| d0ugal | jprovazn: scaleup seems to assume Tuskar, wont that be an issue? https://github.com/openstack/tripleo-common/blob/master/tripleo_common/scale.py#L49 | 07:54 |
| d0ugal | The others seem to have if self.tuskarclient | 07:54 |
| d0ugal | Sorry for the quite possibly silly questions :) | 07:54 |
| jprovazn | d0ugal: scaleup is obsolete since openstack overcloud deploy takes care of this | 07:55 |
| d0ugal | aha | 07:55 |
| jprovazn | d0ugal: tuskar can be removed in future too :) | 07:55 |
| jprovazn | *tuskar support | 07:55 |
| d0ugal | jprovazn: Yup, can't wait for that \o/ | 07:55 |
| jprovazn | lol | 07:56 |
| *** dtantsur|afk is now known as dtantsur | 07:57 | |
| *** paramite|afk is now known as paramite | 07:57 | |
| *** gfidente has joined #tripleo | 08:00 | |
| *** sdake has quit IRC | 08:07 | |
| openstackgerrit | Merged openstack/puppet-tripleo: Allow a user to specify the syslog address for HAProxy https://review.openstack.org/222595 | 08:08 |
| *** derekh has joined #tripleo | 08:13 | |
| *** trellooobot has joined #tripleo | 08:13 | |
| trellooobot | URGENT TRIPLEO TASKS NEED ATTENTION | 08:13 |
| trellooobot | +------------------------------+-------------------------------+-------------------+-------------+ | 08:13 |
| trellooobot | | Title | URL | Members | Last Active | | 08:13 |
| trellooobot | +------------------------------+-------------------------------+-------------------+-------------+ | 08:13 |
| trellooobot | | Package builds failing in CI | https://trello.com/c/1kCLMvMB | **NEEDS MEMBERS** | 1 min | | 08:13 |
| trellooobot | +------------------------------+-------------------------------+-------------------+-------------+ | 08:13 |
| *** trellooobot has quit IRC | 08:13 | |
| *** lucas-dinner has joined #tripleo | 08:16 | |
| *** lucasagomes has joined #tripleo | 08:16 | |
| *** lucasagomes_ has joined #tripleo | 08:16 | |
| *** jistr has joined #tripleo | 08:18 | |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Pin Delorean https://review.openstack.org/223476 | 08:24 |
| *** pelix has joined #tripleo | 08:31 | |
| *** athomas has joined #tripleo | 08:32 | |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Get a reference to the delorean dependencies https://review.openstack.org/223479 | 08:35 |
| derekh | hey look a passing ha job https://review.openstack.org/#/c/223211/ | 08:36 |
| jistr | !!! | 08:38 |
| openstack | jistr: Error: "!!" is not a valid command. | 08:38 |
| jistr | :) | 08:38 |
| openstackgerrit | Merged openstack-infra/tripleo-ci: Include the puppet-pacemaker environment in the HA job https://review.openstack.org/223211 | 08:40 |
| derekh | jistr: this is fixing probably a completly different problem to the one we had pre last week, for starters in instack now and on centos | 08:40 |
| jistr | derekh: yeah i think so. i'm thinking we got around the problem we had before (with an updated corosync) by switching to centos which probably doesn't have the problematic corosync version | 08:41 |
| derekh | jistr: yup | 08:41 |
| gfidente | jistr, was looking at https://review.openstack.org/#/c/218930/1/puppet/manifests/overcloud_controller_pacemaker.pp,cm | 08:46 |
| gfidente | is memecached in that list for future use? | 08:46 |
| jistr | not sure what you mean by future use? | 08:49 |
| gfidente | I think keystone is not using memecahed | 08:50 |
| gfidente | but it might | 08:50 |
| gfidente | oh it's in the ref arch anyway | 08:51 |
| jistr | ah right. Yes it's mainly to get in sync with the ref arch i'd say, as that's what they test failures on etc. | 08:52 |
| jistr | https://bugzilla.redhat.com/show_bug.cgi?id=1257414 | 08:53 |
| openstack | bugzilla.redhat.com bug 1257414 in openstack-tripleo-heat-templates "[HA] critical resource constraints missing from pacemaker config make things go kaboom" [Urgent,On_qa] - Assigned to jstransk | 08:53 |
| gfidente | very creative in the bug title | 08:55 |
| gfidente | I think we need to remove the start-delay=10s for the nova resources as well | 08:55 |
| gfidente | someone pointed me at that the other day | 08:55 |
| jistr | did we add that for some reason though? | 08:56 |
| jistr | oof i just noticed i rechecked the patches in vain as CI is red | 08:56 |
| jistr | i guess this https://review.openstack.org/#/c/223476/ | 08:57 |
| jistr | could help | 08:57 |
| gfidente | jistr, I couldn't figure where they came frm | 08:57 |
| gfidente | I found those for some resources in older commits of astapor | 08:57 |
| gfidente | (for some nova resources) | 08:57 |
| gfidente | so I think was an unfortunate copy | 08:57 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Use alternative trunk repository https://review.openstack.org/223487 | 09:00 |
| *** untriaged-bot has joined #tripleo | 09:00 | |
| untriaged-bot | Untriaged bugs so far: | 09:00 |
| untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1495860 | 09:00 |
| openstack | Launchpad bug 1495860 in diskimage-builder ""disk-image-create ironic-agent fedora iso -o my-deploy-ramdisk" failed" [Undecided,New] | 09:00 |
| *** untriaged-bot has quit IRC | 09:00 | |
| *** mcornea has joined #tripleo | 09:00 | |
| jistr | gfidente: right, if it's not in ref arch it should probably go away | 09:05 |
| gfidente | jistr, waiting for dprince | 09:06 |
| gfidente | extra eyes https://review.openstack.org/#/c/222497/ would help | 09:06 |
| jistr | ack, looking | 09:07 |
| gfidente | it's something in between hard and impossible to test :( | 09:07 |
| gfidente | I'm trying something like deploy two overclouds, abandoning one and setting static ips like the first for the second | 09:08 |
| spredzy | jistr, have you tried to deploy today ? While running openstack overcloud deploy --templates it kicks me with PKI initilization in init-keystone is deprecated and will be removed ... then it tries to ssh into a machine with heat-admin and go a permiession denied | 09:12 |
| jistr | spredzy: i got permission denied yesterday on a different delorean URL i think, that's what made me switch to what's in CI. After the switch i didn't get to a successful deployment. Instack VM didn't get an IP on previous run, i'm trying again now. | 09:14 |
| pino|work | jistr: https://review.openstack.org/#/c/217511/ | 09:17 |
| *** dtantsur is now known as dtantsur|brb | 09:17 | |
| spredzy | jistr, ack, let me know if you get any success, so far - starting from a fresh machine this morning - most repo/package issue were gone | 09:18 |
| jistr | hmm still the same. instack-virt-setup keeps printing | 09:20 |
| jistr | "instack never got an IP address from the libvirt default network." | 09:20 |
| jistr | spredzy: ^ | 09:20 |
| * jistr will reprovision to clean the env | 09:21 | |
| jistr | i think i got further with this delorean url previously | 09:21 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Rewrite baseurl after packages are built https://review.openstack.org/223479 | 09:21 |
| spredzy | jistr, sudo virsh undefine instack. Make sure you have used this url http://trunk.rdoproject.org/centos7/38/1c/381cac9139096bfef49952f3fd67e19451160b61_4bc2d731/delorean.repo and relaunch instack-virt-setup | 09:23 |
| jistr | spredzy: yeah i virsh udefine all the machines | 09:24 |
| jistr | spredzy: i tried with the URL you posted previously, now i was trying with the URL from the CI, which i thought would work, given that CI works... | 09:25 |
| jistr | spredzy: this https://github.com/openstack-infra/tripleo-ci/blob/master/toci_instack.sh#L164 | 09:25 |
| spredzy | jistr, ack. But weird, I got it working this am | 09:26 |
| jistr | trying again with the 381ca URL | 09:30 |
| *** lucas-dinner has quit IRC | 09:30 | |
| *** lucasagomes_ has quit IRC | 09:30 | |
| *** adrianopetrich has quit IRC | 09:32 | |
| jistr | hmm same result... something's wrong :) i really need to reprovision i guess | 09:37 |
| spredzy | :( | 09:38 |
| spredzy | ack | 09:38 |
| spredzy | On my side for more info, the puppet stuff went fine, the post initialisation setup failed (init-keystone) | 09:38 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Rewrite baseurl after packages are built https://review.openstack.org/223479 | 09:55 |
| *** paramite is now known as paramite|afk | 09:55 | |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Use alternative trunk repository https://review.openstack.org/223487 | 09:57 |
| *** pradk has quit IRC | 09:59 | |
| *** jaosorior has quit IRC | 10:00 | |
| *** chlong has quit IRC | 10:09 | |
| *** chlong has joined #tripleo | 10:11 | |
| *** pradk has joined #tripleo | 10:14 | |
| jprovazn | shardy: you around? | 10:17 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Use alternative trunk repository https://review.openstack.org/223487 | 10:18 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Pin Delorean https://review.openstack.org/223476 | 10:18 |
| jprovazn | shardy: when looking at events code in engine, it seems that I can get all events for a tenant https://github.com/openstack/heat/blob/master/heat/engine/service.py#L1254 | 10:19 |
| jprovazn | shardy: I wonder if it's too crazy to try this for getting all events :) | 10:20 |
| *** dtantsur|brb is now known as dtantsur | 10:21 | |
| jprovazn | (supposing that single OC is supported and events would be deleted with stack) | 10:21 |
| *** julim has joined #tripleo | 10:21 | |
| *** paramite|afk is now known as paramite | 10:21 | |
| shardy | jprovazn: Hmm, I'm not sure if/how you can actually hit that code.. | 10:22 |
| shardy | I suspect the API forces you to provide a stack identity | 10:23 |
| shardy | jprovazn: the other issue is you may find the tenant-wide list includes events for previously (soft) deleted stacks | 10:23 |
| jprovazn | shardy: kk, I expected a blocker like this | 10:24 |
| d0ugal | Is there anything stopping us removing Tuskar support now? | 10:24 |
| shardy | jprovazn: what we need is a new "nested_depth" argument there, and replace the get_all_by_tenant query with a recursive DB lookup | 10:24 |
| openstackgerrit | Dougal Matthews proposed openstack/tripleo-common: Remove the unused scaleup command https://review.openstack.org/223504 | 10:24 |
| jprovazn | shardy: yep | 10:25 |
| shardy | jprovazn: Yeah, I can't see a way to hit that via either the native or CFN API, I suspect it's dead code | 10:27 |
| openstackgerrit | Merged openstack-infra/tripleo-ci: Pin Delorean https://review.openstack.org/223476 | 10:28 |
| jprovazn | shardy: kk, nvm, trying to add nested_depth | 10:28 |
| shardy | jprovazn: Ok, cool - you can refer to jdob's recursive validation patch to see what needs changing to update the API & RPC API | 10:29 |
| shardy | https://review.openstack.org/#/c/219016/ | 10:29 |
| shardy | basically the same as his wiring in show_nested | 10:29 |
| *** wendar has quit IRC | 10:31 | |
| *** Hunner has quit IRC | 10:31 | |
| *** wendar has joined #tripleo | 10:31 | |
| jprovazn | shardy: hm, I'm using resource's nested code | 10:32 |
| jprovazn | shardy: should I stick with this new one instead? | 10:33 |
| shardy | jprovazn: It doesn't matter really - I just wanted to point out how to wire in the new parameter | 10:33 |
| jprovazn | ok, thanks | 10:34 |
| *** akrivoka has joined #tripleo | 10:40 | |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Rewrite baseurl after packages are built https://review.openstack.org/223479 | 10:44 |
| openstackgerrit | Pino Toscano proposed openstack/diskimage-builder: dib-lint: validate json/yaml files https://review.openstack.org/223520 | 10:49 |
| openstackgerrit | Dougal Matthews proposed openstack/tripleo-common: [WIP] Remove Tuskar support https://review.openstack.org/223527 | 11:16 |
| *** yamahata has quit IRC | 11:17 | |
| *** Hunner has joined #tripleo | 11:20 | |
| *** Hunner has quit IRC | 11:20 | |
| *** Hunner has joined #tripleo | 11:20 | |
| *** sdake has joined #tripleo | 11:25 | |
| *** sdake_ has joined #tripleo | 11:30 | |
| *** sdake has quit IRC | 11:31 | |
| *** shardy_ has joined #tripleo | 11:36 | |
| *** rhallisey has joined #tripleo | 11:36 | |
| *** shardy has quit IRC | 11:37 | |
| *** athomas_ has joined #tripleo | 11:39 | |
| *** lazy_prince has quit IRC | 11:40 | |
| *** shardy_ has quit IRC | 11:41 | |
| *** shardy has joined #tripleo | 11:42 | |
| *** athomas has quit IRC | 11:42 | |
| *** morazi has joined #tripleo | 11:42 | |
| *** weshay_xchat has joined #tripleo | 11:44 | |
| *** weshay_xchat is now known as weshay | 11:44 | |
| *** stendulker has quit IRC | 11:47 | |
| *** lucasagomes is now known as lucas-hungry | 11:47 | |
| *** trown|outttypeww is now known as trown | 11:49 | |
| *** dprince has joined #tripleo | 11:50 | |
| derekh | shardy: thanks fixed the duplicate day in the commit message https://review.openstack.org/#/c/223538/ | 12:02 |
| *** ibravo has quit IRC | 12:02 | |
| shardy | derekh: Hey, any idea what the status of CI is atm, I got some fast failures earlier failing to build discoverd packages, ring any bells? | 12:03 |
| derekh | shardy: should be ok now https://review.openstack.org/#/c/223476/ | 12:03 |
| slagle | suspect it's the delorean maintenance | 12:03 |
| shardy | derekh: aha, thanks! | 12:04 |
| * shardy rechecks things | 12:04 | |
| derekh | slagle: shardy a commit merged in delorean last night so it now depends on a new liberty repository | 12:04 |
| derekh | slagle: shardy I'm trying a more correct fix here https://review.openstack.org/#/c/223479/ | 12:05 |
| derekh | lunch | 12:05 |
| *** paramite is now known as paramite|afk | 12:06 | |
| *** paramite|afk is now known as paramite | 12:08 | |
| trown | EmilienM: I have a question regarding OPM... looking at the other modules it looks like every option is "optional", if ironic-inspector would not work without an option being set (keystone auth_url for example) is it really optional? | 12:11 |
| jistr | gfidente: did the reviews, had to think a bit to grasp how it all works, it's a complex thing to solve :) looks good | 12:12 |
| *** jaosorior has joined #tripleo | 12:12 | |
| *** athomas_ has quit IRC | 12:13 | |
| *** pradk has quit IRC | 12:14 | |
| *** jayg|g0n3 is now known as jayg | 12:15 | |
| gfidente | jistr, hey thanks! | 12:18 |
| gfidente | I have a sample static-ips.yaml which can be used to test the static ips assignment | 12:18 |
| gfidente | the external lb stuff is harder to test honestly | 12:18 |
| gfidente | and I'm still messing a little with the different network Ports arguments | 12:19 |
| gfidente | static-ips http://fpaste.org/267403/ | 12:22 |
| EmilienM | trown: openstack is providing default values isn't? | 12:22 |
| *** athomas has joined #tripleo | 12:24 | |
| trown | EmilienM: for some things there are not sane defaults | 12:24 |
| trown | like IPs and passwords | 12:24 |
| *** athomas has quit IRC | 12:25 | |
| EmilienM | trown: you would really follow how other puppet modules do | 12:25 |
| EmilienM | trown: if you have any doubt, send the patch and we will review together | 12:26 |
| trown | EmilienM: ok, I should have a WIP today sometime, thanks | 12:27 |
| EmilienM | trown: cool | 12:27 |
| *** jaosorior has quit IRC | 12:29 | |
| *** sdake_ has quit IRC | 12:30 | |
| *** adrianopetrich has joined #tripleo | 12:33 | |
| *** lucas-hungry is now known as lucasagomes | 12:37 | |
| *** shardy_ has joined #tripleo | 12:38 | |
| *** shardy has quit IRC | 12:39 | |
| *** pradk has joined #tripleo | 12:40 | |
| *** shardy_ has quit IRC | 12:43 | |
| *** shardy has joined #tripleo | 12:44 | |
| *** jaosorior has joined #tripleo | 12:45 | |
| jistr | spredzy: one more thing which might be useful in future debugging -- state can creep in from previous instack runs via ~/.cache/image-create or ~/.cache/tripleo (not sure which one it was in my latest issue, but removing those helped) | 12:45 |
| *** aufi has quit IRC | 12:46 | |
| d0ugal | slagle: I am working on a docs patch now :) | 12:46 |
| slagle | cool thx | 12:46 |
| *** Goneri has joined #tripleo | 12:46 | |
| d0ugal | slagle: and going to email the list - proposing that we only remove CLI for now as we don't have an alternative GUI. | 12:47 |
| slagle | ack | 12:47 |
| slagle | gui is actually broken now anyway due to horizon in delorean | 12:47 |
| d0ugal | oh :) | 12:47 |
| slagle | no one must be using it | 12:48 |
| d0ugal | akrivoka: ^ | 12:48 |
| *** tzumainn has joined #tripleo | 13:00 | |
| derekh | slagle: FYI meeting time change has merged | 13:01 |
| slagle | derekh: yo that was quick | 13:01 |
| slagle | i look forward to increased meeting participation! | 13:01 |
| derekh | slagle: wasn't it, I'll mail the list now | 13:02 |
| derekh | slagle: yup, looking forward to being at more of them | 13:02 |
| slagle | it won't just be me complimenting jdob's hairstyle | 13:02 |
| jdob | damnit, that was one of the few motivating factors of my week :( | 13:03 |
| *** masco_ has quit IRC | 13:05 | |
| *** hashar has joined #tripleo | 13:06 | |
| spredzy | jistr, thanks for the heads up | 13:08 |
| spredzy | jistr, did you make it all the way ? | 13:08 |
| jistr | spredzy: running install undercloud now (had to switch the keystone heat domain relationship, re-run is going fine so far) | 13:09 |
| spredzy | :) | 13:10 |
| *** yog_ has quit IRC | 13:24 | |
| *** yog_ has joined #tripleo | 13:24 | |
| *** ukalifon has joined #tripleo | 13:26 | |
| *** jprovazn has quit IRC | 13:29 | |
| dprince | derekh: looks like the DELOREAN repo traffic is really slow this morning | 13:35 |
| dprince | derekh: jobs failing because the disk-images fail to build... | 13:35 |
| derekh | dprince: ya ? in CI or locally ? there | 13:35 |
| dprince | derekh: in CI only | 13:36 |
| dprince | derekh: works for me | 13:36 |
| derekh | dprince: there was talk of switching DNS to a diferent server before tonights 24HR outage | 13:36 |
| dprince | derekh: also, looks like we might should set HOST_IP to something earlier in the script so postci doesn't fail | 13:36 |
| derekh | dprince: there was failures earlier but here gone now, can you point me at a slowness error | 13:37 |
| dprince | derekh: http://logs.openstack.org/87/223487/3/check-tripleo/gate-tripleo-ironic-overcloud-f21puppet-nonha/981d0da/console.html | 13:37 |
| *** jaosorior has quit IRC | 13:38 | |
| pino|work | slagle, thrash: hi, would you be available in a couple of hours to discuss further (better than in a review request) about virt-dib integration? | 13:38 |
| derekh | dprince: that patch is me trying to switch the trunk server we point at | 13:38 |
| derekh | dprince: i.e. I screwed up the patch | 13:38 |
| slagle | pino|work: i can do it now | 13:38 |
| dprince | derekh: oh, I see | 13:38 |
| pino|work | slagle: i cannot now :/ | 13:39 |
| slagle | pino|work: ok, what time works? | 13:39 |
| thrash | pino|work: most likely, yes | 13:39 |
| pino|work | slagle: in ~2h and on | 13:39 |
| slagle | pino|work: how about we just say 1600UTC? that 2h20m from now | 13:40 |
| pino|work | slagle: works for me | 13:40 |
| slagle | ok, i'll be around | 13:41 |
| thrash | pino|work: slagle wfm | 13:41 |
| pino|work | thanks! | 13:41 |
| spredzy | slagle, when the overcloud finish to provision and it runs init-keystone it tries to connect as heat-admin (https://github.com/openstack/python-tripleoclient/blob/master/tripleoclient/v1/overcloud_deploy.py#L700) | 13:47 |
| *** ukalifon has quit IRC | 13:47 | |
| spredzy | to an URL that does not belong to the oc controller | 13:47 |
| spredzy | s/URL/IP | 13:47 |
| spredzy | but when connecting to centos@controller0 everything work as expected | 13:48 |
| spredzy | does that ring a bell ? | 13:48 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: [WIP] Remove workaround to make TE_DATAFILE relative https://review.openstack.org/223595 | 13:49 |
| spredzy | slagle, logs -> http://paste.openstack.org/show/463010/ | 13:51 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: [WIP] Stop creating id_rsa_virt_power https://review.openstack.org/223598 | 13:51 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Remove code to pull clone repositories https://review.openstack.org/222406 | 13:53 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Stop building tuskar-ui-extras https://review.openstack.org/222407 | 13:53 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Stop building ironic-discoverd https://review.openstack.org/222408 | 13:53 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Remove support for mgt-master branches https://review.openstack.org/223599 | 13:53 |
| *** lblanchard has joined #tripleo | 13:53 | |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Stop running delorean twice https://review.openstack.org/223600 | 13:54 |
| *** adrianopetrich has quit IRC | 13:54 | |
| slagle | spredzy: yes, kind of. i suspect you might have old code somewhere | 13:55 |
| spredzy | slagle, ack will redeploy from scratch and remove the ~/.cache as jistr suggested above | 13:55 |
| slagle | just a sec | 13:56 |
| slagle | give me a few to track down how we fixed the heat-admin/centos thing | 13:56 |
| *** egafford has joined #tripleo | 13:56 | |
| slagle | there was a patch to tripleo-heat-templates | 13:56 |
| *** noslzzp has joined #tripleo | 13:57 | |
| slagle | spredzy: do you have this patch in your templates? https://review.openstack.org/#/c/220057 | 13:58 |
| * spredzy checks | 13:58 | |
| slagle | spredzy: and this patch for python-tripleoclient? https://review.gerrithub.io/#/c/243712/ | 13:59 |
| *** julim has quit IRC | 14:01 | |
| *** jprovazn has joined #tripleo | 14:05 | |
| spredzy | slagle, so no and no :) | 14:05 |
| spredzy | That might explain it :) | 14:05 |
| spredzy | I will redeploy cleaning the cache | 14:06 |
| slagle | where do you have those packages installed from? | 14:06 |
| slagle | and what versions, etc | 14:06 |
| *** lsmola has quit IRC | 14:06 | |
| spredzy | slagle, openstack-tripleo-heat-templates-0.8.7-dev185.el7.centos.noarch and python-rdomanager-oscplugin-0.0.9-dev141.el7.centos.noarch | 14:10 |
| slagle | spredzy: hmm, i think those might be ok actually | 14:14 |
| slagle | spredzy: i do have a patch out to update the delorean repos in the docs: https://review.openstack.org/#/c/223293/ | 14:15 |
| slagle | that should be closer to what CI is doing | 14:15 |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Consume the NeutronMechanismDrivers from the hiera data https://review.openstack.org/214649 | 14:15 |
| slagle | i didnt test a full oc deployment though | 14:15 |
| *** david-lyle has quit IRC | 14:20 | |
| *** lblanchard has quit IRC | 14:20 | |
| spredzy | slagle, will rebuild the role with the new repos and let you know | 14:21 |
| openstackgerrit | Rob Pothier proposed openstack/tripleo-heat-templates: Port Cisco Nexus/UCSM ExtraConfig to AllNodes https://review.openstack.org/215694 | 14:25 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Use SEED_IP to decide it we can ssh to the undercloud https://review.openstack.org/223612 | 14:26 |
| *** mcornea has quit IRC | 14:26 | |
| derekh | dprince: ^^ should get rid of the "unbound variable" error you saw | 14:26 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Stop running delorean twice https://review.openstack.org/223600 | 14:28 |
| *** lblanchard has joined #tripleo | 14:28 | |
| dprince | derekh: nice, looking | 14:28 |
| openstackgerrit | Dougal Matthews proposed openstack/tripleo-common: [WIP] Remove Tuskar support https://review.openstack.org/223527 | 14:29 |
| *** kbyrne has quit IRC | 14:31 | |
| *** julim has joined #tripleo | 14:35 | |
| *** kbyrne has joined #tripleo | 14:37 | |
| *** kbyrne has quit IRC | 14:38 | |
| *** kbyrne has joined #tripleo | 14:40 | |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Use SEED_IP to decide if we can ssh to the undercloud https://review.openstack.org/223612 | 14:40 |
| *** mcornea has joined #tripleo | 14:43 | |
| *** aufi has joined #tripleo | 14:47 | |
| *** cody-somerville has joined #tripleo | 14:53 | |
| *** untriaged-bot has joined #tripleo | 15:00 | |
| untriaged-bot | Untriaged bugs so far: | 15:00 |
| untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1495860 | 15:00 |
| *** untriaged-bot has quit IRC | 15:00 | |
| openstack | Launchpad bug 1495860 in diskimage-builder ""disk-image-create ironic-agent fedora iso -o my-deploy-ramdisk" failed" [Undecided,New] | 15:00 |
| *** thrash has quit IRC | 15:01 | |
| *** thrash has joined #tripleo | 15:04 | |
| *** thrash has joined #tripleo | 15:04 | |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Remove workaround to make TE_DATAFILE relative https://review.openstack.org/223595 | 15:04 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Allow for assignment of pre-allocated IPs to the Controller nodes https://review.openstack.org/222497 | 15:04 |
| *** pblaho has quit IRC | 15:11 | |
| jprovazn | shardy, hi again | 15:12 |
| shardy | jprovazn: hi | 15:12 |
| *** david-lyle has joined #tripleo | 15:14 | |
| jprovazn | shardy, I added nested_depth support to events list, but it doesn't help much - I think major bottleneck is going through all resources to find all nested stacks | 15:14 |
| shardy | jprovazn: Ok, so we have to come up with a smarter DB query instead of walking the entire tree in memory I guess | 15:16 |
| jprovazn | shardy, +1, ideally getting all resources in one query (=> this would imply having a "root stack id") | 15:17 |
| openstackgerrit | Timothy Swanson proposed openstack/tripleo-heat-templates: Support new form of servers config for ML2 cisco nexus allnodes config. https://review.openstack.org/223271 | 15:18 |
| jprovazn | shardy, zaneb pointed me to stevebaker's patch https://github.com/openstack/heat/commit/be41d08a44c6769743241515e8afb16e09077422 which did some db optimization for resource counting, I will look at it tomorrow | 15:19 |
| shardy | jprovazn: yeah I was just thinking about that | 15:19 |
| shardy | jprovazn: I guess we may need a new column in the DB tho, similar to the nested_depth that's passed down from the parent | 15:20 |
| shardy | jprovazn: thanks for digging into it - if I get the chance I'll chat to stevebaker about it later | 15:20 |
| jprovazn | shardy, one question about the root id... | 15:21 |
| jprovazn | shardy, it's not possible that some stack would be part of other 2 stacks, is it? | 15:21 |
| *** dtantsur is now known as dtantsur|afk | 15:21 | |
| jprovazn | "part of" == has more than one parent | 15:21 |
| shardy | jprovazn: not unless something has gone very wrong ;) | 15:22 |
| jprovazn | :) ok, cool | 15:22 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Allow for usage of pre-allocated IPs for the Controller nodes https://review.openstack.org/222497 | 15:23 |
| *** aufi has quit IRC | 15:23 | |
| *** adrianopetrich has joined #tripleo | 15:24 | |
| *** shadower has quit IRC | 15:30 | |
| *** slagle has quit IRC | 15:30 | |
| *** marios has quit IRC | 15:31 | |
| *** hashar has quit IRC | 15:34 | |
| *** marios has joined #tripleo | 15:56 | |
| *** alop has joined #tripleo | 15:57 | |
| *** rpothier has joined #tripleo | 15:57 | |
| *** ifarkas has quit IRC | 15:58 | |
| jistr | spredzy: so yeah the same issue -- Permission denied (publickey,gssapi-keyex,gssapi-with-mic). | 16:00 |
| jistr | on overcloud init | 16:00 |
| jistr | somehow the undercloud stack user's key didn't get inserted to the overcloud image probably | 16:01 |
| *** paramite has quit IRC | 16:02 | |
| *** yog_ has quit IRC | 16:02 | |
| spredzy | jistr, I updated the packages in my undercloud with the repo slagle pointed me to https://review.openstack.org/#/c/223293 | 16:03 |
| spredzy | processing, heat stack-delete failed :( so I removed everything and re-buidling pltfm from scratch | 16:03 |
| *** hashar has joined #tripleo | 16:03 | |
| jistr | re updated repo -- i tried with that yesterday, that's the repo used in CI :) | 16:04 |
| *** slagle has joined #tripleo | 16:04 | |
| slagle | pino|work: sorry, i'm hear now | 16:04 |
| slagle | *here | 16:05 |
| jistr | spredzy: though not the rest of what's in that doc, i'll try that next then | 16:05 |
| pino|work | slagle: heya, me too | 16:05 |
| slagle | the cloud instance i use for irc got shutdown! | 16:05 |
| pino|work | planned maintenance | 16:05 |
| slagle | yea, i saw :) | 16:06 |
| pino|work | slagle: so, basically what i'm working on is integrating virt-dib into the image generation workflow (as you can see as well, from the review requests) | 16:06 |
| slagle | yep | 16:07 |
| *** jaosorior has joined #tripleo | 16:07 | |
| pino|work | my approach was to KISS, as it didn't seem much worth in building up an internal infrastructure only for few differencies | 16:07 |
| slagle | pino|work: are you talking about me asking for the cli arg, or the class inheritance? | 16:08 |
| pino|work | this was the latter | 16:08 |
| * pino|work was writing the rest | 16:09 | |
| slagle | ok, so, yea, i think the judging of "simple" here can be subjective | 16:09 |
| slagle | personally i think the class would be simpler, but as i said in the review i'm not going to block or -1 on account of that | 16:09 |
| *** Marga_ has quit IRC | 16:10 | |
| pino|work | regarding the choice of tool, i admit i was a bit biased, being the author of it :) but i guess there needs to be any sort of "policy" for using either | 16:10 |
| pino|work | ah, thrash ^ | 16:10 |
| slagle | pino|work: i dont think we need a policy, but i'm not opposed to one. i am kind of opposed of not having the choice to directly choose though | 16:12 |
| slagle | without having to install/uninstall a package | 16:12 |
| slagle | if i do have virt-dib installed, i should have to uninstall it just to use dib instead | 16:12 |
| pino|work | well, "how to choose which tool to use" is a policy after all, isn't it? ;) | 16:12 |
| slagle | err, "shouldn't have to" | 16:13 |
| slagle | pino|work: thought you meant a coded policy to pick | 16:13 |
| pino|work | i see | 16:13 |
| pino|work | would a cli arg the best approach here? | 16:14 |
| pino|work | (not arguing, just collecting more pro's/con's) | 16:14 |
| slagle | that's my preference | 16:14 |
| slagle | not sure if thrash has an opinion on it | 16:14 |
| pino|work | do we have the concept of configuration, in the cli? | 16:14 |
| slagle | i don't think so really | 16:15 |
| slagle | there are cli args and environment varaible, but no config file for the cli itself | 16:15 |
| pino|work | right | 16:16 |
| *** Marga_ has joined #tripleo | 16:17 | |
| pino|work | slagle: regarding the ci, that was my step after the initial support | 16:20 |
| pino|work | the only issue might be the version of the distro used in the undercloud, when the images are built | 16:20 |
| pino|work | virt-dib is in libguestfs >= 1.30, available since f22+ (and hopefully in rhel 7.3+) | 16:21 |
| slagle | well, if we wanted to ci this (and i'm not saying we do), we'd have to add support for f22 | 16:23 |
| slagle | also dont you need a running libvirt to talk to? | 16:23 |
| *** shivrao has joined #tripleo | 16:24 | |
| *** jistr has quit IRC | 16:24 | |
| slagle | while we have libvirt in our ci environment for our virt testing, it isn't really meant to be used in this way | 16:24 |
| slagle | the udnercloud is already a vm itself | 16:24 |
| pino|work | i know | 16:25 |
| slagle | so there would be no kvm support offered for guests on the undercloud | 16:25 |
| pino|work | an option here could be move the OC image generation out of the UC | 16:25 |
| slagle | true | 16:25 |
| pino|work | considering that libguestfs tools run fine as user, this could improve a bit the deployment workflow | 16:26 |
| pino|work | after all, currently we support deploying with UC and OC images already built, so prebuilding them with virt-dib would not change this | 16:27 |
| *** shivrao_ has joined #tripleo | 16:27 | |
| slagle | yea, i could see that | 16:27 |
| pino|work | and actually, could simplify the workflow, where the only difference between having the images and building them would be an initial step only | 16:28 |
| *** sdake has joined #tripleo | 16:28 | |
| *** shivrao has quit IRC | 16:28 | |
| *** shivrao_ is now known as shivrao | 16:28 | |
| *** cody-somerville has quit IRC | 16:30 | |
| *** sdake_ has joined #tripleo | 16:31 | |
| derekh | slagle: I'll be gone soon the, if the RDO switch over to a temporary server cause problems later, here is my attempt to use our own mirror https://review.openstack.org/#/c/223487/ | 16:31 |
| derekh | slagle: although it probably wont be needed | 16:31 |
| slagle | derekh: ok, thanks | 16:33 |
| slagle | there supposed to be done at 9pm EDT? | 16:34 |
| slagle | i'll keep an eye on rdo-list for any updates | 16:34 |
| *** sdake has quit IRC | 16:35 | |
| derekh | slagle: lol, forget about it, I thought it was starting tonight .... | 16:36 |
| slagle | oh i see :) | 16:37 |
| derekh | well at least I know what needs to happen to mirror it in future when I start creating a new mirror server | 16:37 |
| slagle | it should be a seemless switch back once the new server is up since they are using update the dns record | 16:37 |
| derekh | yup, basically I should have had that patch lined up yesterday not today | 16:38 |
| jroll | other than the whole 'tests are failing' part, which is pretty likely unrelated... any reason we can't land this? https://review.openstack.org/#/c/209079/ | 16:40 |
| jroll | we really want to get that deprecated in Liberty | 16:40 |
| *** tzumainn has quit IRC | 16:41 | |
| *** tzumainn has joined #tripleo | 16:41 | |
| *** marios has quit IRC | 16:45 | |
| pino|work | slagle: would be acceptable to do the "generate OC in seed" step even right now, using dib? | 16:49 |
| thrash | pino|work: slagle I would prefer some type of cli arg | 16:50 |
| thrash | pino|work: there is actually no reason that the image creation needs to happen on the undercloud | 16:51 |
| thrash | that command doesn't require auth. | 16:51 |
| *** lazy_prince has joined #tripleo | 16:51 | |
| pino|work | for auth you mean root? | 16:51 |
| thrash | pino|work: no, OS auth | 16:51 |
| thrash | OS == OpenStack | 16:51 |
| pino|work | ok | 16:52 |
| slagle | pino|work: what do you mean "generate OC in seed"? build the images on the seed or undercloud? we do that now | 16:52 |
| pino|work | slagle: what we just said above: instead of build the OC image in UC, build it on seed or "outside" | 16:52 |
| thrash | tripleoclient rpm should have all the necessary deps to be able to do that. | 16:53 |
| slagle | ah. we only have a single bootstrap node | 16:53 |
| slagle | it's just the undercloud | 16:53 |
| slagle | but yea as thrash says, it's possible | 16:54 |
| pino|work | ok, can start that as well | 16:55 |
| slagle | yea, for diskimage-builder it should work | 16:55 |
| slagle | don't know about virt-dib, you're not on the same host as libvirt | 16:56 |
| pino|work | doesn't matter, it can use libvirt only to run its own vm (called appliance) | 16:56 |
| slagle | right, but that libvirt won't have kvm | 16:56 |
| pino|work | noted; we'll see how it can perform | 16:57 |
| slagle | pino|work: ok. one other thing i thought is that might be good to email the list about this new approach | 16:58 |
| slagle | make sure folks are aware of the patches, etc | 16:58 |
| pino|work | openstack-dev? | 16:59 |
| slagle | yea | 16:59 |
| pino|work | ok | 16:59 |
| slagle | i dont know that anyone would object | 16:59 |
| slagle | but, we should give it a little more awareness i think | 16:59 |
| pino|work | makes sense | 16:59 |
| *** trown is now known as trown|lunch | 17:00 | |
| *** hashar has quit IRC | 17:04 | |
| lucasagomes | hey folks, can I get some eyes on https://review.openstack.org/#/c/194161/2 ? This is a requirement for deprecating the bash ramdisk | 17:07 |
| lucasagomes | we want to do it in L | 17:07 |
| lucasagomes | the patch is hanging there for a long time already | 17:08 |
| slagle | lucasagomes: it's actually not clear to me what the deprecation relationship is between dib and ironic | 17:09 |
| lucasagomes | slagle, the bash ramdisk is only built by using DIB | 17:10 |
| lucasagomes | and that ramdisk is now deprecated in Ironic | 17:10 |
| lucasagomes | it uses specific endpoints of the API to talk to Ironic and those endpoints are already deprecated in Ironic | 17:11 |
| *** derekh has quit IRC | 17:11 | |
| *** yamahata has joined #tripleo | 17:11 | |
| lucasagomes | slagle, so with those patches we want to, 1) convert the tripleo-CI to use IPA instead of the bash ramdisk 2) add a deprecation warning when one build the bash ramdisk in DIB so they will know it may not work with future version of ironic | 17:11 |
| slagle | lucasagomes: right, so given dib is independently released, when we deprecate the bash ramdisk, how long does it have to stay there? | 17:11 |
| slagle | well, tripleo-ci doesnt use devtest anymore | 17:12 |
| pino|work | slagle, thrash: btw, thanks for the discussions and inputs | 17:12 |
| lucasagomes | slagle, we want to keep it deprecated in Mitaka and remove it after it | 17:12 |
| lucasagomes | slagle, right, but it doesn't use the tripleo-incubator scripts to build the images? (re https://review.openstack.org/#/c/210461/) | 17:13 |
| *** sthillma has joined #tripleo | 17:14 | |
| slagle | it doesnt, we switched a little bit ago | 17:14 |
| slagle | i'm not saying we can't deprecate | 17:14 |
| slagle | tripleo uses deprecated stuff all the time (unfortunately) | 17:14 |
| slagle | but we'd need to update | 17:14 |
| lucasagomes | slagle, right, can we add the deprecated messages to the deploy-ironic element | 17:15 |
| lucasagomes | and in parallel I can modify the new way we run the tripleo-ci to test IPA instead? | 17:15 |
| lucasagomes | (so it won't use deprecated stuff) | 17:15 |
| *** lazy_prince has quit IRC | 17:16 | |
| slagle | yea, we can add the message. i think the reason that hasn't happened is b/c the patch has not passed ci | 17:17 |
| lucasagomes | slagle, this one https://review.openstack.org/#/c/209079/ ? | 17:17 |
| * lucasagomes checks | 17:17 | |
| slagle | i gather the ci failure is unrelated | 17:17 |
| lucasagomes | I will remove the Depends-On from the commit message there as well and it will rerun the tests | 17:17 |
| lucasagomes | (the depends-on is pointing to the tripleo-incubator scripts) | 17:18 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/diskimage-builder: Deprecate the deploy-ironic element https://review.openstack.org/209079 | 17:18 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/diskimage-builder: Deprecate the deploy-ironic element https://review.openstack.org/209079 | 17:18 |
| slagle | i'll have another look when i can, but it's hard to get motivated to merge stuff when we have enough ci issues as well | 17:18 |
| slagle | even though it's unrelated | 17:18 |
| lucasagomes | slagle, yeah, I understand, it's just that we don't want to say it's deprecated in ironic and not in DIB | 17:19 |
| lucasagomes | it may be confusing for the users | 17:19 |
| lucasagomes | slagle, but anyway, we plan to leave the endpoints there for the M cycle so it gives us time to fix the CI | 17:19 |
| slagle | ok | 17:19 |
| *** adam_g has joined #tripleo | 17:20 | |
| *** adam_g has joined #tripleo | 17:20 | |
| rhallisey | https://review.openstack.org/#/c/223182/ | 17:21 |
| *** cody-somerville has joined #tripleo | 17:22 | |
| lucasagomes | thanks! | 17:24 |
| *** penick has joined #tripleo | 17:26 | |
| *** lucasagomes is now known as lucas-dinner | 17:27 | |
| *** mcornea has quit IRC | 17:28 | |
| *** sthillma has quit IRC | 17:32 | |
| *** sdake_ is now known as sdake | 17:33 | |
| *** tzumainn has quit IRC | 17:36 | |
| *** tzumainn has joined #tripleo | 17:36 | |
| *** sdake_ has joined #tripleo | 17:41 | |
| *** sdake has quit IRC | 17:44 | |
| *** kbyrne has quit IRC | 17:49 | |
| *** dsneddon has quit IRC | 17:52 | |
| *** dsneddon has joined #tripleo | 17:52 | |
| *** dsneddon has quit IRC | 17:52 | |
| *** dsneddon has joined #tripleo | 17:52 | |
| *** dprince has quit IRC | 17:54 | |
| *** dprince has joined #tripleo | 17:56 | |
| *** sthillma has joined #tripleo | 18:04 | |
| *** trown|lunch is now known as trown | 18:11 | |
| *** noslzzp has quit IRC | 18:13 | |
| *** noslzzp has joined #tripleo | 18:14 | |
| *** dprince has quit IRC | 18:14 | |
| *** dprince has joined #tripleo | 18:15 | |
| *** shivrao has quit IRC | 18:15 | |
| *** cody-somerville has quit IRC | 18:16 | |
| *** shardy is now known as shardy_afk | 18:22 | |
| *** dsneddon has quit IRC | 18:36 | |
| *** dsneddon has joined #tripleo | 18:37 | |
| *** dsneddon has quit IRC | 18:37 | |
| *** dsneddon has joined #tripleo | 18:37 | |
| *** sdake_ is now known as sdake | 18:45 | |
| *** shivrao has joined #tripleo | 18:49 | |
| trown | EmilienM: thanks for the review on https://review.openstack.org/#/c/223690/ I have a follow-up PS ready to go up, but I have a couple questions | 18:50 |
| trown | 1) For the acceptance tests, ironic-inspector would need to be on delorean? | 18:51 |
| EmilienM | trown: yes for 1/ | 18:51 |
| *** pelix has quit IRC | 18:51 | |
| trown | 2) for your comment about puppet not conflicting with packaging, do you just mean to remove the directory part? or the files too? | 18:52 |
| trown | EmilienM: ok, I am working on getting it to delorean...it is not there yet though | 18:52 |
| *** cody-somerville has joined #tripleo | 18:54 | |
| EmilienM | trown: if there is no package today, then let's move forward and add acceptance later | 18:55 |
| *** akrivoka has quit IRC | 18:55 | |
| EmilienM | I don't want to block the dev here | 18:55 |
| trown | EmilienM: cool, I will take on adding the acceptance tests after getting it into delorean | 18:56 |
| *** egafford has quit IRC | 18:58 | |
| *** dprince has quit IRC | 19:00 | |
| openstackgerrit | Andreas Jaeger proposed openstack/tuskar-ui: Remove .tx/config https://review.openstack.org/223760 | 19:00 |
| *** dprince has joined #tripleo | 19:01 | |
| openstackgerrit | Merged openstack/tripleo-heat-templates: switch to vxlan by default https://review.openstack.org/215726 | 19:04 |
| *** noslzzp_ has joined #tripleo | 19:09 | |
| openstackgerrit | Ben Nemec proposed openstack/tripleo-heat-templates: Add YAML sanity check https://review.openstack.org/223766 | 19:09 |
| *** noslzzp has quit IRC | 19:12 | |
| *** shardy_afk is now known as shardy | 19:18 | |
| *** shivrao has quit IRC | 19:23 | |
| *** noslzzp_ has quit IRC | 19:27 | |
| *** noslzzp has joined #tripleo | 19:28 | |
| openstackgerrit | venkatamahesh proposed openstack/tripleo-heat-templates: Added home-page value with openstack.org https://review.openstack.org/223776 | 19:33 |
| openstackgerrit | Tzu-Mainn Chen proposed openstack/tripleo-specs: Spec for TripleO overcloud deployment library and API https://review.openstack.org/219754 | 19:38 |
| *** Goneri has quit IRC | 19:41 | |
| *** dprince has quit IRC | 19:56 | |
| *** jprovazn has quit IRC | 20:21 | |
| *** SlickNik has left #tripleo | 20:24 | |
| *** shardy has quit IRC | 20:29 | |
| *** jayg is now known as jayg|g0n3 | 20:29 | |
| *** shadower has joined #tripleo | 20:29 | |
| openstackgerrit | Ben Nemec proposed openstack/tripleo-heat-templates: Add YAML sanity check https://review.openstack.org/223766 | 20:32 |
| *** morazi has quit IRC | 20:33 | |
| openstackgerrit | Rob Pothier proposed openstack/tripleo-heat-templates: Enable Cisco Nexus and UCSM plugins https://review.openstack.org/198754 | 20:37 |
| *** rhallisey has quit IRC | 20:41 | |
| bnemec | Crap, we broke the docs job for instack-undercloud since I proposed that infra change. | 20:41 |
| *** SlickNik has joined #tripleo | 20:54 | |
| *** cody-somerville has quit IRC | 20:57 | |
| *** pradk has quit IRC | 20:59 | |
| *** untriaged-bot has joined #tripleo | 21:00 | |
| untriaged-bot | Untriaged bugs so far: | 21:00 |
| untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1495860 | 21:00 |
| openstack | Launchpad bug 1495860 in diskimage-builder ""disk-image-create ironic-agent fedora iso -o my-deploy-ramdisk" failed" [Undecided,New] | 21:00 |
| *** untriaged-bot has quit IRC | 21:00 | |
| *** jaosorior has quit IRC | 21:02 | |
| *** lucas-dinner has quit IRC | 21:06 | |
| *** rpothier has quit IRC | 21:07 | |
| *** lblanchard has quit IRC | 21:19 | |
| bnemec | slagle: We're going to need https://review.openstack.org/#/c/223810/ to unblock instack-undercloud. | 21:19 |
| slagle | looking | 21:21 |
| slagle | do we have heat templates in instack-undercloud? | 21:23 |
| slagle | oh, so we do. forgot about those | 21:24 |
| slagle | we should probably remove those | 21:24 |
| slagle | anyway, docs are accurate as is! | 21:24 |
| bnemec | Oh right, dib elements was actually the thing I probably should have called out there. | 21:33 |
| bnemec | It's kind of just placeholder to get the docs job working anyway though. | 21:34 |
| *** zaneb has quit IRC | 21:36 | |
| *** zaneb has joined #tripleo | 21:51 | |
| *** gfidente has quit IRC | 21:55 | |
| *** noslzzp_ has joined #tripleo | 22:03 | |
| *** noslzzp has quit IRC | 22:06 | |
| *** weshay has quit IRC | 22:07 | |
| *** penick has quit IRC | 22:08 | |
| *** panda has quit IRC | 22:10 | |
| *** noslzzp_ has quit IRC | 22:10 | |
| *** panda has joined #tripleo | 22:10 | |
| *** shivrao has joined #tripleo | 22:14 | |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Ensure mysql root can only connect from localhost https://review.openstack.org/208993 | 22:21 |
| openstackgerrit | Dan Sneddon proposed openstack/tripleo-heat-templates: Configure ctlplane network with a static IP https://review.openstack.org/206022 | 22:21 |
| openstackgerrit | Merged openstack-infra/tripleo-ci: Stop running delorean twice https://review.openstack.org/223600 | 22:22 |
| bnemec | All cores: We need the following patches merged to unblock instack-undercloud and tripleoclient: | 22:49 |
| bnemec | https://review.openstack.org/#/c/223858/ | 22:49 |
| bnemec | https://review.openstack.org/#/c/223810/ | 22:49 |
| *** sdake has quit IRC | 22:56 | |
| *** julim has quit IRC | 23:08 | |
| *** openstackgerrit has quit IRC | 23:16 | |
| *** openstackgerrit has joined #tripleo | 23:16 | |
| *** david-lyle has quit IRC | 23:17 | |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Use a specific fedora mirror on the jenkins node https://review.openstack.org/223868 | 23:21 |
| openstackgerrit | Merged openstack/diskimage-builder: dib-lint: validate json/yaml files https://review.openstack.org/223520 | 23:23 |
| *** beagles has quit IRC | 23:48 | |
| *** beagles has joined #tripleo | 23:48 | |
| *** beagles is now known as Guest94042 | 23:49 | |
| *** Guest94042 is now known as b3nt_pin | 23:49 | |
| *** alop has quit IRC | 23:53 | |
| openstackgerrit | Merged openstack/diskimage-builder: Update fedora elements README from free text to table formatting https://review.openstack.org/221995 | 23:56 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!