| *** shadower has quit IRC | 00:23 | |
| *** shadower has joined #tripleo | 00:23 | |
| openstackgerrit | Ben Nemec proposed openstack/instack-undercloud: Fix similar vm name on instack-virt-setup script. https://review.openstack.org/239957 | 00:28 |
|---|---|---|
| openstackgerrit | Ben Nemec proposed openstack/tripleo-incubator: Add quoting variable for vnname. https://review.openstack.org/239939 | 00:29 |
| *** alop has quit IRC | 00:30 | |
| *** jerrygb has joined #tripleo | 00:31 | |
| *** jerrygb has quit IRC | 00:32 | |
| openstackgerrit | Ben Nemec proposed openstack/tripleo-heat-templates: Enable notifications on the overcloud https://review.openstack.org/241421 | 00:34 |
| *** Marga_ has quit IRC | 00:39 | |
| *** Marga_ has joined #tripleo | 00:40 | |
| *** beagles has quit IRC | 00:51 | |
| *** b3nt_pin has joined #tripleo | 00:53 | |
| *** b3nt_pin is now known as Guest46631 | 00:53 | |
| *** lblanchard has joined #tripleo | 01:22 | |
| *** panda has quit IRC | 01:40 | |
| *** panda has joined #tripleo | 01:41 | |
| *** rlandy has quit IRC | 01:43 | |
| *** julim has joined #tripleo | 01:46 | |
| *** tiswanso has joined #tripleo | 01:47 | |
| *** noslzzp has joined #tripleo | 01:56 | |
| *** rlandy has joined #tripleo | 01:56 | |
| openstackgerrit | greghaynes proposed openstack/diskimage-builder: Add missing six requirement for svc-map element https://review.openstack.org/232962 | 01:57 |
| *** weshay has quit IRC | 02:00 | |
| *** shivrao has quit IRC | 02:01 | |
| *** SlickNik has left #tripleo | 02:21 | |
| *** tiswanso has quit IRC | 02:25 | |
| *** tiswanso has joined #tripleo | 02:25 | |
| *** lblanchard has quit IRC | 02:31 | |
| *** sthillma has quit IRC | 02:57 | |
| *** sthillma has joined #tripleo | 02:58 | |
| *** sthillma has quit IRC | 03:01 | |
| *** tzumainn has quit IRC | 03:02 | |
| *** yuanying_ has joined #tripleo | 03:21 | |
| *** yuanying has quit IRC | 03:21 | |
| *** shivrao has joined #tripleo | 03:33 | |
| openstackgerrit | Merged openstack/diskimage-builder: Fix uniqueness check of initrd in fedora-minimal https://review.openstack.org/235705 | 03:36 |
| *** yuanying_ has quit IRC | 03:36 | |
| *** shivrao_ has joined #tripleo | 03:36 | |
| *** shivrao has quit IRC | 03:38 | |
| *** shivrao_ is now known as shivrao | 03:38 | |
| *** lazy_prince has joined #tripleo | 03:39 | |
| *** untriaged-bot has joined #tripleo | 04:01 | |
| untriaged-bot | Untriaged bugs so far: | 04:01 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1510901 | 04:01 |
| openstack | Launchpad bug 1510901 in tripleo "seed's nova-compute service start fail" [Undecided,In progress] - Assigned to Cameron.C (cameron-c) | 04:01 |
| untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1511715 | 04:01 |
| openstack | Launchpad bug 1511715 in diskimage-builder "Images using draycut should have the option to pass drivers to include in build" [Undecided,New] - Assigned to Julia Kreger (juliaashleykreger) | 04:01 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1509148 | 04:01 |
| openstack | Launchpad bug 1509148 in tripleo "keystone initialize fails if undercloud can't reach overcloud public vip" [Undecided,New] | 04:01 |
| untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1505900 | 04:01 |
| openstack | Launchpad bug 1505900 in diskimage-builder "Failed to create 'wily' release with DIB_RELEASE" [Undecided,New] | 04:01 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1504263 | 04:01 |
| openstack | Launchpad bug 1504263 in tripleo "os-net-config mapping by MAC fails if trying to map in alternate order" [Undecided,Triaged] | 04:01 |
| *** untriaged-bot has quit IRC | 04:01 | |
| *** Slower has joined #tripleo | 04:05 | |
| *** rlandy has quit IRC | 04:07 | |
| *** anande has joined #tripleo | 04:08 | |
| *** yuanying has joined #tripleo | 04:08 | |
| *** masco__ has joined #tripleo | 04:09 | |
| *** anande has quit IRC | 04:11 | |
| *** shivrao has quit IRC | 04:12 | |
| *** anande has joined #tripleo | 04:12 | |
| *** anande has quit IRC | 04:12 | |
| *** anande has joined #tripleo | 04:23 | |
| *** rlandy has joined #tripleo | 04:25 | |
| *** links has joined #tripleo | 04:26 | |
| *** tiswanso has quit IRC | 04:31 | |
| *** tzumainn has joined #tripleo | 04:32 | |
| *** anande has quit IRC | 04:41 | |
| *** morazi has quit IRC | 04:48 | |
| *** rlandy has quit IRC | 05:20 | |
| *** tzumainn has quit IRC | 05:29 | |
| *** athomas has quit IRC | 05:41 | |
| *** jaosorior has joined #tripleo | 05:53 | |
| *** killer_prince has joined #tripleo | 06:31 | |
| *** lazy_prince has quit IRC | 06:35 | |
| *** lazy_prince has joined #tripleo | 06:43 | |
| *** killer_prince has quit IRC | 06:43 | |
| *** jprovazn has joined #tripleo | 06:43 | |
| *** jtomasek has joined #tripleo | 06:50 | |
| *** lazy_prince has quit IRC | 06:52 | |
| *** lazy_prince has joined #tripleo | 06:52 | |
| *** mcornea has joined #tripleo | 06:55 | |
| *** lazy_prince has quit IRC | 06:58 | |
| *** rasca has joined #tripleo | 06:58 | |
| *** shivrao has joined #tripleo | 07:12 | |
| *** shivrao_ has joined #tripleo | 07:15 | |
| *** shivrao has quit IRC | 07:17 | |
| *** shivrao_ is now known as shivrao | 07:17 | |
| *** lsmola has joined #tripleo | 07:20 | |
| *** stendulker has joined #tripleo | 07:23 | |
| *** ukalifon1 has joined #tripleo | 07:37 | |
| *** sthillma has joined #tripleo | 07:38 | |
| *** shivrao has quit IRC | 07:40 | |
| *** Marga_ has quit IRC | 07:40 | |
| *** tremble has joined #tripleo | 07:41 | |
| *** sthillma_ has joined #tripleo | 07:42 | |
| *** devvesa has joined #tripleo | 07:42 | |
| *** Marga_ has joined #tripleo | 07:44 | |
| *** sthillma has quit IRC | 07:44 | |
| *** sthillma_ is now known as sthillma | 07:44 | |
| *** zigo has quit IRC | 07:53 | |
| jaosorior | tremble: Sup dude | 07:56 |
| *** zigo has joined #tripleo | 07:56 | |
| tremble | Hey jaosorior | 07:56 |
| jaosorior | tremble: Well, it seems that this commit merged https://review.openstack.org/#/c/230375/39 | 07:56 |
| jaosorior | but it will be reverted because of the update problems | 07:57 |
| jaosorior | which I've been trying to figure out, with no luck so far | 07:57 |
| tremble | yeah I noticed, there's a new CR to revert it | 07:57 |
| tremble | https://review.openstack.org/#/c/241688/ | 07:57 |
| jaosorior | yup | 07:57 |
| jaosorior | so, if we want the refacter to land, we need to figure out why it's failing | 07:58 |
| tremble | jaosorior: Yup, just running an upgrade now to figure out what's going on. | 07:58 |
| jaosorior | tremble: Let me know if you get the same issues I was getting yesterday | 07:58 |
| jaosorior | tremble: On the other hand, if you can review the TLS patch it would be appreciated https://review.openstack.org/#/c/231930/ | 07:59 |
| tremble | Hmm, with mine something tried to trigger a host rebuild... | 07:59 |
| jaosorior | tremble: what's the latest commit you have checked out? | 08:00 |
| *** ifarkas has joined #tripleo | 08:00 | |
| tremble | 805368cc3d34a4dfe24a027eef5408a0a71e69ac | 08:00 |
| jaosorior | tremble: Which is that? | 08:02 |
| tremble | http://paste.openstack.org/show/478059/ | 08:02 |
| tremble | Oh, I know what triggered the rebuild, that's my mistake | 08:03 |
| jaosorior | what triggered it? | 08:04 |
| tremble | It's my hack to inject the new puppet-tripleo | 08:04 |
| tremble | (local change) | 08:04 |
| * tremble cleans up and starts again | 08:04 | |
| *** sthillma has quit IRC | 08:04 | |
| jaosorior | lol dammit | 08:04 |
| *** shardy has joined #tripleo | 08:04 | |
| tremble | I actually really hate that updating userdata *always* rebuilds the nodes. There are use cases (especially where userdata just hands off to some config management tool) where that's annoying behaviour | 08:05 |
| tremble | It'd be kinda nice to be able to specify that behaviour | 08:06 |
| tremble | jaosorior: What was the bug you hit yesterday that prevented deletion? | 08:10 |
| jaosorior | tremble: the engine not finding the heat domain user | 08:11 |
| jaosorior | tremble: Are you hitting the same bug? | 08:11 |
| tremble | 12015-11-05 08:11:19.393 21664 WARNING heat.engine.service [req-384d1e36-b825-4eb5-9017-a0962bc5beef - 293dc9af98c84037a0f327a80e35fb17-75c4d51d-1944-43d8-a5a4-7495c36] Access denied to resource NovaCompute | 08:11 |
| jaosorior | funky | 08:11 |
| jaosorior | tremble: What's the status reason you get from the resource? | 08:12 |
| tremble | It's just sat there in Stack DELETE started with nothing actually changing | 08:13 |
| jaosorior | whaaa | 08:14 |
| tremble | I think my overcloud has gone a little wonky | 08:15 |
| *** openstackgerrit has quit IRC | 08:16 | |
| jaosorior | tremble: Well, if it's a similar issue, this is what fixed it https://review.openstack.org/#/c/234707/1/heat/engine/resources/stack_user.py | 08:16 |
| *** openstackgerrit has joined #tripleo | 08:17 | |
| *** Marga_ has quit IRC | 08:27 | |
| *** Marga_ has joined #tripleo | 08:28 | |
| *** devvesa has quit IRC | 08:32 | |
| *** devvesa has joined #tripleo | 08:32 | |
| *** openstack has joined #tripleo | 08:33 | |
| *** regebro has joined #tripleo | 08:38 | |
| *** paramite has joined #tripleo | 08:51 | |
| *** aufi has joined #tripleo | 08:53 | |
| *** jaosorior has quit IRC | 08:53 | |
| *** lazy_prince has joined #tripleo | 08:54 | |
| *** stendulker_ has joined #tripleo | 09:00 | |
| *** stendulker has quit IRC | 09:02 | |
| *** stendulker_ has quit IRC | 09:04 | |
| *** jaosorior has joined #tripleo | 09:09 | |
| *** derekh has joined #tripleo | 09:09 | |
| jaosorior | marios: ping | 09:18 |
| marios | jaosorior: hey o/ just reviewed the tls inject patch | 09:19 |
| jaosorior | marios: I just responded | 09:19 |
| marios | jaosorior: sec looking | 09:20 |
| marios | jaosorior: so you're not using the nodeindex to identify the controller node, sorry. it is used to specify which certificate goes to which controller. | 09:21 |
| shardy | jaosorior: Hey, yesterday you hit some issues on update, which led me to believe the keystone management of endpoints is broken (at least for updates), is that correct? | 09:22 |
| *** mbound has joined #tripleo | 09:22 | |
| shardy | jaosorior: we inadvertantly landed the keystone/puppet patch, and bnemec proposed a revert in https://review.openstack.org/#/c/241688/ | 09:22 |
| jaosorior | marios: Yes, that's the case | 09:22 |
| jaosorior | shardy: I saw, I was discussing that with tremble, and he's doing update testing on his side | 09:23 |
| shardy | jaosorior: Ok, thanks - manually testing isn't ideal, but it's our only option until we have CI update testing in place | 09:23 |
| marios | jaosorior: sec, thinking... but it seems to me we can re-use the existing controllerextraconfigpre (it would just simplify the change is all, but it looks like it would work fine this way) | 09:23 |
| cmyster | shardy: got a minute ? | 09:23 |
| tremble | marios: I thought the Extra ones were reserved for Operators to do extra things | 09:23 |
| *** gfidente has joined #tripleo | 09:24 | |
| jaosorior | shardy: Currently I'm doing update testing on the TLS injection and CA injection patch sets, and the update from not having these resources at all to having them and injecting certs and root certs works (update was successful) | 09:24 |
| *** jistr has joined #tripleo | 09:24 | |
| shardy | cmyster: hi! sure | 09:24 |
| marios | shardy: when you have a sec can you have a look at the https://review.openstack.org/#/c/231930/ its the TLS change which jaosorior has updated to use the ssl injection | 09:24 |
| marios | tremble: indeed | 09:24 |
| shardy | jaosorior: Ok, that's a useful datapoint, thanks | 09:24 |
| jaosorior | now I'll do an update to changing the cert/private key | 09:24 |
| tremble | marios: And TLS/SSL is a pretty standard thing that Operators will want. | 09:24 |
| jaosorior | and see what happens | 09:24 |
| marios | tremble: so this would be the problem, if more than one was to be specified. and i hadn't thought that this is something we may want to .. yeah | 09:24 |
| jaosorior | marios, more than one what? | 09:25 |
| marios | that (going to be a standard thing)... was just wondering if it needed its own pre-config interface is all | 09:25 |
| shardy | and you've based your update on a master commit before 62f2237cebe941f1d5ca1b3c86df74ff1df4319a to your patch, right? | 09:25 |
| jaosorior | shardy: Yeah, before spredzy's commits | 09:25 |
| shardy | jaosorior: Ok, thanks for clarifying | 09:25 |
| marios | jaosorior: i meant you wished to specify more than one pre-config (right now we use it if cisco env, or if bigswitch env etc) | 09:25 |
| jaosorior | spredzy's commits seem to fail at some point when creating keystone endpoints | 09:26 |
| shardy | pretty paranoid about update regressions atm as there's a bunch of downstream pain related to them L:( | 09:26 |
| jaosorior | trying to assign the nova endpoint, and not finding the nova service related to it | 09:26 |
| jaosorior | at least that's how it failed for me | 09:26 |
| jaosorior | waiting to see if tremble stumbles upon the same | 09:26 |
| shardy | marios: FWIW the ResourceChain heat BP jdob has been working on is aimed at enabling easier chaining of e.g multiple ExtraConfig things together | 09:26 |
| shardy | for now, the expectation is that folks will just write a tiny wrapper template | 09:27 |
| shardy | https://github.com/openstack/tripleo-heat-templates/blob/master/puppet/extraconfig/pre_deploy/controller/multiple.yaml | 09:27 |
| marios | shardy: i see thanks | 09:27 |
| marios | jaosorior: will update my vote, thanks for pinging | 09:28 |
| marios | shardy: yeah, so in theory, if we are able to chain multiple, we don't need to define a new preconfig interface per feature | 09:28 |
| jaosorior | marios: Thanks for reviewing | 09:28 |
| shardy | marios: Yeah, that's the idea, whether it makes sense to just add one in this case is a judgement call tho | 09:28 |
| shardy | the idea was really that *ExtraConfig was mostly for operator enhancements and third-party integration, so I get the argument that SSL is more of a core feature which could be wired in directly | 09:29 |
| marios | shardy: is that something jdob is working on for this cycle ? | 09:30 |
| shardy | jaosorior: Ok, thanks, if tremble also sees the nova failure, I guess we go ahead and revert, then do more investigation | 09:30 |
| shardy | marios: Yeah, I think so, but it's still at the spec interface-bikeshedding stage I think | 09:31 |
| shardy | https://review.openstack.org/#/c/228615/ | 09:31 |
| shardy | I'll try to find out later if another revision is in the pipleline | 09:31 |
| marios | shardy: thanks. well in any case it doesn' apply here, especially given the 'this is a common thing we will want to do anyway' aspect | 09:32 |
| shardy | the actual coding shouldn't be that much work | 09:33 |
| gfidente | shardy, /me applying https://review.openstack.org/#/c/238194/ to 7.0 | 09:34 |
| gfidente | shardy, is that fair to do? | 09:35 |
| shardy | gfidente: cool, be interested to hear how it goes | 09:35 |
| shardy | gfidente: Yes, there may be some internal rework, but it should be functional | 09:35 |
| gfidente | yeah I wanted to make sure that is intended to work on 7.0 too | 09:35 |
| shardy | gfidente: it should, but I've only tested on mitaka heat | 09:36 |
| jaosorior | yay, updating privatekeys/certs works | 09:37 |
| *** tosky has joined #tripleo | 09:37 | |
| shardy | derekh: Hey, just looking at the trigger merge conflicts in https://review.openstack.org/#/c/240938/ | 09:44 |
| shardy | are you already working on another revision, or shall I attempt to rebase/push? | 09:44 |
| * shardy puzzles over if we need a periodic job for the stable branch.. | 09:47 | |
| *** olap has joined #tripleo | 09:51 | |
| derekh | shardy: opps, I hadn't noticed the merge conflict | 09:52 |
| derekh | shardy: I think we do need a periodic job, I'm hoping to get them to save artifacts (e.g. images) to be used in other CI jobs | 09:53 |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Fix dependency for Ceilometer agent when using MySQL https://review.openstack.org/239373 | 09:53 |
| *** adrianopetrich has quit IRC | 09:53 | |
| derekh | shardy: if you have a rebase got for it, if not I'll sort it out in a few minutes | 09:53 |
| derekh | *go | 09:53 |
| derekh | brb | 09:53 |
| shardy | derekh: Ok, I'll push what I have, cheers | 09:53 |
| shardy | derekh: Ok, pushed my rebase | 09:56 |
| jaosorior | when adding the TLS parts, expecting the VIPDeployment, NodeExtraConfig and some PostDeployments to run again is normal, right? | 09:57 |
| jaosorior | I guess the fact that those get updated too is to be expected | 09:58 |
| tremble | jaosorior: Grah, something still triggered a rebuild of the nodes. | 09:58 |
| shardy | jaosorior, gfidente: thanks for your comments on the project-config patch, replied | 09:59 |
| jaosorior | tremble: uhm... are you sure you took away the domain override commit? | 09:59 |
| jaosorior | tremble: Can' | 09:59 |
| shardy | it's not testing upgrades at all (yet), so I think we should rename later when it actually does | 09:59 |
| jaosorior | Can't think of anything else that would trigger rebuilding the nodes | 09:59 |
| jaosorior | shardy: Doesn't it? | 09:59 |
| shardy | jaosorior: no | 09:59 |
| *** untriaged-bot has joined #tripleo | 10:00 | |
| untriaged-bot | Untriaged bugs so far: | 10:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1510901 | 10:00 |
| openstack | Launchpad bug 1510901 in tripleo "seed's nova-compute service start fail" [Undecided,In progress] - Assigned to Cameron.C (cameron-c) | 10:00 |
| untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1511715 | 10:00 |
| openstack | Launchpad bug 1511715 in diskimage-builder "Images using draycut should have the option to pass drivers to include in build" [Undecided,New] - Assigned to Julia Kreger (juliaashleykreger) | 10:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1509148 | 10:00 |
| openstack | Launchpad bug 1509148 in tripleo "keystone initialize fails if undercloud can't reach overcloud public vip" [Undecided,New] | 10:00 |
| untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1505900 | 10:00 |
| openstack | Launchpad bug 1505900 in diskimage-builder "Failed to create 'wily' release with DIB_RELEASE" [Undecided,New] | 10:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1504263 | 10:00 |
| openstack | Launchpad bug 1504263 in tripleo "os-net-config mapping by MAC fails if trying to map in alternate order" [Undecided,Triaged] | 10:00 |
| *** untriaged-bot has quit IRC | 10:00 | |
| jaosorior | shardy: Ah! I think I understand now | 10:01 |
| shardy | jaosorior: first step is just CI testing the stable branch, then we can add upgrade/update testing when we have both stable and master tested in isolation | 10:01 |
| tremble | jaosorior: yeah | 10:01 |
| shardy | currently we cut the branches but they have zero CI, this fixes that | 10:01 |
| shardy | then we have to implement update/upgrade testing next | 10:01 |
| jaosorior | that makes sense | 10:01 |
| shardy | this first step just runs the existing CI pointing at the stable delorean | 10:02 |
| cmyster | shardy: update upgrade ? | 10:02 |
| shardy | cmyster: there are (at least) two different things to test | 10:02 |
| cmyster | only two? :) | 10:02 |
| shardy | one is update all the things and prove nothing breaks | 10:02 |
| shardy | I'm calling that "update" | 10:03 |
| cmyster | k | 10:03 |
| shardy | e.g update the undercloud, prove you can still deploy an overcloud (potentially with mismatched versions of overcloud) | 10:03 |
| shardy | and maybe just do a no-op update of the overcloud to the new templates, prove nothing explodes or is replaced | 10:03 |
| shardy | "upgrade" is actually upgrading e.g liberty->mitaka on the overcloud | 10:04 |
| shardy | we have zero coverage for any of that yet, so I'm trying to start simple ;) | 10:04 |
| shardy | actually, let's write this down in an etherpad | 10:04 |
| cmyster | +1 | 10:05 |
| shardy | https://etherpad.openstack.org/p/tripleo-update-testing | 10:05 |
| cmyster | shardy: alo there is stuff zaneb started, you can probably add things from it | 10:05 |
| cmyster | also * | 10:05 |
| shardy | cmyster: Ah, yeah, can you link that pls if you have it bookmarked? | 10:05 |
| cmyster | I'll try and find it, it was in a mailing list | 10:06 |
| cmyster | thankfully I can just grep | 10:06 |
| jaosorior | shardy, marios, gfidente: if you guys have time to review these two patches https://review.openstack.org/#/c/231930/35 https://review.openstack.org/#/c/241200/ it would be appreciated. And the state of those is: | 10:08 |
| jaosorior | shardy, marios, gfidente: They do successfully inject the specified certificates/keys and the trust anchors (cert keys need to be specified as a list depending on the number of controllers) | 10:09 |
| *** rasca has quit IRC | 10:10 | |
| jaosorior | and the certificate is actually configured for haproxy | 10:10 |
| jaosorior | now | 10:10 |
| *** akrivoka has joined #tripleo | 10:10 | |
| jaosorior | that by itself works | 10:10 |
| jaosorior | but it's not very useful | 10:10 |
| jaosorior | because we need the keystone endpoints properly configured to use TLS | 10:10 |
| cmyster | shardy: https://etherpad.openstack.org/p/tripleo-kilo-to-liberty-upgrades | 10:10 |
| jaosorior | which is what we were attempting to do in the subsequents patches | 10:10 |
| jaosorior | but I guess the patches starting from this one: https://review.openstack.org/#/c/238887/ will be problematic since they depend on spredzy's patches that configure keystone from puppet | 10:11 |
| jaosorior | but anyway, at least https://review.openstack.org/#/c/231930/35 and https://review.openstack.org/#/c/241200/ seem to be alright regarding updates | 10:12 |
| jaosorior | I've cherry picked them to the previous patch before the upgrade issues | 10:13 |
| jaosorior | and tried the following scenarios: | 10:13 |
| jaosorior | * install overcloud with base templates without TLS and CA injection patches | 10:14 |
| jaosorior | * update overcloud with TLS and CA injection patches (without certs/keys and root CA) | 10:14 |
| tremble | jaosorior: Grah *something* has changed in the userdata.... | 10:14 |
| jaosorior | * update overcloud injecting certs/keys and root CA | 10:15 |
| jaosorior | that scenario udpates ok | 10:15 |
| jaosorior | also doing an upgrade directly putting the certs/keys from the base templates to the new ones updates without errors | 10:16 |
| tremble | jaosorior: I'm possibly updating from further back... | 10:17 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Wire Neutron ML2 plugin and OVS agent settings as arrays https://review.openstack.org/187506 | 10:18 |
| jaosorior | tremble: What commits do you have checked out? | 10:19 |
| tremble | jaosorior: OSP-D 7.1 and 805368cc3d34a4dfe24a027eef5408a0a71e69ac | 10:20 |
| tremble | jaosorior: shardy : http://paste.openstack.org/show/478069/ | 10:22 |
| *** rasca has joined #tripleo | 10:23 | |
| tremble | jaosorior: You know what I said about userdata triggering rebuilds being annoying... | 10:23 |
| tremble | The simple act of injecting | 10:24 |
| tremble | #cloud-config | 10:24 |
| *** devvesa has quit IRC | 10:24 | |
| tremble | {user: heat-admin} | 10:24 |
| shardy | tremble: that should be included by default, thus won't change on update | 10:31 |
| shardy | but I agree, it's a gotcha to look out for | 10:32 |
| tremble | shardy: From osp-d 7.1 ? | 10:32 |
| derekh | shardy: comments added, want to update or will I ? | 10:32 |
| shardy | tremble: Hmm, good point - we had to explcitly add that to cloud-config at one point because heat stopped injecting it | 10:33 |
| shardy | we need to confirm the exact timeline, that's going to be annoying if we have to exclude the userdata on update :( | 10:33 |
| tremble | Yup... | 10:34 |
| shardy | derekh: If you're happy to then go for it, thanks :) | 10:34 |
| derekh | shardy: will do | 10:34 |
| shardy | tremble: There was a blueprint for heat at one point proposing a template flag which restricted what could happen on update | 10:34 |
| shardy | that sounds like what we need here, e.g lock all server and network resources by default so they can never be updated in a destructive way | 10:35 |
| tremble | shardy: That would have been really nice... I can think of multiple use cases... | 10:35 |
| shardy | https://review.openstack.org/#/c/135444/ | 10:36 |
| shardy | the spec landed, let me look for code.. | 10:36 |
| shardy | No code :( | 10:37 |
| tremble | shardy: Trouble is that would move things to UPDATE_FAILED based on the spec... | 10:37 |
| shardy | I'll chat to ryansb about it later | 10:37 |
| shardy | tremble: that's probably OK, at least as a first pass, I'd rather have a failed upate than rebuild my entire cloud | 10:38 |
| tremble | Yeah | 10:38 |
| shardy | perhaps an "ignore" mode would also be useful tho | 10:38 |
| shardy | integrating the existing stack update --dry-run behavior into the flow will also help | 10:39 |
| shardy | at least then you can prompt with a list of what will be replaced, and the operator can bail out if it looks wrong | 10:39 |
| *** paramite is now known as paramite|afk | 10:50 | |
| *** paramite|afk is now known as paramite | 10:50 | |
| *** paramite is now known as paramite|afk | 10:50 | |
| *** adrianopetrich has joined #tripleo | 10:52 | |
| *** jerrygb has joined #tripleo | 10:58 | |
| *** masco__ has quit IRC | 11:07 | |
| *** tzumainn has joined #tripleo | 11:21 | |
| gfidente | shardy, I might be running the wrong test but I got the VirtualIP resources deleted and recreated | 11:26 |
| gfidente | shardy, want to give a look yourself? | 11:26 |
| gfidente | shardy, if you have sample templates to test the patch I could test those first to make sure the patch applied correctly | 11:29 |
| *** paramite|afk is now known as paramite | 11:48 | |
| *** Marga_ has quit IRC | 11:57 | |
| *** devvesa has joined #tripleo | 11:59 | |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Test a update in trunk repository version https://review.openstack.org/229789 | 12:02 |
| devvesa | Hi! I run a clean environment and I get the following error: | 12:05 |
| devvesa | Error: Could not find class ::ironic::inspector for instack on node instack | 12:06 |
| devvesa | And yes, /etc/puppet/modules/ironic/manifest/inspector.pp is not in my undercloud machine | 12:06 |
| *** Marga_ has joined #tripleo | 12:06 | |
| devvesa | (which I built using 'instack-virt-setup') | 12:06 |
| devvesa | Which is the best way to update it to current master branch? | 12:07 |
| devvesa | Sorry: this error happened running the `tripleo.sh --undercloud` statement | 12:08 |
| *** jerrygb has quit IRC | 12:10 | |
| shardy | devvesa: Hi! Sorry I've not seen that one, dtantsur|afk or trown|outttypeww might know when they return | 12:12 |
| slagle | devvesa: what version of of openstack-puppet-modules is on the undercloud? | 12:12 |
| shardy | devvesa: If you re-run tripleo.sh --repo-setup it should point to the latest trunk delorean | 12:13 |
| slagle | yea i was just checking the latest version there and it does have the inspector puppet classes in it | 12:13 |
| slagle | openstack-puppet-modules-2015.2-dev.28f2f4cff9b8f35511b3b15668b10cc32419c236_41994f7f.el7.centos.noarch.rpm | 12:13 |
| devvesa | weird. I run the repo-setup setp just 5 minutes before run the undercloud one | 12:14 |
| *** adrianopetrich has quit IRC | 12:15 | |
| devvesa | openstack-puppet-modules-2015.2-dev.232f6155275c8defe1d50c675d702041c108cb30_8eef2746.el7.centos.noarch | 12:16 |
| slagle | ok so you have the version from http://trunk.rdoproject.org/centos7/current-tripleo/ instead of http://trunk.rdoproject.org/centos7/current/ | 12:18 |
| slagle | the --repo-setup is supposed to set this up so that openstack-puppet-modules is preferred from the current repo via an includepkgs in the yum repo conf file | 12:18 |
| slagle | try a yum list --show-duplicates openstack-puppet-modules | 12:19 |
| slagle | and check the priorities, and for the includepkgs, in the .repo files | 12:19 |
| devvesa | So that means that running delorean-setup is enough? | 12:20 |
| devvesa | I don't need to run repo-setup? | 12:20 |
| *** thrash|g0ne is now known as thrash | 12:22 | |
| devvesa | [stack@instack yum.repos.d]$ yum list --show-duplicates | grep openstack-puppet-modules | 12:22 |
| devvesa | openstack-puppet-modules.noarch 2015.2-dev.232f6155275c8defe1d50c675d702041c108cb30_8eef2746.el7.centos | 12:22 |
| devvesa | openstack-puppet-modules.noarch 2015.2-dev.232f6155275c8defe1d50c675d702041c108cb30_8eef2746.el7.centos | 12:22 |
| *** rasca has quit IRC | 12:22 | |
| *** rasca has joined #tripleo | 12:23 | |
| *** athomas has joined #tripleo | 12:24 | |
| *** athomas has quit IRC | 12:25 | |
| slagle | you do need to run --repo-setup | 12:25 |
| *** athomas has joined #tripleo | 12:25 | |
| devvesa | I ran both, then | 12:26 |
| *** athomas has quit IRC | 12:27 | |
| *** athomas has joined #tripleo | 12:27 | |
| slagle | can you run --repo-setup -x again, and capture the output? | 12:28 |
| *** olap has quit IRC | 12:30 | |
| devvesa | http://fpaste.org/287214/44672664/ | 12:30 |
| devvesa | (this is a new clean VM) | 12:31 |
| devvesa | and after the repo-setup: | 12:32 |
| devvesa | http://fpaste.org/287216/44672673/ | 12:32 |
| devvesa | still being the dev.232f61... etc | 12:32 |
| devvesa | Can it be that I need to update the instack project? | 12:33 |
| slagle | no | 12:34 |
| *** rasca has quit IRC | 12:35 | |
| slagle | it looks like openstack-puppet-modules is gone from includepkgs, it was removed. i wasn't aware of that | 12:35 |
| slagle | so, what you need to do then is run "export DIB_INSTALLTYPE_puppet_modules=source" | 12:35 |
| *** rasca has joined #tripleo | 12:35 | |
| slagle | before tripleo.sh --undercloud | 12:35 |
| slagle | i guess that's required now | 12:35 |
| *** regebro is now known as regebro|lunch | 12:35 | |
| slagle | ok yea, tripleo.sh was updated to match what CI does. that makes sense | 12:36 |
| slagle | so, you need to run that export first | 12:37 |
| derekh | slagle: we should probably put "export DIB_INSTALLTYPE_puppet_modules=source | 12:37 |
| derekh | slagle: into tripleo.sh and remove it from toci | 12:37 |
| *** olap has joined #tripleo | 12:38 | |
| slagle | +1, default it to source, but allow people to override to package if they want to | 12:42 |
| *** yamahata has joined #tripleo | 12:45 | |
| devvesa | running the 'undercloud' now... let's see if it works | 12:46 |
| devvesa | meanwhile, can I ask you why there are two different repos in delorean? the current and the current-tripleo ? | 12:46 |
| *** Goneri has quit IRC | 12:50 | |
| *** jistr has quit IRC | 12:52 | |
| *** adrianopetrich has joined #tripleo | 12:54 | |
| *** dprince has joined #tripleo | 13:02 | |
| *** Marga_ has quit IRC | 13:03 | |
| *** paramite is now known as paramite|afk | 13:07 | |
| *** jistr has joined #tripleo | 13:11 | |
| *** Goneri has joined #tripleo | 13:11 | |
| *** jerrygb has joined #tripleo | 13:11 | |
| *** trown|outttypeww is now known as trown | 13:14 | |
| gfidente | dprince, there is something about the external lb patches and upgrades which I have been through with shardy earlier today | 13:16 |
| dprince | gfidente: okay, whats up? | 13:16 |
| *** jerrygb has quit IRC | 13:16 | |
| *** jayg|g0n3 is now known as jayg | 13:16 | |
| gfidente | dprince, mainly to make upgrades to work, in addition to a heat change, we shouldn't change the port resource name from https://github.com/openstack/tripleo-heat-templates/blob/master/network/ports/internal_api.yaml#L21 into https://github.com/openstack/tripleo-heat-templates/blob/master/network/ports/vip.yaml#L31 | 13:18 |
| gfidente | dprince, we were trying to figure what are the solutions | 13:18 |
| shardy | dprince: the problem is, even with my heat fix to not replace nested stacks but instead update them in place, we still replace the port resource | 13:18 |
| shardy | because the name of the port changes | 13:18 |
| shardy | inside the nested stack template | 13:18 |
| shardy | Heat (correctly) interprets that as a requirement for a new resource with a different name | 13:19 |
| dprince | shardy: okay, probably fine | 13:19 |
| dprince | do we have other ideas on the table? | 13:19 |
| shardy | dprince: the problem is we need to enforce naming consistency so if someone swaps around the various templates e.g in network/ports/foo.yaml we don't replace things when not needed | 13:20 |
| shardy | e.g we could enforce all port resources are called "Port" in that directory | 13:20 |
| shardy | however, folks are using the existing templates, which don't do that | 13:20 |
| shardy | gfidente: you have an etherpad with some ideas don't you? | 13:21 |
| gfidente | https://etherpad.openstack.org/p/external-lb-and-upgrades | 13:22 |
| shardy | dprince: basically the solution which won't break existing users is to duplicate vip.yaml so we maintain the name over upgrades | 13:23 |
| *** paramite|afk is now known as paramite | 13:24 | |
| gfidente | shardy, dprince which seems to go well together with the existing multiple internal_api, storage, external ports implementations, even though those are effectively duplicates as well | 13:25 |
| *** morazi has joined #tripleo | 13:27 | |
| dprince | gfidente: I think that sounds reasonable | 13:28 |
| dprince | gfidente, shardy: if we had our upgrades job in place upstream, would it have caught this as a failure? | 13:29 |
| shardy | dprince: Yes, assuming we also tested network isolation | 13:29 |
| dprince | shardy: oh right, we'd need both I guess | 13:30 |
| gfidente | shardy, so first I'd like to test the idea on the patched heat | 13:30 |
| gfidente | shardy, dprince to make sure it works | 13:30 |
| shardy | dprince: there's a ton of stuff like this we need to catch, but initially I'd like a job which deploys from t-h-t master, then does a stack-update to the posted patch | 13:30 |
| dprince | shardy: ++, lets do it | 13:31 |
| shardy | particularly for t-h-t that would catch a lot of things without even implementing the whole upgrade from stable thing | 13:31 |
| shardy | (which, obviously, we also need to test) | 13:31 |
| shardy | dprince: Ok, I'll look into it, I guess we'd need a way to mask the delorean package built for the under-review t-h-t change, deploy, then pull in the updated package and do another overcloud deploy --templates to stack-update | 13:32 |
| dprince | shardy: or just don't use a packages for the t-h-t patch in the upgrade job. Just git clone it directly, fetch the refspec, and then send the --templates location along to tripleoclient | 13:33 |
| dprince | shardy: the green field (normal, non-upgrades jobs) would catch any packaging issues w/ t-h-t I think anyways | 13:34 |
| shardy | dprince: Yeah, that may be easier | 13:36 |
| *** Goneri has quit IRC | 13:37 | |
| *** rhallisey has joined #tripleo | 13:37 | |
| shardy | things would get tricky if a t-h-t Depends-On a puppet change that is built into the image though | 13:37 |
| shardy | I'm not sure how we handle that | 13:37 |
| *** weshay_xchat has joined #tripleo | 13:38 | |
| *** panda has quit IRC | 13:40 | |
| EmilienM | marios: your comment on https://review.openstack.org/#/c/191195/26/os-apply-config/controller.yaml,cm is chineese for me | 13:41 |
| openstackgerrit | Ryan Hallisey proposed openstack/tripleo-common: Deploy Docker compute role with tripleo.sh https://review.openstack.org/232853 | 13:41 |
| *** panda has joined #tripleo | 13:41 | |
| EmilienM | marios: can you take over the patch so we can make progress on this loooooooong story? | 13:41 |
| marios | EmilienM: hey o/ i just mean you don't need to include this file in the review. we aren't updating the os-apply-config templates any longer | 13:44 |
| *** CheKoLyN has joined #tripleo | 13:45 | |
| EmilienM | ok | 13:45 |
| marios | EmilienM: sorry if the comment was confusing. but the -1 was mainly for the ports thing bnemec pointed out | 13:45 |
| EmilienM | marios: I started this patch on Jun 12 3:27 PM - can we land it soon? | 13:45 |
| marios | EmilienM: (it really doesn't matter if you do/don't include that o-a-c file just saying). we could land it right now, it would then need an immediate update for those duplicate ports | 13:46 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: Implement Advanced Firewalling support https://review.openstack.org/191195 | 13:48 |
| EmilienM | marios: done ^ | 13:48 |
| dprince | EmilienM: +2, thanks for seeing this through | 13:49 |
| EmilienM | cool my pleasure | 13:49 |
| marios | EmilienM: thanks updated | 13:49 |
| dprince | EmilienM: if, when we split up the roles... this firewall stuff looks like it will split out fairly cleanly | 13:50 |
| * dprince hopes it does anyway | 13:50 | |
| marios | am gonna approve based on v 26 green ci | 13:50 |
| marios | objections? | 13:50 |
| EmilienM | marios: I suggest to wait | 13:51 |
| dprince | marios: lets wait | 13:51 |
| shardy | Can we land https://review.openstack.org/#/c/238839/ - it's a Depends-On for the related project-config change enabling stable branch CI | 13:51 |
| shardy | please :) | 13:51 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: neutron: enable nova-event-callback by default https://review.openstack.org/225962 | 13:52 |
| marios | ack.. EmilienM hehe i was trying to appease the rage "can we land it soon?"! | 13:52 |
| dprince | shardy: +A | 13:52 |
| shardy | dprince: thanks! | 13:52 |
| openstackgerrit | Merged openstack/tripleo-common: tripleo.sh initial support for stable branches https://review.openstack.org/238839 | 13:55 |
| openstackgerrit | Merged openstack/tripleo-common: tripleo.sh initial support for stable branches https://review.openstack.org/238839 | 13:55 |
| *** akuznetsov has joined #tripleo | 13:57 | |
| *** jaosorior has quit IRC | 13:57 | |
| EmilienM | can someone look this easy one: https://review.openstack.org/#/c/240538/ ? | 13:57 |
| EmilienM | marios, dprince ^ | 13:58 |
| *** jaosorior has joined #tripleo | 13:58 | |
| EmilienM | and also https://review.openstack.org/#/c/234325/ | 13:58 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: Implement Workers parameters https://review.openstack.org/229471 | 13:58 |
| marios | EmilienM: looking | 14:00 |
| *** links has quit IRC | 14:01 | |
| dprince | shardy: FWIW, if you have 2 +2's I think it is totally fine to +A your own patch | 14:02 |
| dprince | shardy: unless it is something you want extra assurance on, feel free I think | 14:03 |
| *** jerrygb has joined #tripleo | 14:04 | |
| openstackgerrit | Merged openstack/instack-undercloud: neutron: enable nova-event-callback by default https://review.openstack.org/234325 | 14:05 |
| shardy | dprince: Ok, thanks, if it was a heat patch I probably would've done so, but I wasn't sure on the TripleO convention in that situation | 14:06 |
| tremble | Bother, jaosorior my update is failing due to https://bugzilla.redhat.com/show_bug.cgi?id=1275986 | 14:07 |
| openstack | bugzilla.redhat.com bug 1275986 in rhel-osp-director "Director update to 7.1 fails due to httpd service restart failure" [High,New] - Assigned to calfonso | 14:07 |
| *** david-lyle has joined #tripleo | 14:08 | |
| openstackgerrit | Merged openstack/tripleo-heat-templates: compute/ceilometer: use internalURL for os endpoint type https://review.openstack.org/240538 | 14:08 |
| *** rlandy has joined #tripleo | 14:09 | |
| openstackgerrit | Rhys Oxenham proposed openstack/tripleo-heat-templates: Add network templates for multiple NIC configuration https://review.openstack.org/204494 | 14:10 |
| *** rbrady-afk has quit IRC | 14:11 | |
| *** rbrady has joined #tripleo | 14:13 | |
| *** tiswanso has joined #tripleo | 14:18 | |
| openstackgerrit | Gael Chamoulaud proposed openstack/instack-undercloud: Add Puppet 4.x lint checks https://review.openstack.org/240505 | 14:22 |
| openstackgerrit | Gael Chamoulaud proposed openstack/instack-undercloud: Make puppet manifests compliant with Puppet 4.x https://review.openstack.org/240507 | 14:22 |
| *** rgowrishankar has joined #tripleo | 14:25 | |
| *** absubram has joined #tripleo | 14:26 | |
| jaosorior | tremble I see :/ | 14:26 |
| *** jaosorior has quit IRC | 14:26 | |
| *** jaosorior has joined #tripleo | 14:26 | |
| tremble | jaosorior: Unrelated to our work, so probably *shouldn't* block what we're doing. | 14:27 |
| jaosorior | tremble: Well, my update worked | 14:27 |
| jaosorior | the HA one | 14:27 |
| tremble | jaosorior: Yeah it's a race condition so sometimes it'll work, sometimes not. | 14:27 |
| tremble | jaosorior: And I was testing the Endpoints stuff too | 14:28 |
| *** lblanchard has joined #tripleo | 14:29 | |
| *** regebro|lunch is now known as regebro | 14:30 | |
| jaosorior | well, I guess now we just need reviweers | 14:32 |
| *** Goneri has joined #tripleo | 14:36 | |
| *** pradk has joined #tripleo | 14:38 | |
| *** paramite is now known as paramite|afk | 14:38 | |
| *** akuznetsov has quit IRC | 14:39 | |
| *** adrianopetrich is now known as apetrich | 14:39 | |
| *** paramite|afk is now known as paramite | 14:40 | |
| *** rlandy has quit IRC | 14:42 | |
| *** pradk has quit IRC | 14:43 | |
| *** athomas has quit IRC | 14:43 | |
| *** pradk has joined #tripleo | 14:44 | |
| *** fabbione has quit IRC | 14:44 | |
| *** fabbione has joined #tripleo | 14:45 | |
| *** athomas has joined #tripleo | 14:45 | |
| openstackgerrit | Gael Chamoulaud proposed openstack/tripleo-heat-templates: Add Puppet 4.x lint checks https://review.openstack.org/234150 | 14:53 |
| openstackgerrit | Gael Chamoulaud proposed openstack/tripleo-heat-templates: Make puppet manifests compliant with Puppet 4.x https://review.openstack.org/225664 | 14:53 |
| *** trown is now known as trown|outttypeww | 14:57 | |
| tremble | jistr: You familiar with the Pacemaker/stonith code for the controllers? | 14:58 |
| jistr | tremble: somewhat yes :) | 14:59 |
| tremble | jistr: I've just hit https://bugzilla.redhat.com/show_bug.cgi?id=1275986 | 14:59 |
| openstack | bugzilla.redhat.com bug 1275986 in rhel-osp-director "Director update to 7.1 fails due to httpd service restart failure" [High,New] - Assigned to calfonso | 14:59 |
| jistr | i think i saw some patches for that fly by IIRC | 15:00 |
| * jistr searches | 15:00 | |
| tremble | jistr: I'm wondering if we should *remove* the stonith resources if $enable_fencing == false | 15:00 |
| jistr | tremble: probably not as that could get rid of any manually created stonith resources. Anyway this is probably not related to stonith resources at all, just service resources i think. | 15:02 |
| *** tiswanso has quit IRC | 15:02 | |
| tremble | jistr: It looks like something else is running service restarts at the same time as puppet | 15:03 |
| *** tiswanso has joined #tripleo | 15:03 | |
| *** rlandy has joined #tripleo | 15:04 | |
| jistr | oh i mistook it for this https://review.openstack.org/#/c/241572/ | 15:04 |
| tremble | jistr: Only thing I can imagine restarting services would be pcs... | 15:04 |
| jistr | which is a different BZ | 15:04 |
| jistr | tremble: which service did you have problem with? httpd? as jprovazn mentions on the BZ, the slow httpd startup can be caused by some horizon pre-processing on startup | 15:05 |
| *** david-lyle has quit IRC | 15:06 | |
| tremble | jistr: The issue wasn't it being slow, the issue was that something *else* issued a service restart too... | 15:06 |
| tremble | jistr: I hit nova and httpd | 15:06 |
| jistr | hmm interesting... perhaps the timeout amendment patch i linked above could help with it | 15:08 |
| jistr | but in httpd case it's probably more of a horizon bug rather than tripleo | 15:08 |
| *** lazy_prince has quit IRC | 15:08 | |
| jistr | i remember that on a slow machine some time ago i had to disable horizon because it wasn't able to start up on the first try | 15:09 |
| *** athomas has quit IRC | 15:09 | |
| jistr | the preprocessing took too long so systemd gave up on it | 15:09 |
| tremble | jistr: http://paste.openstack.org/ | 15:09 |
| tremble | Something triggered systemctl to try and *stop* httpd while it was still coming up | 15:10 |
| openstackgerrit | Merged openstack/python-tripleoclient: Do not throw an error on validation when tunneling is disabled https://review.openstack.org/228446 | 15:11 |
| d0ugal | thrash: https://review.openstack.org/#/c/226790/ - the defaults that you have removed here, do they all match t-h-t? | 15:11 |
| thrash | d0ugal: yes | 15:12 |
| thrash | the ones that remain are the ones that differed. | 15:12 |
| d0ugal | thrash: cool, thanks | 15:12 |
| tremble | jistr: http://paste.openstack.org/show/478105/ (sorry) | 15:13 |
| d0ugal | thrash: and the args removed from the docs like --cinder-lvm, did they get removed previously but never from the docs? | 15:13 |
| thrash | d0ugal: yes | 15:13 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: neutron: enable nova-event-callback by default https://review.openstack.org/225962 | 15:13 |
| thrash | If they weren't in the code, I removed it from the docs. | 15:13 |
| d0ugal | thrash: cool, thanks - I guess we need to do a better job of looking out for that :) | 15:14 |
| thrash | hehe, yeah | 15:14 |
| thrash | d0ugal: this one too... https://review.openstack.org/#/c/235569/ | 15:15 |
| jistr | tremble: it's stopping 10:09:09, when is the start time? | 15:15 |
| tremble | jistr: It's taking about 15 seconds. | 15:16 |
| *** SlickNik has joined #tripleo | 15:18 | |
| tremble | jistr: The trigger seems to be: | 15:23 |
| tremble | Nov 5 10:09:09 localhost crmd[26468]: notice: Initiating action 34: stop httpd_stop_0 on overcloud-controller-0 (local) | 15:23 |
| *** olap has quit IRC | 15:26 | |
| jistr | ok then i'm hoping gfidente's patch i linked above might help. I didn't follow the development 100%, but if i caught it right, it effectively postpones first monitor action after stop/start of the services, so if the reason pcmk stopped it there ^^ was that it thought httpd failed to start up, it could help maybe. I'm starting to look into updates now as well, i'll see which of those issues i'll hit. | 15:28 |
| jistr | tremble: ^ | 15:28 |
| tremble | jistr: ack | 15:28 |
| tremble | jistr: I hit it because I was testing a patch which ended up restarting most of the openstack services on the controller | 15:29 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Add net_vip_map_external to be used for an external balancer https://review.openstack.org/222189 | 15:30 |
| gfidente | sorry but I'm deep down into this balancer stuff and upgrades guys :( | 15:31 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Make load balancer deployment optional via template param https://review.openstack.org/222305 | 15:32 |
| *** paramite has quit IRC | 15:34 | |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Make load balancer deployment optional via template param https://review.openstack.org/222305 | 15:35 |
| *** dprince has quit IRC | 15:37 | |
| EmilienM | slagle: if I had to install monitoring/logging tools, where would you see them? on the controller nodes? | 15:38 |
| jistr | sure thing, sorry for the pingnoise | 15:38 |
| slagle | EmilienM: the centralized services for those tools would probably need to be on the undercloud | 15:39 |
| slagle | if your monitoring tool were on the overcloud, and the overcloud goes down, that doesnt seem ideal :) | 15:40 |
| EmilienM | slagle: your undercloud is HA now? | 15:40 |
| shardy | another possiblility would be to deploy some extra nodes via a separate heat stack | 15:40 |
| shardy | and run the services on them | 15:41 |
| slagle | EmilienM: no | 15:42 |
| slagle | shardy: that could be an option. it comes down to either how big we can make the undercloud or how many nodes do we have to require | 15:43 |
| slagle | it's a tradeoff either way | 15:43 |
| shardy | slagle: yup, either approach could work depending on your requirements | 15:43 |
| slagle | i just don't think we'd want them co-located on the oc controllers | 15:44 |
| slagle | at least not until we have the fully composable roles | 15:44 |
| slagle | but then, it's not even really a controller. it's just whatever you define it to be | 15:44 |
| EmilienM | I think using undercloud makes 100% sense | 15:45 |
| *** rwsu has joined #tripleo | 15:47 | |
| *** paramite has joined #tripleo | 15:50 | |
| *** aufi has quit IRC | 15:52 | |
| *** jistr is now known as jistr|afkmtg | 15:53 | |
| *** shivrao has joined #tripleo | 15:54 | |
| *** bnemec has quit IRC | 15:55 | |
| *** yamahata has quit IRC | 15:56 | |
| *** untriaged-bot has joined #tripleo | 16:01 | |
| untriaged-bot | Untriaged bugs so far: | 16:01 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1510901 | 16:01 |
| openstack | Launchpad bug 1510901 in tripleo "seed's nova-compute service start fail" [Undecided,In progress] - Assigned to Cameron.C (cameron-c) | 16:01 |
| untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1511715 | 16:01 |
| openstack | Launchpad bug 1511715 in diskimage-builder "Images using draycut should have the option to pass drivers to include in build" [Undecided,New] - Assigned to Julia Kreger (juliaashleykreger) | 16:01 |
| untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1513508 | 16:01 |
| openstack | Launchpad bug 1513508 in diskimage-builder "Ironic IPA does not work when Swift is using https" [Undecided,New] | 16:01 |
| untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1505900 | 16:01 |
| openstack | Launchpad bug 1505900 in diskimage-builder "Failed to create 'wily' release with DIB_RELEASE" [Undecided,New] | 16:01 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1504263 | 16:01 |
| openstack | Launchpad bug 1504263 in tripleo "os-net-config mapping by MAC fails if trying to map in alternate order" [Undecided,Triaged] | 16:01 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1509148 | 16:01 |
| openstack | Launchpad bug 1509148 in tripleo "keystone initialize fails if undercloud can't reach overcloud public vip" [Undecided,New] | 16:01 |
| *** untriaged-bot has quit IRC | 16:01 | |
| *** tiswanso has quit IRC | 16:02 | |
| *** paramite has quit IRC | 16:02 | |
| *** tiswanso has joined #tripleo | 16:02 | |
| *** dsneddon has quit IRC | 16:08 | |
| *** dsneddon has joined #tripleo | 16:08 | |
| *** dsneddon has quit IRC | 16:08 | |
| *** dsneddon_ has joined #tripleo | 16:08 | |
| *** jerrygb has quit IRC | 16:13 | |
| *** jerrygb has joined #tripleo | 16:15 | |
| *** sthillma has joined #tripleo | 16:15 | |
| openstackgerrit | Mark Chappell proposed openstack/tripleo-heat-templates: Use PCS's maintenance mode while updating configuration https://review.openstack.org/242141 | 16:20 |
| openstackgerrit | Ben Nemec proposed openstack/instack-undercloud: Fix similar vm name on instack-virt-setup script. https://review.openstack.org/239957 | 16:23 |
| *** ukalifon1 has quit IRC | 16:25 | |
| *** Guest46631 has quit IRC | 16:29 | |
| *** jerrygb has quit IRC | 16:31 | |
| *** jerrygb has joined #tripleo | 16:31 | |
| *** mkovacik has quit IRC | 16:32 | |
| *** pino|work_ has joined #tripleo | 16:32 | |
| *** tosky_ has joined #tripleo | 16:32 | |
| openstackgerrit | Mark Chappell proposed openstack/tripleo-heat-templates: Use PCS's maintenance mode while updating configuration https://review.openstack.org/242141 | 16:33 |
| *** b3nt_pin has joined #tripleo | 16:35 | |
| *** b3nt_pin is now known as Guest91242 | 16:36 | |
| *** pino|work has quit IRC | 16:36 | |
| *** tiswanso has quit IRC | 16:36 | |
| *** tosky has quit IRC | 16:37 | |
| *** pino|work has joined #tripleo | 16:37 | |
| *** tosky__ has joined #tripleo | 16:37 | |
| *** tremble has quit IRC | 16:37 | |
| *** pino|work_ has quit IRC | 16:37 | |
| *** tosky_ has quit IRC | 16:38 | |
| *** mcornea has quit IRC | 16:38 | |
| *** thrash is now known as thrash|f00dz | 16:41 | |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Add network templates for multiple NIC configuration https://review.openstack.org/204494 | 16:42 |
| *** ifarkas has quit IRC | 16:46 | |
| *** tiswanso has joined #tripleo | 16:46 | |
| *** Guest91242 is now known as beagles | 16:47 | |
| *** jcooley_ has joined #tripleo | 16:51 | |
| *** jaosorior_ has joined #tripleo | 16:52 | |
| *** marios_ has joined #tripleo | 16:54 | |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Test a update in trunk repository version https://review.openstack.org/229789 | 16:54 |
| *** StevenK_ has joined #tripleo | 16:54 | |
| *** tosky__ has quit IRC | 16:55 | |
| *** beagles has quit IRC | 16:56 | |
| *** beagles has joined #tripleo | 16:57 | |
| *** shivrao has quit IRC | 16:58 | |
| *** barra204 has joined #tripleo | 16:58 | |
| *** dhellmann_ has joined #tripleo | 17:00 | |
| *** tiswanso has quit IRC | 17:01 | |
| *** tiswanso has joined #tripleo | 17:02 | |
| *** lsmola has quit IRC | 17:02 | |
| *** dprince has joined #tripleo | 17:03 | |
| *** Ng_ has joined #tripleo | 17:03 | |
| *** ChanServ sets mode: +v Ng_ | 17:03 | |
| *** jaosorior has quit IRC | 17:04 | |
| *** dhellmann has quit IRC | 17:04 | |
| *** marios has quit IRC | 17:04 | |
| *** Ng has quit IRC | 17:04 | |
| *** StevenK has quit IRC | 17:04 | |
| *** jcooley has quit IRC | 17:04 | |
| *** shakamunyi has quit IRC | 17:04 | |
| *** jcooley_ is now known as jcooley | 17:04 | |
| *** dhellmann_ is now known as dhellmann | 17:05 | |
| *** tosky__ has joined #tripleo | 17:05 | |
| *** Ng_ is now known as Ng | 17:06 | |
| *** alop has joined #tripleo | 17:07 | |
| *** apetrich has quit IRC | 17:08 | |
| *** bnemec has joined #tripleo | 17:10 | |
| *** jistr|afkmtg is now known as jistr | 17:14 | |
| *** pradk has quit IRC | 17:18 | |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Allow for usage of pre-allocated IPs for the controller nodes https://review.openstack.org/222497 | 17:20 |
| *** pradk has joined #tripleo | 17:21 | |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Add sample environment file to document usage of pre-defined IPs https://review.openstack.org/241606 | 17:21 |
| *** dsneddon_ is now known as dsneddon | 17:25 | |
| *** sthillma has quit IRC | 17:27 | |
| *** sthillma has joined #tripleo | 17:28 | |
| openstackgerrit | Ben Nemec proposed openstack/python-tripleoclient: Always install yum-plugin-priorities https://review.openstack.org/242169 | 17:34 |
| *** thrash|f00dz is now known as thrash | 17:34 | |
| * bnemec doing some hack shaving | 17:35 | |
| *** panda has quit IRC | 17:41 | |
| *** panda has joined #tripleo | 17:41 | |
| openstackgerrit | Rohit Pagedar proposed openstack/tripleo-heat-templates: Adding nova parameteres to controller: instance_name_template use_forwarded_for https://review.openstack.org/234965 | 17:41 |
| *** rpothier has joined #tripleo | 17:46 | |
| *** derekh has quit IRC | 17:48 | |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Add sample environment file to document usage of pre-defined IPs https://review.openstack.org/241606 | 17:49 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Allow for usage of pre-allocated IPs for the controller nodes https://review.openstack.org/222497 | 17:49 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Do not rename the nested port resource name for the VIPs https://review.openstack.org/242176 | 17:49 |
| gfidente | dprince, shardy turns out we didn't even need to create $network_vip.yaml becaise $network.yaml is enough https://review.openstack.org/242176 | 17:50 |
| gfidente | so that effectively maps back the new resource type into the pre-existing definition | 17:51 |
| *** sthillma has quit IRC | 17:51 | |
| *** sthillma has joined #tripleo | 17:51 | |
| dprince | gfidente: so NetName is all that was needed? | 17:53 |
| gfidente | no NetName spilled in there, it's the vip.yaml to $network.yaml all that was needed | 17:54 |
| gfidente | NetName was just a cleanup because there is no such a parameter anywhere, it's usually named something like ExternalNetName | 17:54 |
| gfidente | let me strip that out from the change | 17:55 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Do not rename the nested port resource name for the VIPs https://review.openstack.org/242176 | 17:56 |
| gfidente | done | 17:58 |
| gfidente | the port resource in vip.yaml is named VipPort, while in the per-network.yaml is named InternalApiPort or ExternalPort | 17:58 |
| *** mbound has quit IRC | 17:58 | |
| *** marcusvrn_ has joined #tripleo | 17:59 | |
| *** jistr has quit IRC | 18:00 | |
| *** sthillma has quit IRC | 18:06 | |
| *** akrivoka has quit IRC | 18:06 | |
| gfidente | shardy, one thing I'm left dealing with is this: https://github.com/openstack/tripleo-heat-templates/blob/master/overcloud-without-mergepy.yaml#L1105 | 18:06 |
| gfidente | shardy, that is an actual neutron port, not a custom type | 18:06 |
| *** tiswanso has quit IRC | 18:06 | |
| gfidente | if I needed to make it overridable from the registry, can I create it from a nested stack with "same name" ? | 18:07 |
| gfidente | ControlVirtualIP ? | 18:07 |
| openstackgerrit | Ben Nemec proposed openstack/python-tripleoclient: Always install yum-plugin-priorities https://review.openstack.org/242169 | 18:08 |
| *** tiswanso has joined #tripleo | 18:09 | |
| shardy | gfidente: No, not if you want to avoid the port being replaced | 18:13 |
| gfidente | shardy, yeah which we want to avoid | 18:13 |
| gfidente | so we're stuck with that kinda forever | 18:13 |
| shardy | the parent stack will see the type change from OS::Neutron::Port to a nested stack, which would trigger replacement | 18:13 |
| gfidente | yeah | 18:14 |
| gfidente | so I'll leave it as is | 18:14 |
| gfidente | as a result | 18:14 |
| gfidente | when deploying with external balancer we can't noop ctlplane | 18:14 |
| gfidente | vip | 18:14 |
| gfidente | but we can *not* use it though; so users will see in the undercloud neutron some control_virtual_ip which we don't use | 18:14 |
| gfidente | not terrible | 18:14 |
| shardy | gfidente: Oh, wait | 18:14 |
| shardy | if you just want to noop that one resource, that is probably possible | 18:15 |
| gfidente | yes in general yes the only purpose is to noop it I think | 18:15 |
| shardy | gfidente: yeah, I think we can do that without making a nested stack, sec | 18:15 |
| gfidente | we can work on it later | 18:15 |
| gfidente | the side-effect of not doing it is that we get an ip we don't use | 18:15 |
| gfidente | but that is only for the ctplane virtual ip | 18:16 |
| shardy | http://docs.openstack.org/developer/heat/template_guide/environment.html#override-a-resource-with-a-custom-resource | 18:16 |
| shardy | we can override just that one resource to OS::Heat::None | 18:16 |
| shardy | without changing the type in the template | 18:16 |
| gfidente | sec | 18:17 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Add net_vip_map_external to be used for an external balancer https://review.openstack.org/222189 | 18:18 |
| openstackgerrit | Merged openstack/diskimage-builder: Add missing six requirement for svc-map element https://review.openstack.org/232962 | 18:18 |
| shardy | http://paste.fedoraproject.org/287363/44674746 | 18:18 |
| shardy | gfidente: something like that | 18:18 |
| shardy | although for kilo you will need a noop.yaml instead of OS::Heat::None | 18:19 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Make load balancer deployment optional via template param https://review.openstack.org/222305 | 18:19 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Do not rename the nested port resource name for the VIPs https://review.openstack.org/242176 | 18:19 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Add sample environment file to document usage of pre-defined IPs https://review.openstack.org/241606 | 18:19 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Allow for usage of pre-allocated IPs for the controller nodes https://review.openstack.org/222497 | 18:19 |
| gfidente | Ok I think this needs testing only | 18:19 |
| gfidente | the nested resource_registry thing is very cool | 18:20 |
| shardy | Yeah, we could actually use it to reduce some of the per-role template duplication I think | 18:22 |
| *** devvesa has quit IRC | 18:23 | |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Allow for usage of pre-allocated IPs for the controller nodes https://review.openstack.org/222497 | 18:32 |
| *** weshay_xchat has quit IRC | 18:32 | |
| *** weshay_xchat has joined #tripleo | 18:34 | |
| gfidente | bnemec, how about we merge ? https://review.openstack.org/#/c/212664/9 | 18:37 |
| gfidente | we just need to make sure the params are passed via parameter_defaults: | 18:37 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Allow customization of the Ceph pool names https://review.openstack.org/212664 | 18:39 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Allow customization of Ceph client user https://review.openstack.org/212686 | 18:39 |
| gfidente | bnemec ^^ rebased | 18:39 |
| bnemec | Did they need a rebase? | 18:39 |
| gfidente | yeah the topmost | 18:40 |
| *** chem has joined #tripleo | 18:41 | |
| bnemec | gfidente: I think I'm fine with those merging. It's going to be a problem for people already using the top-level params, but that's our fault for rushing it downstream. | 18:41 |
| gfidente | well | 18:42 |
| gfidente | it's a problem only if they were using parameters; | 18:42 |
| gfidente | not if they were using parameter_defaults: | 18:42 |
| bnemec | Right, but I know at least one customer that was configuring with parameters. | 18:42 |
| gfidente | AND if they were using parameters: it won't actually break things but fail validation | 18:42 |
| bnemec | Ah, that's actually very good. | 18:43 |
| chem | hi, where can I find the source of "setup-endpoints" used in undercloud-post-config/os-refresh-config/post-configure.d/98-undercloud-setup in instack-undercloud, I've got this | 18:43 |
| chem | https://github.com/openstack/tripleo-incubator/blob/master/scripts/setup-endpoints | 18:43 |
| chem | but it doesn't take the "-s" option, so that can't be this one. Could someone point | 18:43 |
| chem | me to the right direction ? | 18:43 |
| bnemec | chem: It's in os-cloud-config now. | 18:43 |
| chem | bnemec: thank you | 18:44 |
| bnemec | https://github.com/openstack/os-cloud-config/blob/master/os_cloud_config/cmd/setup_endpoints.py | 18:44 |
| gfidente | bnemec, this one instead I am less comfortable https://review.openstack.org/#/c/210402/ | 18:44 |
| gfidente | I think puppet will notice the param was changed (even though its value is what the default was) | 18:44 |
| *** alop has quit IRC | 18:45 | |
| gfidente | and try a config refresh | 18:45 |
| gfidente | needs testing though | 18:45 |
| bnemec | gfidente: Yeah, I'd say let's hold off on that until we have the upgrade job. | 18:45 |
| bnemec | The redis one too. | 18:45 |
| gfidente | redis sure would break yes because ceilometer coordingation_url is changed | 18:46 |
| gfidente | coordination_url | 18:46 |
| *** regebro has quit IRC | 18:46 | |
| bnemec | It's something we obviously need to do, but we also need to make sure there's some sort of upgrade path available. | 18:46 |
| gfidente | yeah leaving for today, have fun | 18:46 |
| bnemec | Have a good one. | 18:47 |
| *** gfidente has quit IRC | 18:48 | |
| *** alop has joined #tripleo | 18:53 | |
| chem | bnemec: sorry to bother you again. Then in instack-deploy-overcloud (in instack-overcloud), "tripleo setup-endpoints" is used. It seems to match the interface of the aforementioned script is that correct or is it another script ? | 18:55 |
| bnemec | chem: We're not using instack-deploy-overcloud anymore. The endpoint setup is done either in tripleoclient now or by puppet (a change just merged to do the latter, but it will probably have to be reverted because it breaks upgrades). | 18:56 |
| bnemec | chem: The current workflow we're using is documented at http://docs.openstack.org/developer/tripleo-docs/ | 18:57 |
| *** jerrygb has quit IRC | 18:57 | |
| *** jerrygb has joined #tripleo | 18:58 | |
| bnemec | chem: You might also be interested in tripleo.sh: https://github.com/openstack/tripleo-common/blob/master/scripts/tripleo.sh | 18:58 |
| bnemec | It scripts a lot of the steps from the docs. | 18:58 |
| chem | bnemec: oki, thanks I have a look. My goal would be to fix an old "bug", https://bugzilla.redhat.com/show_bug.cgi?id=1236989, so I grep around to find a fix. | 18:59 |
| openstack | bugzilla.redhat.com bug 1236989 in rhel-osp-director "RFE: change computev3 service name to novav3" [Unspecified,New] - Assigned to calfonso | 18:59 |
| bnemec | chem: Okay, the old place to fix it would have been https://github.com/openstack/os-cloud-config/blob/master/os_cloud_config/keystone.py#L70 | 19:02 |
| bnemec | I'm not sure where it's defined in puppet. | 19:02 |
| bnemec | But puppet may not have had the bug in the first place, so it may not need to be fixed there. | 19:02 |
| chem | bnemec: yeah, that's exactly that :) | 19:03 |
| *** jerrygb has quit IRC | 19:03 | |
| bnemec | Although we should probably just stop registering that endpoint, since nova v3 is no longer a thing. | 19:03 |
| *** jprovazn has quit IRC | 19:03 | |
| chem | bnemec: puppet-keystone module currently chocks on service which are different only by type | 19:04 |
| chem | bnemec: here you have nova/computev3 and nova/compute | 19:04 |
| *** shivrao has joined #tripleo | 19:04 | |
| chem | bnemec: puppet-keystone will be fixed but it's a non backward comptatible fix and may takes some times | 19:04 |
| chem | bnemec: in the meantime a simple fix would be to follow the convention described in the bug report : novav3/computev3 and nova/compute | 19:05 |
| bnemec | chem: Right, although I still think we need to completely remove the v3 endpoint anyway. Nova decided against pursuing their v3 api so I can't imagine the endpoint actually works anymore. | 19:07 |
| bnemec | I'm fine with just fixing the name in the meantime though. | 19:07 |
| chem | bnemec: either way is ok by me, my ultimate goal is to quickly unlock this https://review.openstack.org/#/c/234704 | 19:08 |
| chem | (quickly == not waiting the change in puppet to have been accepted) | 19:09 |
| chem | bnemec: so it's ok if I propose a patch that fix the the rhbz and you can decide later if the removal is appropriate, even if your point is very strong: no api, no endpoint, makes a lot of senses :) | 19:11 |
| bnemec | chem: Yeah, that's fine. | 19:12 |
| chem | bnemec: great | 19:12 |
| openstackgerrit | Dan Prince proposed openstack/instack-undercloud: Set nova ironic admin_url correctly https://review.openstack.org/242195 | 19:14 |
| *** electrofelix has quit IRC | 19:15 | |
| openstackgerrit | Dan Prince proposed openstack/diskimage-builder: Use ironic-agent for source-repositories https://review.openstack.org/233088 | 19:17 |
| *** rasca has quit IRC | 19:17 | |
| *** rasca has joined #tripleo | 19:18 | |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/os-cloud-config: Change nova service's name for v3 to novav3. https://review.openstack.org/242198 | 19:24 |
| chem | bnemec: ^ | 19:25 |
| *** tosky__ has quit IRC | 19:26 | |
| jaosorior_ | Hey bnemec, do you have some time to review the TLS patch? | 19:26 |
| jaosorior_ | https://review.openstack.org/#/c/231930/ | 19:27 |
| *** Marga_ has joined #tripleo | 19:30 | |
| *** noslzzp has quit IRC | 19:31 | |
| *** adrianopetrich has joined #tripleo | 19:33 | |
| openstackgerrit | Mark Chappell proposed openstack/tripleo-heat-templates: Use PCS's maintenance mode while updating configuration https://review.openstack.org/242141 | 19:33 |
| *** tremble has joined #tripleo | 19:34 | |
| *** tremble has joined #tripleo | 19:34 | |
| jaosorior_ | tremble: was that the fix for the bug we were talking about earlier? O_O | 19:36 |
| tremble | jaosorior_: With a little luck, yes | 19:37 |
| jaosorior_ | duuuude | 19:37 |
| jaosorior_ | you da man | 19:37 |
| tremble | We'll see if it works, or if the more senior devs hate it. | 19:37 |
| jaosorior_ | tremble, If you get the chanse, can you score the root CA injection patch? | 19:38 |
| tremble | jaosorior_: Sure, could you take a quick look at https://review.openstack.org/#/c/241574/ too | 19:39 |
| jaosorior_ | sure | 19:40 |
| jaosorior_ | tremble: That looks quite straight forward | 19:41 |
| tremble | jaosorior_: Yup, just adds some outputs that show which certs/keys were injected in a way that doesn't compromise them | 19:42 |
| jaosorior_ | does the pacemaker puppet module have no way to se the maintenance mode? | 19:45 |
| tremble | nope, ultimately it should move in there | 19:46 |
| jaosorior_ | ok, looks good to me in the meantime | 19:46 |
| tremble | Yeah if it works I probably need to figure out how to update the commit message to reference the appropriate bugs too | 19:48 |
| jaosorior_ | tremble is there more than one bug reported for this>? | 19:49 |
| tremble | There's a Red Hat BZ, dunno if/what the tripleo bug might be | 19:50 |
| jaosorior_ | you can reference the red hat BZ to the patch | 19:51 |
| *** mbound has joined #tripleo | 19:53 | |
| *** jerrygb has joined #tripleo | 19:53 | |
| *** aufi has joined #tripleo | 19:58 | |
| *** jaosorior_ has quit IRC | 19:58 | |
| *** johnsom has joined #tripleo | 19:58 | |
| *** jobewan has joined #tripleo | 20:00 | |
| *** jaosorior has joined #tripleo | 20:00 | |
| johnsom | Hi, I'm with the Octavia team. We use diskimage-builder to generate our amphora service vm images. In the past week or two the virtual size as reported by qemu-img has gone from 2147483648 bytes to 2214658048 bytes when specifying --image-size 2. Would anyone have a pointer as to what changed? | 20:01 |
| johnsom | Nova will no longer boot these images with a 2GB flavor | 20:02 |
| *** Marga_ has quit IRC | 20:03 | |
| *** Marga_ has joined #tripleo | 20:03 | |
| *** aufi has quit IRC | 20:04 | |
| *** jaosorior has quit IRC | 20:04 | |
| *** noslzzp has joined #tripleo | 20:13 | |
| *** Marga_ has quit IRC | 20:20 | |
| *** Marga_ has joined #tripleo | 20:21 | |
| *** ryansb has quit IRC | 20:26 | |
| *** Marga_ has quit IRC | 20:26 | |
| *** ryansb has joined #tripleo | 20:28 | |
| *** ryansb has joined #tripleo | 20:28 | |
| openstackgerrit | Pierre Blanc proposed openstack/tripleo-heat-templates: Fix the wrong selinux context when glance use nfs backend https://review.openstack.org/242211 | 20:34 |
| *** dsneddon is now known as dsneddon_biab | 20:34 | |
| *** adrianopetrich has quit IRC | 20:43 | |
| openstackgerrit | Mark Chappell proposed openstack/tripleo-heat-templates: Use PCS's maintenance mode while updating configuration https://review.openstack.org/242141 | 20:46 |
| *** lblanchard has quit IRC | 20:58 | |
| openstackgerrit | Pierre Blanc proposed openstack/tripleo-heat-templates: Fix the wrong selinux context when glance uses nfs backend https://review.openstack.org/242211 | 21:14 |
| bnemec | johnsom: My guess would be this: https://github.com/openstack/diskimage-builder/commit/f84b910f152220294a723836a71586974efadede | 21:14 |
| johnsom | Thank you! I will check that out. I was just pulling a commit ID from a last known good git log. | 21:15 |
| bnemec | Maybe we need to use 1000 as the base for calculating the size in K? I'm not sure how the old truncate call would have been doing that. | 21:16 |
| johnsom | Yeah, not sure. I think we do need to make sure that 2GB in diskimage-builder == 2GB in nova flavor | 21:17 |
| bnemec | But I would definitely call it a bug if you build a 2 GB image and it won't boot in a 2 GB flavor. | 21:17 |
| bnemec | Yeah | 21:17 |
| johnsom | It worked until recently | 21:17 |
| johnsom | I will open a bug while I try to track it down better. | 21:19 |
| bnemec | Yeah, that change only merged a couple of weeks ago. I'm guessing that's when you started having trouble with it. | 21:19 |
| bnemec | Sounds good, thanks. | 21:19 |
| bnemec | johnsom: Oh, I bet it's because of the $du_size + 65536 that is being run even when image size is specified now. | 21:21 |
| bnemec | Basically we're adding 65 MB to the image size unconditionally, which is about what the difference you mentioned above is. | 21:22 |
| johnsom | It is close | 21:23 |
| *** dsneddon_biab is now known as dsneddon | 21:25 | |
| *** jayg is now known as jayg|g0n3 | 21:25 | |
| bnemec | We should probably just put the ext4 block that's messing with size and mkfs opts back into the else so it doesn't get run when an image size is specified. | 21:26 |
| johnsom | https://bugs.launchpad.net/diskimage-builder/+bug/1513622 | 21:29 |
| openstack | Launchpad bug 1513622 in diskimage-builder "diskimage-builder 2GB image does not boot in 2GB flavor" [Undecided,New] | 21:29 |
| johnsom | bnemec I just ran through the numbers. This is the issue. I might take a pass at a patch. | 21:37 |
| openstackgerrit | Ben Nemec proposed openstack/tripleo-common: Remove yum-plugin-priorities hack https://review.openstack.org/242228 | 21:39 |
| bnemec | johnsom: Cool, if you push it then I can +2 it. :-) | 21:39 |
| *** dprince has quit IRC | 21:41 | |
| *** panda has quit IRC | 21:41 | |
| *** panda has joined #tripleo | 21:42 | |
| *** absubram has quit IRC | 21:44 | |
| johnsom | bnemec What do you think about removing the + 65536 on the size but leave the journal settings outside? It just means that the usable space is slightly smaller, but the image size is what was requested. | 21:45 |
| johnsom | I think the journal setting is valid and want to retain that | 21:46 |
| *** tiswanso has quit IRC | 21:46 | |
| *** noslzzp has quit IRC | 21:49 | |
| bnemec | johnsom: I'd be okay with that. The journal setting is designed to be "safe", so it probably makes sense to always apply it. | 21:49 |
| johnsom | Ok, cool. Testing now | 21:50 |
| *** shardy has quit IRC | 21:58 | |
| *** untriaged-bot has joined #tripleo | 22:00 | |
| untriaged-bot | Untriaged bugs so far: | 22:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1510901 | 22:00 |
| openstack | Launchpad bug 1510901 in tripleo "seed's nova-compute service start fail" [Undecided,In progress] - Assigned to Cameron.C (cameron-c) | 22:00 |
| *** rhallisey has quit IRC | 22:00 | |
| untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1511715 | 22:00 |
| openstack | Launchpad bug 1511715 in diskimage-builder "Images using draycut should have the option to pass drivers to include in build" [Undecided,New] - Assigned to Julia Kreger (juliaashleykreger) | 22:00 |
| untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1513508 | 22:00 |
| openstack | Launchpad bug 1513508 in diskimage-builder "Ironic IPA does not work when Swift is using https" [Undecided,New] | 22:00 |
| untriaged-bot | https://bugs.launchpad.net/diskimage-builder/+bug/1505900 | 22:00 |
| openstack | Launchpad bug 1505900 in diskimage-builder "Failed to create 'wily' release with DIB_RELEASE" [Undecided,New] | 22:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1504263 | 22:00 |
| openstack | Launchpad bug 1504263 in tripleo "os-net-config mapping by MAC fails if trying to map in alternate order" [Undecided,Triaged] | 22:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1509148 | 22:00 |
| openstack | Launchpad bug 1509148 in tripleo "keystone initialize fails if undercloud can't reach overcloud public vip" [Undecided,New] | 22:00 |
| *** untriaged-bot has quit IRC | 22:00 | |
| openstackgerrit | Pierre Blanc proposed openstack/tripleo-heat-templates: Fix the wrong selinux context when glance uses nfs backend https://review.openstack.org/242211 | 22:00 |
| *** thrash is now known as thrash|g0ne | 22:03 | |
| *** weshay_xchat has quit IRC | 22:08 | |
| *** tzumainn has quit IRC | 22:09 | |
| *** tzumainn has joined #tripleo | 22:13 | |
| *** rpothier has quit IRC | 22:14 | |
| *** Goneri has quit IRC | 22:15 | |
| *** Marga_ has joined #tripleo | 22:20 | |
| *** weshay_xchat has joined #tripleo | 22:23 | |
| *** pradk has quit IRC | 22:25 | |
| *** cwolferh has quit IRC | 22:29 | |
| *** alop has quit IRC | 22:50 | |
| *** wfoster has quit IRC | 22:51 | |
| *** alop has joined #tripleo | 22:51 | |
| *** wfoster has joined #tripleo | 22:52 | |
| *** morazi has quit IRC | 22:56 | |
| *** jerrygb has quit IRC | 23:08 | |
| *** StevenK_ is now known as StevenK | 23:21 | |
| *** nibalizer has quit IRC | 23:24 | |
| *** jobewan has quit IRC | 23:32 | |
| *** jerrygb has joined #tripleo | 23:33 | |
| *** sthillma has joined #tripleo | 23:35 | |
| openstackgerrit | Pierre Blanc proposed openstack/tripleo-heat-templates: Fix the wrong selinux context when glance uses nfs backend https://review.openstack.org/242211 | 23:35 |
| *** Marga__ has joined #tripleo | 23:37 | |
| *** Marga_ has quit IRC | 23:40 | |
| openstackgerrit | Shiva Prasad Rao proposed openstack/tripleo-heat-templates: Add support to tag patch port for Cisco Nexus1000v VSM bringup https://review.openstack.org/242265 | 23:41 |
| openstackgerrit | Rhys Oxenham proposed openstack/tripleo-heat-templates: [WIP] Adding in support for automated Calamari Integration https://review.openstack.org/242267 | 23:44 |
| *** jobewan has joined #tripleo | 23:44 | |
| *** jerrygb_ has joined #tripleo | 23:49 | |
| *** jerrygb has quit IRC | 23:50 | |
| *** alop_ has joined #tripleo | 23:52 | |
| *** alop has quit IRC | 23:54 | |
| *** alop_ is now known as alop | 23:54 | |
| *** julim has quit IRC | 23:56 | |
| *** alop has quit IRC | 23:57 | |
| *** julim has joined #tripleo | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!