| *** noslzzp has joined #tripleo | 01:14 | |
| *** pcrews has quit IRC | 01:43 | |
| *** lblanchard has quit IRC | 02:24 | |
| *** noslzzp has quit IRC | 02:24 | |
| *** adam_g has quit IRC | 02:43 | |
| *** adam_g has joined #tripleo | 02:44 | |
| *** adam_g has joined #tripleo | 02:44 | |
| *** ddieterl_ has quit IRC | 02:54 | |
| *** untriaged-bot has joined #tripleo | 03:00 | |
| untriaged-bot | Untriaged bugs so far: | 03:00 |
|---|---|---|
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1351263 | 03:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1354509 | 03:00 |
| uvirtbot | Launchpad bug 1351263 in tripleo "Seed init failed, nova list returned 503" [Undecided,New] | 03:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1347848 | 03:00 |
| uvirtbot | Launchpad bug 1354509 in tripleo "Glance requirements.txt for osprofiler too loose" [Undecided,Confirmed] | 03:00 |
| uvirtbot | Launchpad bug 1347848 in tripleo "/etc/hosts information about cloud hsots lost on reboot" [Undecided,In progress] | 03:00 |
| *** untriaged-bot has quit IRC | 03:00 | |
| *** eghobo has joined #tripleo | 03:17 | |
| *** eghobo has quit IRC | 04:03 | |
| *** eghobo has joined #tripleo | 04:23 | |
| *** cwolfe__ has joined #tripleo | 04:33 | |
| *** cwolferh_ has quit IRC | 04:37 | |
| openstackgerrit | James Polley proposed a change to openstack/tripleo-incubator: Add a second ssh key for tchaypo https://review.openstack.org/113110 | 05:15 |
| *** dshulyak_ has joined #tripleo | 05:17 | |
| *** eghobo has quit IRC | 05:29 | |
| lifeless | adam_g: you're welcome :) | 05:32 |
| openstackgerrit | A change was merged to openstack/tripleo-heat-templates: Fix overcloud controller scaling https://review.openstack.org/112538 | 05:45 |
| tchaypo | lifeless: so I've been poking at the bastions | 05:50 |
| tchaypo | If I haven't missed anything, it looks like the things we might need to update aren't in that repo - we just need to get that repo up-to-date on hp2. It looks like all the config it's going to use lives one level higher | 05:51 |
| lifeless | thats the intent yes | 05:51 |
| lifeless | but the repo refs file is probably still in the repo on hp2 | 05:51 |
| lifeless | so checking for git diff output would be a good starting point | 05:51 |
| tchaypo | it's still in the repo on hp1 too *thinks more* wait, you mean, look for what's changed on hp2 between the git revision and what's live | 05:53 |
| tchaypo | hp2 has a longer list of patches, which look roughly like the list of things we'd need to get the vlan stuff working | 05:55 |
| tchaypo | so I'm guessing we want to commit then; then merge with what's on hp1? | 05:55 |
| *** vinsh has quit IRC | 06:04 | |
| *** cwolferh has joined #tripleo | 06:06 | |
| *** cwolfe__ has quit IRC | 06:10 | |
| *** eghobo has joined #tripleo | 06:19 | |
| *** akuznetsov has joined #tripleo | 07:03 | |
| *** eghobo has quit IRC | 07:45 | |
| *** tchaypo has quit IRC | 07:54 | |
| *** tchaypo has joined #tripleo | 07:54 | |
| *** akuznetsov has quit IRC | 08:05 | |
| *** akuznetsov has joined #tripleo | 08:20 | |
| *** akuznetsov has quit IRC | 08:25 | |
| *** akuznetsov has joined #tripleo | 08:27 | |
| *** akuznetsov has quit IRC | 08:31 | |
| *** jmoffitt has quit IRC | 08:47 | |
| *** jmoffitt has joined #tripleo | 08:48 | |
| *** untriaged-bot has joined #tripleo | 09:00 | |
| untriaged-bot | Untriaged bugs so far: | 09:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1351263 | 09:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1354509 | 09:00 |
| uvirtbot | Launchpad bug 1351263 in tripleo "Seed init failed, nova list returned 503" [Undecided,New] | 09:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1347848 | 09:00 |
| uvirtbot | Launchpad bug 1354509 in tripleo "Glance requirements.txt for osprofiler too loose" [Undecided,Confirmed] | 09:00 |
| uvirtbot | Launchpad bug 1347848 in tripleo "/etc/hosts information about cloud hsots lost on reboot" [Undecided,In progress] | 09:00 |
| *** untriaged-bot has quit IRC | 09:00 | |
| *** jang1 has quit IRC | 10:25 | |
| *** jang has quit IRC | 10:25 | |
| *** jang has joined #tripleo | 10:25 | |
| *** jang has quit IRC | 10:28 | |
| *** rlandy has joined #tripleo | 10:30 | |
| *** jang has joined #tripleo | 10:31 | |
| *** jang has quit IRC | 10:32 | |
| *** jang has joined #tripleo | 10:33 | |
| *** jang has quit IRC | 10:36 | |
| *** jang has joined #tripleo | 10:39 | |
| *** jang has quit IRC | 10:40 | |
| *** jang has joined #tripleo | 10:41 | |
| *** akuznetsov has joined #tripleo | 10:55 | |
| *** jang1 has joined #tripleo | 12:12 | |
| *** dshulyak_ has quit IRC | 12:33 | |
| *** boris-42 has quit IRC | 14:02 | |
| *** dshulyak_ has joined #tripleo | 14:04 | |
| *** noslzzp has joined #tripleo | 14:57 | |
| *** untriaged-bot has joined #tripleo | 15:00 | |
| untriaged-bot | Untriaged bugs so far: | 15:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1351263 | 15:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1354509 | 15:00 |
| uvirtbot | Launchpad bug 1351263 in tripleo "Seed init failed, nova list returned 503" [Undecided,New] | 15:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1347848 | 15:00 |
| *** untriaged-bot has quit IRC | 15:00 | |
| uvirtbot | Launchpad bug 1354509 in tripleo "Glance requirements.txt for osprofiler too loose" [Undecided,Confirmed] | 15:00 |
| uvirtbot | Launchpad bug 1347848 in tripleo "/etc/hosts information about cloud hsots lost on reboot" [Undecided,In progress] | 15:00 |
| *** dshulyak_ has quit IRC | 15:03 | |
| *** noslzzp has quit IRC | 15:10 | |
| *** Leslie_ has joined #tripleo | 15:14 | |
| Leslie_ | Hello Everyone | 15:14 |
| Leslie_ | I'm studying tripleo for now | 15:14 |
| Leslie_ | I got one question: | 15:15 |
| Leslie_ | when building seed VM, /etc/os-collect-config.conf only has default command, but after VM boot, seems like some applications append collector=heat_local, then seems like cfn-init-data are got triggered | 15:16 |
| Leslie_ | so does anyone know the detail work flow | 15:16 |
| Leslie_ | I'm curious which applications read /var/lib/heat-cfn-tools/cfn-init-data at when | 15:18 |
| *** rlandy has quit IRC | 16:35 | |
| *** eghobo has joined #tripleo | 16:37 | |
| *** eghobo has quit IRC | 16:48 | |
| *** rwsu has quit IRC | 16:50 | |
| *** dshulyak_ has joined #tripleo | 17:37 | |
| *** dshulyak_ has quit IRC | 17:53 | |
| *** noslzzp has joined #tripleo | 17:55 | |
| *** noslzzp has quit IRC | 18:07 | |
| *** noslzzp has joined #tripleo | 18:16 | |
| *** noslzzp has quit IRC | 18:28 | |
| *** pensu has joined #tripleo | 19:08 | |
| *** penick has joined #tripleo | 19:15 | |
| *** jml has quit IRC | 19:18 | |
| *** jml has joined #tripleo | 19:19 | |
| *** dshulyak_ has joined #tripleo | 19:26 | |
| *** dshulyak_ has quit IRC | 19:33 | |
| *** dshulyak_ has joined #tripleo | 19:34 | |
| *** pensu has quit IRC | 19:34 | |
| *** pensu has joined #tripleo | 19:35 | |
| *** akuznetsov has quit IRC | 19:37 | |
| *** eghobo has joined #tripleo | 19:41 | |
| *** eghobo has quit IRC | 19:41 | |
| *** eghobo has joined #tripleo | 19:42 | |
| *** pensu has quit IRC | 19:50 | |
| *** dshulyak_ has quit IRC | 19:54 | |
| *** dshulyak_ has joined #tripleo | 19:59 | |
| *** dshulyak_ has quit IRC | 19:59 | |
| *** openstack has joined #tripleo | 20:17 | |
| *** eguz has joined #tripleo | 20:36 | |
| *** eghobo has quit IRC | 20:38 | |
| *** penick has quit IRC | 20:40 | |
| *** robsparker has quit IRC | 20:42 | |
| *** robsparker has joined #tripleo | 20:57 | |
| *** untriaged-bot has joined #tripleo | 21:00 | |
| untriaged-bot | Untriaged bugs so far: | 21:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1351263 | 21:00 |
| uvirtbot | Launchpad bug 1351263 in tripleo "Seed init failed, nova list returned 503" [Undecided,New] | 21:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1354509 | 21:00 |
| uvirtbot | Launchpad bug 1354509 in tripleo "Glance requirements.txt for osprofiler too loose" [Undecided,Confirmed] | 21:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1347848 | 21:00 |
| uvirtbot | Launchpad bug 1347848 in tripleo "/etc/hosts information about cloud hsots lost on reboot" [Undecided,In progress] | 21:00 |
| *** untriaged-bot has quit IRC | 21:00 | |
| *** boris-42 has joined #tripleo | 21:15 | |
| tchaypo | morning lifeless | 21:19 |
| openstackgerrit | Steve Baker proposed a change to openstack/tripleo-heat-templates: Switch to heat_template_version: 2013-05-23 https://review.openstack.org/113148 | 21:21 |
| lifeless | tchaypo: morning! | 21:25 |
| lifeless | tchaypo: lets deploy a cloud | 21:25 |
| lifeless | tchaypo: commit what? | 21:30 |
| tchaypo | repo_refs.yaml - it looks like we need to merge between current repo state on hp2, current working state on hp2, and current repo state on hp1 | 21:34 |
| tchaypo | well, lots of stuff on hp1 needs to be updated on hp2, but it looks like we need to also merge in the currently-live repo_refs.yaml, which looks to me as though it's been set up for the vlan work we've been doing | 21:35 |
| tchaypo | *thinks* this would imply that we want to have consistent repo_refs across all regions. I'm pretty sure that's accurate - I think regional differences are going to be dealt with as config elsewhere, not as which patches are being applied? | 21:36 |
| *** jang1 has quit IRC | 21:48 | |
| lifeless | tchaypo: config is patches is config | 21:49 |
| lifeless | tchaypo: repo_refs.yaml in the tripleo-end-to-end repo in hp1 is no longer relevant | 21:49 |
| lifeless | tchaypo: in the same place on hp2 its relevant, because thats how the old tripleo-end-to-end works | 21:50 |
| tchaypo | the hp2 version looks like what we want | 21:50 |
| lifeless | tchaypo: we need to migrate it out of there into the same structure as hp1 uses | 21:50 |
| lifeless | tchaypo: and then reconcile the two sets of patches | 21:50 |
| tchaypo | specifically - the format for repos: has changed, the gerrit_refs structure looks the same | 21:51 |
| lifeless | also where we store the refs | 21:52 |
| lifeless | like the file path on disk | 21:52 |
| lifeless | totally different | 21:52 |
| tchaypo | so: I'll commit what's on hp2 now, then do another commit with the same content but the new format for repos, then add a new upstream pointing at the same place as the hp1 upstream, then merge.. | 21:52 |
| lifeless | nooo | 21:52 |
| lifeless | please | 21:52 |
| lifeless | lets get consensus on next step first | 21:53 |
| lifeless | whats the file path to repo_refs.yaml on hp1 | 21:53 |
| lifeless | whats the file path to repo_refs.yaml on hp2 | 21:53 |
| tchaypo | both have /home/shared/tripleo-end-to-end/repo_refs.yaml; hp1 also has /home/shared/repo_refs.yaml | 21:54 |
| tchaypo | agree about getting consensus first. above was me trying to start that by proposing a series of steps. maybe i made it sound like I'd already reached a decision? | 21:54 |
| lifeless | yah | 21:55 |
| lifeless | ok so /home/shared/tripleo-end-to-end/repo_refs.yaml is irrelevant in hp1 | 21:55 |
| lifeless | see the README - its an example file | 21:55 |
| tchaypo | the ../repo_refs.yaml on hp1 is newer by about 90 minutes | 21:55 |
| tchaypo | ah right. So where I was speculating about whether we *want* to have the same repo_refs in all regions, the answer is "no" | 21:56 |
| tchaypo | which I think is what you meant by "config is patches is config" | 21:56 |
| lifeless | some of our config | 21:56 |
| lifeless | is stored in tripleo-image-elements | 21:56 |
| tchaypo | hence upating the patches to use for that repo == updating the config to use for this region | 21:57 |
| lifeless | yes | 21:57 |
| tchaypo | I'd like to git init /home/shared; give it a .gitignore of .*, then manually add in config files like repo_refs.yaml so that I can track and revert my changes. any objection? | 22:01 |
| lifeless | uhm | 22:02 |
| lifeless | yeah | 22:02 |
| lifeless | if we want to version track meta config like this | 22:02 |
| lifeless | we shouldn't make it a containing dir of other things | 22:02 |
| lifeless | as thats an awesome way to confuse people | 22:03 |
| tchaypo | better to make a subdir called regional_configs (or something), track that, and make the files at the top level symlinks into the repo? | 22:03 |
| lifeless | I've no objection to having local history of this stuff (though by definition, since its the thing tracking what we haven't committed, it can't ever be CId etc) | 22:03 |
| lifeless | tchaypo: we *have* regional configs. | 22:04 |
| lifeless | tchaypo: in tie | 22:04 |
| tchaypo | you're talking about elements/tripleo-cd/configs? the files with "hp1" in the name? | 22:05 |
| lifeless | yes | 22:05 |
| lifeless | and the ones with hp2 etc | 22:05 |
| tchaypo | hp1 hasn't been updated to have the commit that added those yet | 22:06 |
| lifeless | they haven't landed yet | 22:06 |
| lifeless | and hp1 doesn't need that config | 22:06 |
| tchaypo | I think that if we tracked repo_refs.yaml in there it would lead to what I'm going to call "fun" but which a person more interested in a sane working system might call "not working" | 22:07 |
| lifeless | tchaypo: so, if you want to vc these files - and like I say, I've no objection, I would be ok with a new subdir 'repo_refs' and put the file in there | 22:07 |
| lifeless | right | 22:08 |
| lifeless | repo_refs can't be sensibly versioned in the product trees | 22:08 |
| lifeless | since its the 'and this is why our product doesn't work right now' set of patches | 22:08 |
| * tchaypo notes the rollup_1_2_3_4 branch names | 22:08 | |
| * tchaypo approves of how easy this makes it to compare state between regions | 22:08 | |
| lifeless | so with your new understanding, whats your proposed action plan? | 22:10 |
| tchaypo | btw I took the liberty of adding some aliases to ~/.gitconfig on both bastions | 22:10 |
| tchaypo | create /home/shared/repo_refs, copy repo_refs.yaml in, commit. Modify to new format based on reading hp1's sample, commit, ask for review. Add new remote to hp2, fetch changes in the new remote, do a diff and make sure it looks as I expect (ie, should be identical to " git diff -r f5fc0e7b -r fbdbef8" on hp1, except for the changes to repo_refs.yaml on disk) | 22:12 |
| tchaypo | no wait - once I've copied repo_refs off to the new location I can revert the one in tripleo-end-to-end to match the repo, which will simplify the diff. | 22:13 |
| lifeless | by simplify you mean 'should be no diff' ? | 22:13 |
| tchaypo | At that point I'm going to pause and re-read the readme and take a look at the config files under /home/shared | 22:13 |
| tchaypo | yes, should be no diff :) | 22:13 |
| lifeless | so that sounds fine to me; please change hp1 in the same way of course | 22:23 |
| tchaypo | what, we want consistency? *gasp* | 22:27 |
| tchaypo | I know we have some puppet configs, which I *think* touch on these machines, but I haven't looked. | 22:29 |
| lifeless | nope | 22:31 |
| lifeless | we aren't operating these with puppet at all | 22:31 |
| tchaypo | nope, looking at site.pp in the config repo it doesn't seem to touch these machines | 22:31 |
| tchaypo | and there's no /etc/puppet | 22:31 |
| *** eguz has quit IRC | 22:32 | |
| lifeless | tchaypo: so while you mull on that | 22:55 |
| lifeless | I'm going to revert the HOT patches | 22:55 |
| lifeless | ah stevebaker has proposed it. Yay | 22:56 |
| stevebaker | lifeless: yes, take a look | 22:57 |
| stevebaker | check looks happy | 22:57 |
| lifeless | tchaypo: I'm taking a lock on hp1 | 23:00 |
| tchaypo | ack | 23:01 |
| lifeless | tchaypo: the state there seems inconsistent - there is a repo_refs subdir now, but still a repo_refs.yaml above it | 23:01 |
| lifeless | tchaypo: where were you up to ? | 23:01 |
| tchaypo | yes, I got interrupted partway through making that repo | 23:01 |
| lifeless | tchaypo: ok | 23:01 |
| lifeless | tchaypo: I will finish up, since I am trying to recover this :) | 23:02 |
| lifeless | tchaypo: when you say interrupted | 23:02 |
| tchaypo | repo_refs now has git history | 23:02 |
| lifeless | tchaypo: what specifically do you mean ? | 23:02 |
| tchaypo | if you make changes in the parent I can copy them in, make a new commit, and then pivot to a symlink | 23:03 |
| tchaypo | or would you prefer to pivot first? | 23:03 |
| lifeless | tchaypo: what symlink? | 23:03 |
| tchaypo | interrupted by cats walking past | 23:03 |
| tchaypo | repo_refs.yaml -> repo_refs/repo_refs.yaml | 23:03 |
| lifeless | tchaypo: I had $ git add repo_refs.yaml | 23:03 |
| lifeless | fatal: Unable to create '/home/shared/repo_refs/.git/index.lock': File exists. | 23:03 |
| lifeless | tchaypo: which makes me think that you didn't down tools after I took the lock :) | 23:03 |
| tchaypo | Oh. I thought you were going to be working in the parent directory and the repo_refs subdir would be out of your way :( | 23:04 |
| lifeless | no | 23:04 |
| lifeless | if we're moving stuff to a subdir | 23:04 |
| * tchaypo backs right off | 23:04 | |
| lifeless | we're moving stuff to a subdir | 23:04 |
| lifeless | /home/shared/repo_refs.yaml{,.variables} deleted | 23:05 |
| lifeless | updating overcloud now | 23:07 |
| lifeless | tchaypo: you have the lock on hp2 though | 23:07 |
| tchaypo | what generates repo_refs.yaml.variables? I thought it must be in the loop at the bootom of seed.hp2.tripleo.org.rc but I can't see it writing anything | 23:08 |
| tchaypo | and seed.hp1.tripleo.org.rc doesn't seem to have that loop anyway | 23:08 |
| lifeless | tchaypo: you need to read the tripleo-end-to-end README more closely | 23:08 |
| tchaypo | the README (on hp1) says to source the file but doesn't mention it being created. | 23:09 |
| lifeless | tchaypo: thats true | 23:10 |
| tchaypo | From the placement I'm guessing that prep_source_repos must create it | 23:10 |
| lifeless | tchaypo: :) | 23:10 |
| lifeless | tchaypo: yes, perhaps that should be added to the README | 23:10 |
| lifeless | oh yay | 23:11 |
| lifeless | 2014-08-10 23:11:28.647 1946 WARNING os_collect_config.cfn [-] 400 Client Error: InvalidParameterValue | 23:11 |
| lifeless | stevebaker: WAT | 23:11 |
| stevebaker | ? | 23:12 |
| lifeless | nvm, I think I'm confused | 23:12 |
| lifeless | separately | 23:12 |
| lifeless | I just had keepalived spit the dummy at me | 23:13 |
| lifeless | and I'm thoroughly confused as to why | 23:13 |
| lifeless | + echo 'Configuration for keepalived is not yet known. Refusing to proceed.' | 23:13 |
| lifeless | stevebaker: actually no - I think this cfn thing matters | 23:13 |
| lifeless | stevebaker: the collector is failing | 23:13 |
| lifeless | stevebaker: on this live cloud I just stack-updated | 23:14 |
| stevebaker | what did the update include? | 23:14 |
| lifeless | stevebaker: well, the HOT changes | 23:14 |
| lifeless | stevebaker: and I added a Debug parameter | 23:15 |
| lifeless | so I can get debug output from the 3 control plane nodes | 23:15 |
| lifeless | et voila | 23:16 |
| lifeless | /lib/python2.7/site-packages/heat/engine/parameters.py\", line 480, in _validate_user_parameters\n raise exception.UnknownUserParameter(key=param)\n\nUnknownUserParameter: The Parameter (Debug) was not defined in template.\n", "type": "UnknownUserParameter"}, "title": "Bad Request"} to_json /opt/stack/venvs/heat/local/lib/python2.7/site-packages/heat/common/serializers.py:43 | 23:16 |
| lifeless | though I'm stumped why that would be throwing an error on cfn metadata pollin | 23:16 |
| lifeless | # heat stack-list | 23:16 |
| lifeless | ERROR: The Parameter (Debug) was not defined in template. | 23:16 |
| lifeless | stevebaker: So.... I think I have wedged this heat :) | 23:17 |
| stevebaker | problem solved! | 23:17 |
| stevebaker | (bad taste joke) | 23:17 |
| lifeless | the parameter is in the template I used to stack-update | 23:17 |
| lifeless | Debug: | 23:17 |
| lifeless | default: '' | 23:17 |
| lifeless | description: Set to truthy to enable debugging on all services. | 23:17 |
| lifeless | type: string | 23:17 |
| lifeless | I'm going to guess that there is a bug in stack-updates that add a parameter | 23:18 |
| stevebaker | lifeless: so the update is complete but metadata polling isn't working? | 23:18 |
| stevebaker | lifeless: or is the stack in a failed state | 23:19 |
| lifeless | stevebaker: I can't tell | 23:19 |
| lifeless | stevebaker: stack-list is broken | 23:19 |
| lifeless | stevebaker: see above | 23:19 |
| stevebaker | ugh | 23:20 |
| stevebaker | I wish someone would properly deal to our validate flows | 23:21 |
| lifeless | stevebaker: https://bugs.launchpad.net/heat/+bug/1354962 | 23:22 |
| uvirtbot | Launchpad bug 1354962 in heat "stack-update adding a parameter leads to heat stack-list breaking" [Undecided,New] | 23:22 |
| lifeless | stevebaker: I'm not sure that its a validate problem | 23:22 |
| lifeless | stevebaker: I'm going to guess that because the update hasn't completed | 23:22 |
| lifeless | stevebaker: but the parameters have been updated | 23:22 |
| lifeless | stevebaker: its trying to validate the old template with the new environment | 23:22 |
| stevebaker | lifeless: stack-list should always work though | 23:22 |
| lifeless | stevebaker: oh, I totally agree. | 23:23 |
| lifeless | ok, so its now at update-failed | 23:24 |
| lifeless | I've disabled that check :) | 23:24 |
| *** ci-overcloud has joined #tripleo | 23:28 | |
| ci-overcloud | ************** ci-overcloud complete status=1 ************ | 23:28 |
| *** ci-overcloud has quit IRC | 23:28 | |
| *** ci-overcloud has joined #tripleo | 23:29 | |
| ci-overcloud | ************** ci-overcloud complete status=1 ************ | 23:29 |
| *** ci-overcloud has quit IRC | 23:29 | |
| stevebaker | lifeless: once we have functional tests in-tree and https://review.openstack.org/#/c/112938/ lands then we can fairly quickly write tests which verify all the combinations of template changes when stack updates | 23:30 |
| lifeless | now I'm getting StackValidationFailed: Property error : NovaCompute0: | 23:30 |
| lifeless | key_name Value must be a string | 23:31 |
| stevebaker | I would be very suprised if the transition to HOT would work with a stack-update. It didn't even remotely occur to me that it was a requirement | 23:33 |
| lifeless | should we rollback the whole thing? | 23:34 |
| stevebaker | lifeless: is a rebuild an option? | 23:40 |
| lifeless | stevebaker: tear down folks production clouds on upgrade? | 23:41 |
| lifeless | stevebaker: [checking that that is what you mean] | 23:41 |
| stevebaker | ok, that would be a no | 23:41 |
| openstackgerrit | A change was merged to openstack/tripleo-specs: Also bind public services to a dedicated VIP https://review.openstack.org/100151 | 23:42 |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-specs: Tweak HAProxy spec to deal with mixed services. https://review.openstack.org/107956 | 23:42 |
| *** xuhaiwei has joined #tripleo | 23:46 | |
| stevebaker | lifeless: I don't see this working until the undercloud heat is upgradable | 23:47 |
| lifeless | stevebaker: it is upgradeable, using the seed heat. | 23:48 |
| stevebaker | lifeless: ...and then testing for and fixing this particular update bug | 23:49 |
| openstackgerrit | A change was merged to openstack/tripleo-specs: Update the Tuskar storage spec to reference Heat https://review.openstack.org/107684 | 23:50 |
| lifeless | stevebaker: right | 23:50 |
| lifeless | stevebaker: so I think we need to rollback to pre the hot patches | 23:50 |
| stevebaker | ;( yes | 23:50 |
| lifeless | re-propose all the pending work that was open when they landed | 23:50 |
| lifeless | ok, so I'm going to take a breather, then come back to help | 23:51 |
| *** noslzzp has joined #tripleo | 23:53 | |
| stevebaker | lifeless: lets poke at it a little more. Can you paste your overcloud template? | 23:54 |
| lifeless | stevebaker: http://paste.ubuntu.com/8012166/ | 23:54 |
| tchaypo | lifeless: looking at the testenv.jsons - the details of the public_vlan doesn't seem to have been added to the spreadsheet | 23:56 |
| lifeless | tchaypo: its spread on two tabs | 23:56 |
| lifeless | tchaypo: internal and public | 23:56 |
| tchaypo | oh | 23:57 |
| tchaypo | found it. hp2 is listed as tag 25 in the sheet but 1598 in the json | 23:57 |
| tchaypo | *updates sheet on assumption that json is correct* | 23:58 |
| stevebaker | lifeless: back in a bit | 23:58 |
| tchaypo | the json on hp1 has an "undercloud" key that's missing from hp2 | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!