| tchaypo | *sigh* | 00:01 |
|---|---|---|
| tchaypo | Usually that's one of the first things I check - that the commit message satisfies the first point in https://wiki.openstack.org/wiki/GitCommitMessages#Information_in_commit_messages - "Do not assume the reviewer understands what the original problem was" | 00:02 |
| tchaypo | ah - lxsli asked, and gfidente responded saying something similar to what I've been thinking | 00:04 |
| *** eguz has joined #tripleo | 00:04 | |
| tchaypo | ie, that allowing the user more control is nicer | 00:04 |
| tchaypo | so that's one of the two people who suggested it who apparently don'st really have a use case for it | 00:05 |
| tchaypo | and as for xuhaiwei - the change he originally proposed only hard-coded the location of the files under $TRIPLEO_ROOT, the actual suggestion to make it configureable came from.. | 00:06 |
| tchaypo | *digs* | 00:06 |
| bnemec | lifeless: tchaypo: So, my understanding of the history behind that was $SOMEBODY wanted to maintain multiple separate password files for different stacks. I don't remember who. | 00:07 |
| tchaypo | gfidente in a comment on patchset 5 | 00:07 |
| bnemec | There was some sort of change related to that code, and I -1'd because of $SOMEBODY. | 00:07 |
| tchaypo | bnemec: yeah, and that seems reasonable to me, and that's why I was thinking the testenv.json would be the place to store it | 00:07 |
| bnemec | I believe that led to this whole discussion over what to do with it. | 00:07 |
| *** eghobo has quit IRC | 00:08 | |
| bnemec | tchaypo: Yeah, honestly I think that was discussed as a future direction back when I added the "add password" functionality to the generate scripts. | 00:08 |
| * bnemec looks for that review | 00:08 | |
| tchaypo | but since xuhaiwei's original change here was just to hard-code it at $TRIPLEO_ROOT (simplifying the current logic where it checks $PWD first), maybe we could revery back to that in *this* change, and then make the path configurable in a subsequent change? | 00:08 |
| bnemec | Related: My last real comment on https://review.openstack.org/#/c/87039/ | 00:10 |
| tchaypo | I agree that the current logic allows for multiple stacks, and we'd be breaking that -but no-one has commented on https://review.openstack.org/#/c/97657 or on the list asying that they actually use that, so maybe we're trying to support an imaginary use case | 00:11 |
| bnemec | tchaypo: Personally I'd like to just leave it the heck alone, maybe add a comment on why we allow different paths to avoid developer confusion, and then deprecate _all_ of the old password files in favor of ones in JSON. | 00:11 |
| tchaypo | bnemec: your last comment aside from "recheck bug blah" ;) | 00:11 |
| bnemec | tchaypo: Exactly :-) | 00:11 |
| lifeless | so big picture | 00:11 |
| lifeless | all this needs to live in tuskar/barbican/something eventually | 00:12 |
| tchaypo | right. so my idea of sticking the passwords in the testenv.json isn't new... | 00:12 |
| lifeless | if we'regoing to spend time changing it right now | 00:12 |
| lifeless | lets do so with that in mind | 00:12 |
| *** eghobo has joined #tripleo | 00:12 | |
| tchaypo | I don't know enough about tuskar/barbican/something (especially the latter, it's very nebulous) to know what we'd have to do to make it easy to switch to using those | 00:14 |
| tchaypo | except to suggest a generic abstraction layer | 00:14 |
| tchaypo | eg, a script called retrieve-password-for that knows how to look for the files now, testenv.json shortly, or barbican in the future | 00:15 |
| bnemec | tchaypo: +1. I would think that once these are stored somewhere else, we'll need an explicit call to retrieve the values anyway. | 00:16 |
| bnemec | As opposed to the method today where we rely on env vars being set. | 00:16 |
| tchaypo | which seems like a fair chunk of work just to end up supporting something we already support today - if you have seperate $CWD or $TRIPLEO_ROOT for each stack, you can already have seperate sets of passwords for each stack | 00:16 |
| *** eguz has quit IRC | 00:16 | |
| lifeless | all of the ones for the heat environment | 00:16 |
| lifeless | can live in the heat environment right now | 00:16 |
| tchaypo | we don't rely on any env vars today - we look in $CWD and then in $TRIPLEO_ROOT | 00:16 |
| lifeless | its only asserted user details we need to preserve | 00:16 |
| tchaypo | https://review.openstack.org/#/c/97657/ would switch to relying on env vars, but has been downvoted because of that | 00:17 |
| bnemec | tchaypo: I just mean that when we retrieve the passwords, I think we just source the *-passwords file and use the env vars that result. | 00:17 |
| tchaypo | bnemec: oh yes. *those* env vars | 00:17 |
| *** yamahata has joined #tripleo | 00:18 | |
| lifeless | depends onthe file | 00:19 |
| *** eghobo has quit IRC | 00:19 | |
| lifeless | one sort we parse | 00:19 |
| lifeless | the other is environment variable yes | 00:19 |
| *** eghobo has joined #tripleo | 00:19 | |
| tchaypo | so I'm going to propose that short-term we simply add comments to the scripts to explain how someone with multiple stacks can have multiple sets of passwords by running the scripts from different paths; for the long-term, I'll open a bug saying that we should have a retrieve-password-for script so we can easily migrate from the current files to storing them in $TE_DATAFILE/Heat/Barbican/elsewhere | 00:20 |
| bnemec | lifeless: So asserted user details would be stuff like admin password and demo password? | 00:20 |
| lifeless | bnemec: see assert-users | 00:21 |
| bnemec | Okay, I think the asserted users are a separate thing though, right? Those aren't stored in the tripleo-*-passwords files AFAICT. | 00:28 |
| lifeless | bnemec: ah | 00:31 |
| lifeless | yah | 00:31 |
| greghaynes | If its just for multiple clouds, why not just store them in tripleo-$STACKNAME-passwords for now | 00:33 |
| bnemec | Part of the problem is I don't actually know what the original use case for this was. | 00:36 |
| bnemec | So I have no idea which suggestions would satisfy it. | 00:36 |
| greghaynes | I can see the multiple cloud being an issue - if you try to make multiple overcloud's youll overwrite that file | 00:36 |
| bnemec | I'm perfectly happy to have them always read out of TRIPLEO_ROOT. Maintaining multiple sets of configuration sounds like asking for trouble. :-) | 00:37 |
| *** rwsu has quit IRC | 00:37 | |
| tchaypo | so I'm going to go ahead with my suggestion above, with an extra detail of pointing out that we don't understand what the potential use case for either forcing $TRIPLEO_ROOT *or* for configurability is, and asking people to explain that if they want to reopen this | 00:46 |
| *** rwsu has joined #tripleo | 00:50 | |
| *** spzala has quit IRC | 00:50 | |
| *** andreaf_ has joined #tripleo | 00:54 | |
| *** andreaf has quit IRC | 00:56 | |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Fix diff return code in hosts element https://review.openstack.org/115134 | 00:59 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Add security group rules to tripleo-ci/default https://review.openstack.org/109161 | 01:01 |
| *** ccrouch has joined #tripleo | 01:05 | |
| *** jang has quit IRC | 01:06 | |
| *** jang1 has joined #tripleo | 01:07 | |
| *** jang has joined #tripleo | 01:08 | |
| tchaypo | whee. my overcloud is failing to build because the scripts think I need 16 nodes but hypervisor-stats is only reporting 15 | 01:10 |
| *** eghobo has quit IRC | 01:14 | |
| *** eguz has joined #tripleo | 01:14 | |
| *** uberj has quit IRC | 01:15 | |
| greghaynes | tchaypo: 16!? | 01:17 |
| tchaypo | yerp. | 01:19 |
| tchaypo | i should clarify that this is in hp2 | 01:20 |
| tchaypo | I'm assuming atht his point that that's the number of nodes we pass in to testenv.json or something | 01:20 |
| lifeless | tchaypo: its the result of your scaling factors | 01:20 |
| tchaypo | yep, just found the line | 01:21 |
| tchaypo | computescale+controlscale | 01:21 |
| tchaypo | OVERCLOUD_COMPUTESCALE is set to 15 in my env | 01:24 |
| tchaypo | I'm not sure how it got there | 01:24 |
| *** eguz has quit IRC | 01:27 | |
| *** eghobo has joined #tripleo | 01:27 | |
| *** eguz has joined #tripleo | 01:28 | |
| *** eghobo has quit IRC | 01:28 | |
| lifeless | perhaps in the rc file | 01:28 |
| lifeless | or | 01:28 |
| lifeless | perhaps from deploy-ci-overcloud | 01:28 |
| lifeless | (which you are using, yes) | 01:28 |
| *** eguz has quit IRC | 01:31 | |
| *** eghobo has joined #tripleo | 01:31 | |
| tchaypo | i was not aware of the latter, no. *hunts* | 01:34 |
| tchaypo | ... it's in the cd element, of course. | 01:34 |
| lifeless | it needs to be taught to use a config | 01:35 |
| lifeless | as its overly hp1 hard coded today | 01:35 |
| lifeless | bnemec: any chance you can review the 103449 patch? | 01:36 |
| tchaypo | driving instructor just arrived, so I'm going for a lesson and then lunch. Will look over deploy-ci-overcloud this afternoon and see what needs to be done. | 01:37 |
| StevenK | lifeless: Can you please look at https://review.openstack.org/#/c/111606/ again? CI is happy with it, hopefully it can land. | 01:37 |
| tchaypo | I'm confused about why it's not in the bin dir of the element with the other scripts, maybe i can figure that out by looking at it in detail. Also I think it might be helpful to update the readme for tripleo-end-to-end to mention something about "configs and tools specific to the TripleO CD/CI environment are located over here"? | 01:39 |
| StevenK | tchaypo: sicheres Fahren | 01:39 |
| lifeless | tchaypo: hysterical raisins | 01:40 |
| lifeless | tchaypo: tripleo-end-to-end is independent of this, i think the pointer would be confusing to the 99% rest of the world | 01:40 |
| *** nosnos has joined #tripleo | 01:47 | |
| *** jang has quit IRC | 01:48 | |
| *** jang1 has quit IRC | 01:48 | |
| *** Yanyanhu has joined #tripleo | 01:57 | |
| *** jang has joined #tripleo | 01:57 | |
| *** ChuckC has joined #tripleo | 01:58 | |
| *** eguz has joined #tripleo | 02:05 | |
| *** eguz has quit IRC | 02:05 | |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-image-elements: Reduce resource requirements set aside for testenvs https://review.openstack.org/106352 | 02:07 |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-image-elements: Fix prepare-ci-overcloud floating IP handling. https://review.openstack.org/109168 | 02:07 |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-image-elements: Refresh the glance images for squid and te-broker https://review.openstack.org/109169 | 02:07 |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-image-elements: Create the h1.ci flavor on ci-overclouds https://review.openstack.org/109630 | 02:07 |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-image-elements: Move tripleo-bm-test network to 172.16.0.0/22 https://review.openstack.org/109261 | 02:07 |
| *** eghobo has quit IRC | 02:08 | |
| *** morganfainberg is now known as morganfainberg_Z | 02:14 | |
| StevenK | Blah, fixtures needs more changes | 02:15 |
| StevenK | But in other news: | 02:16 |
| StevenK | py33 runtests: commands[0] | python setup.py testr --slowest --testr-args= | 02:16 |
| StevenK | ... | 02:16 |
| StevenK | py33: commands succeeded | 02:16 |
| lifeless | StevenK: ok, I'll wait for the extra changes | 02:39 |
| *** morazi has quit IRC | 02:44 | |
| *** ccrouch 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/1356637 | 03:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1355898 | 03:00 |
| uvirtbot | Launchpad bug 1356637 in tripleo "os-refresh-config fails in hosts element when update is different" [Undecided,New] | 03:00 |
| uvirtbot | Launchpad bug 1355898 in tripleo "mariadb doesn't start after an upgrade" [Undecided,New] | 03:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1357346 | 03:00 |
| *** untriaged-bot has quit IRC | 03:00 | |
| uvirtbot | Launchpad bug 1357346 in tripleo "undercloud failing to come up on devtest due to pacemaker in os-refresh-config" [Undecided,New] | 03:00 |
| *** sdake_ has joined #tripleo | 03:13 | |
| *** wuhg has joined #tripleo | 03:16 | |
| *** ramishra has joined #tripleo | 03:16 | |
| *** vinsh has quit IRC | 03:20 | |
| SpamapS | greghaynes: thanks for keeping the update train rolling. Here keeping the Heat train rolling. ;) | 03:27 |
| *** pradeep has joined #tripleo | 03:31 | |
| *** ramishra has quit IRC | 03:31 | |
| *** ramishra has joined #tripleo | 03:32 | |
| *** Azher has joined #tripleo | 03:42 | |
| *** ramishra has quit IRC | 03:55 | |
| *** sdake_ has quit IRC | 03:57 | |
| *** ramishra has joined #tripleo | 04:03 | |
| *** eghobo has joined #tripleo | 04:40 | |
| StevenK | Huh. Now you can buy books with Qantas Frequent Flyer points. This could be terrible. | 04:59 |
| *** pensu has joined #tripleo | 05:07 | |
| *** pcrews has quit IRC | 05:08 | |
| *** eghobo has quit IRC | 05:13 | |
| *** eghobo has joined #tripleo | 05:14 | |
| *** pensu has quit IRC | 05:14 | |
| *** pradipta_away is now known as pradipta | 05:16 | |
| *** chuckC_ has joined #tripleo | 05:16 | |
| *** loki184 has joined #tripleo | 05:21 | |
| *** killer_prince is now known as lazy_prince | 05:22 | |
| *** dshulyak_ has joined #tripleo | 05:25 | |
| * StevenK peers at EEM | 05:27 | |
| StevenK | Hah. It's a lunch from Raleigh, it only took almost a month to hit my amex | 05:29 |
| *** chuckC_ has quit IRC | 05:32 | |
| *** bmahalakshmi has joined #tripleo | 05:34 | |
| *** sirushti has joined #tripleo | 05:35 | |
| *** k4n0 has joined #tripleo | 05:36 | |
| *** CaptTofu_ has quit IRC | 05:36 | |
| *** bmahalakshmi has quit IRC | 05:39 | |
| *** CaptTofu_ has joined #tripleo | 05:39 | |
| *** bmahalakshmi has joined #tripleo | 05:40 | |
| *** morganfainberg_Z is now known as morganfainberg | 05:41 | |
| tchaypo | oh wow | 05:41 |
| *** sirushti has left #tripleo | 05:41 | |
| *** rwsu has quit IRC | 05:51 | |
| greghaynes | SpamapS: By Heat train I assume you mean that Raleigh is blisteringly hot? | 05:56 |
| tchaypo | SpamapS is still in raleigh? | 05:58 |
| *** rwsu has joined #tripleo | 06:07 | |
| greghaynes | heat sprint is going on there right now I think | 06:07 |
| *** ramishra has quit IRC | 06:09 | |
| *** pensu has joined #tripleo | 06:10 | |
| tchaypo | oh lol | 06:13 |
| *** ramishra has joined #tripleo | 06:21 | |
| *** sdake has quit IRC | 06:26 | |
| *** ramishra has quit IRC | 06:35 | |
| *** rdopieralski has joined #tripleo | 06:35 | |
| *** rdopieralski has quit IRC | 06:35 | |
| *** rdopieralski has joined #tripleo | 06:36 | |
| *** dshulyak_ has quit IRC | 06:38 | |
| openstackgerrit | Giulio Fidente proposed a change to openstack/tripleo-incubator: Remove cinder-volume from controller image. https://review.openstack.org/115191 | 06:43 |
| tchaypo | gfidente: why would you do that? | 06:43 |
| tchaypo | gfidente: hrm. okay, it's 8:43am, that's acceptable, as long as you've had coffee first. | 06:43 |
| *** jprovazn has joined #tripleo | 06:44 | |
| gfidente | tchaypo, yeah I see from past conversations that lifeless asked why about a couple of other things too :P | 06:44 |
| gfidente | so I put a why in the commit message this time :) | 06:44 |
| gfidente | basically we ended up with cinder-volume service, running on controller nodes, but unconfigured if one didn't also install cinder-{tgt,lio} on the controller node itself | 06:45 |
| gfidente | I hope the use case is clear, if it isn't please comment in the review | 06:46 |
| gfidente | I'm thinking about a scenario where one has an arbitrary number of block storage nodes deployed and doesn't want cinder-volume to run on the controllers | 06:48 |
| lifeless | why is important, its even in the wiki page on commits :) | 06:49 |
| gfidente | lifeless, it is, I wasn't questioning that | 06:51 |
| gfidente | sorry if previous message gave you the impression, I was trying to be ironic about myself, not the whys | 06:51 |
| lifeless | gfidente: it didn't, all good | 06:52 |
| jprovazn | lifeless: Hi, yesterday I left before you got back from swimming, I'd like to finish the conversation about backwards incompatibility and adding new required metadata to heat template - AIUI from yesterday, adding new metadata should be fine in all cases exccept the one: no-HOT->HOT update for self hosted undercloud (which is not used anywhere). Correct? | 06:57 |
| lifeless | jprovazn: cool | 06:58 |
| lifeless | jprovazn: uhm | 06:58 |
| lifeless | jprovazn: so generally adding mandatory new metadata will break deployments won't it? | 06:58 |
| jprovazn | lifeless: is these metadata have default value in heat template? | 06:59 |
| jprovazn | s/is/if/ | 06:59 |
| jprovazn | well, yes | 06:59 |
| lifeless | jprovazn: we have to be able to land the patches separately | 06:59 |
| jprovazn | lifeless: if THT part is merged first, it's OK | 07:00 |
| jprovazn | lifeless: what is my problem is combination of these two statements in backwards compat proposal: | 07:02 |
| jprovazn | "We will strive to avoid breaking backwards compatibility at all, and17 | 07:02 |
| jprovazn | if we have to it will be because of extenuating circumstances such as security18 | 07:02 |
| jprovazn | fixes with no other way to fix things." | 07:02 |
| jprovazn | and | 07:02 |
| jprovazn | " Increasing the required metadata to use an element." | 07:02 |
| lifeless | jprovazn: so the assumption in my head is that the only reason to increase required parameters is to increase functionality | 07:03 |
| lifeless | if the new functionality is made optional | 07:03 |
| lifeless | we should be able to get by without the new parameters | 07:03 |
| lifeless | for instance | 07:03 |
| lifeless | DVR is now here | 07:03 |
| lifeless | so pacemaker for l3 isn't needed | 07:03 |
| lifeless | unless you want to run non-dvr | 07:04 |
| jprovazn | lifeless: well, yes - for "leacy" mode pacemaker is still needed | 07:04 |
| jprovazn | lifeless: but pacemaker is still needed for ceilometer | 07:04 |
| jprovazn | lifeless: so the problem remains | 07:05 |
| jprovazn | let me show you example I linked from yesterday | 07:05 |
| lifeless | did ceilometer work before ? | 07:05 |
| jprovazn | lifeless: before pacemaker? we ran central agent on all instances | 07:06 |
| *** dtantsur|afk is now known as dtantsur | 07:06 | |
| jprovazn | https://github.com/openstack/tripleo-image-elements/blob/master/elements/corosync/os-apply-config/etc/corosync/corosync.conf#L15 | 07:07 |
| jprovazn | lifeless: finding a default value in replacement for corosync.nodes is difficult, I could use e.g. haproxy.nodes or something like this, but this is dirty hack | 07:08 |
| lifeless | jprovazn: so I don't think we need a default for corosync.nodes | 07:09 |
| lifeless | we need to make depending on corosync optional | 07:10 |
| jprovazn | lifeless: corosync is requirement for pacemaker, being optional for pacemaker means being optional for HA | 07:10 |
| lifeless | so HA is optional, folk can deploy one node | 07:11 |
| lifeless | I'm not (in this discussion :P) arguing that we support arbitrary service monitor replacements | 07:11 |
| jprovazn | lifeless: you still use same setup (pacemaker) with single node too | 07:11 |
| jprovazn | lifeless: having HA optional is exact opposite of what you stated on last mid-cycle - have service in HA mode only | 07:12 |
| lifeless | so could one make [localhost] be the default? | 07:12 |
| lifeless | I'm still keen on that! | 07:13 |
| jprovazn | I can't use localhost - list of controller nodes in overcloud is needed to setup corosync cluster | 07:14 |
| jprovazn | lifeless: yes, I can use check "if corosync metadata are present, then start services with pacemaker else start them as a normal service" | 07:15 |
| jprovazn | but this is what we should not do and for waht I dont' see a reason | 07:15 |
| jprovazn | as there is a way to update existing stack with stack-update and feed new metadata | 07:16 |
| lifeless | I think we're having two discussions at once | 07:17 |
| lifeless | one about whether new mandatory metadata on existing image builds is indeed backwards incompatible | 07:18 |
| lifeless | and another about whether we need this change to be backwards compatible | 07:18 |
| jprovazn | yes | 07:18 |
| lifeless | so I don't see how new mandatory metadata can be backwards compatible | 07:19 |
| lifeless | it will require changes to t-h-t or tuskar, always | 07:20 |
| lifeless | the reason for doing backwards compatibility is to make things reliable for folk pulling new images, without them having to pull all of trunk all the time | 07:20 |
| lifeless | not the only, but one of the reasons | 07:21 |
| gfidente | lifeless, I have to admit, I'm not sure I understand this requirement though | 07:21 |
| jprovazn | lifeless: "without them having to pull all of trunk" == making sure their template is compatible with newely pulled elements? | 07:22 |
| *** rushiagr_away is now known as rushiagr | 07:22 | |
| gfidente | lifeless, why would I download a K image and expect it to work with my J undercloud? | 07:22 |
| gfidente | lifeless, this is the point I raised in a comment in the spec | 07:22 |
| lifeless | gfidente: because we're pursing a single-branch strategy rather than a fork-and-forget strategy? | 07:23 |
| lifeless | gfidente: So not a 'K image', a 'J image built from elements that also know how to deploy K' | 07:24 |
| lifeless | jprovazn: yes | 07:24 |
| lifeless | jprovazn: as long as they haven't added or removed elements, of course | 07:24 |
| openstackgerrit | James Polley proposed a change to openstack/tripleo-incubator: Explain why we look for passwords in $CWD first https://review.openstack.org/115198 | 07:24 |
| gfidente | lifeless, yeah I understand the stable branch strategy but the questions remains, a J image built from elements that are meant to deploy K ? | 07:25 |
| lifeless | gfidente: I or J or K | 07:26 |
| lifeless | gfidente: and when I goes EOL, then J or K or L or M, or whatever the window is | 07:26 |
| *** jcoufal has joined #tripleo | 07:26 | |
| jprovazn | lifeless: as I commented in previous version of the spec - this pushes us to: 1) do things in a wrong way - define defaults on multiple places instead of having them on one place (template), 2) using defaults doesn't assure that new/changed element will work with old template - with defaults you just guessing | 07:27 |
| lifeless | jprovazn: the spec calls for a test that it will work | 07:28 |
| lifeless | jprovazn: the alternative, forking, means we need a test anyway, except its in a different branch - its no less work | 07:28 |
| lifeless | jprovazn: I don't understand 1) | 07:28 |
| *** jtomasek has joined #tripleo | 07:29 | |
| jprovazn | lifeless: the example of 1): this means that the default values will have to be added to config templates and to any orc scripts if needed, so e.g. corosync.nodes - I would have to add a default block into config template, also I would have to define defaults in ORC script | 07:31 |
| jprovazn | any change in defaults means changing it on multiple places | 07:32 |
| lifeless | jprovazn: we have default values in nearly all our oac code | 07:33 |
| lifeless | jprovazn: in elements | 07:33 |
| lifeless | jprovazn: what you're arguing will happen if we do this we're already doing and have been most of the time | 07:33 |
| lifeless | jprovazn: defaults don't belong in templates | 07:34 |
| lifeless | jprovazn: templates should be primarily calculating things, which is very much not defaults | 07:34 |
| *** ramishra has joined #tripleo | 07:35 | |
| lifeless | jprovazn: (obviously for openstack things we use the defaults that are in their repo - no config == default) | 07:37 |
| lifeless | jprovazn: what do you think we should do differently? | 07:38 |
| jprovazn | lifeless: IIRC in current oac code we rely on default setup wherever possible, what I mean is that if I expect e.g. corosync.nodes to be set, I have to set defaults on multiple places | 07:38 |
| lifeless | jprovazn: or not drag in corosync automatically | 07:38 |
| openstackgerrit | Nicholas Randon proposed a change to openstack/tripleo-heat-templates: Add specific pass-through config for each role https://review.openstack.org/108415 | 07:39 |
| lifeless | jprovazn: possibly there's a third route | 07:39 |
| jprovazn | lifeless: not drag corosync is == HA is optional | 07:39 |
| *** ramishra has quit IRC | 07:40 | |
| lifeless | so the options I see are: | 07:41 |
| lifeless | - make corosync not break things if its not configured | 07:41 |
| lifeless | - make corosync not be brought in to images with the prior set of elements | 07:41 |
| lifeless | - decide to break backwards compat here | 07:42 |
| *** pblaho has joined #tripleo | 07:42 | |
| lifeless | the question I think you raised before is | 07:42 |
| lifeless | 'if a new THT solves things, are new parameters really backwards incompatible?' | 07:42 |
| jprovazn | yes, this is probably better formulated ;) | 07:43 |
| lifeless | you said it just fine | 07:43 |
| lifeless | I'm just trying to bring it all together | 07:43 |
| jprovazn | so yes to the above summary | 07:44 |
| lifeless | I think the last point is hard to answer | 07:44 |
| lifeless | I mean | 07:44 |
| lifeless | consider nova | 07:44 |
| lifeless | depends on libvirt right | 07:44 |
| jprovazn | yes | 07:44 |
| lifeless | if it started requiring libvirt 1.3 | 07:44 |
| lifeless | when it used to work with 1.2 | 07:44 |
| lifeless | would folk consider that backwards incompatible? | 07:44 |
| lifeless | I think folk would hesitate to merge a patch to one of the stable release branches that made such a change | 07:45 |
| lifeless | I think they'd hesitate because that requirement cascades - all of a sudden RHEL would have to update libvirt to get whatever patch it was | 07:46 |
| *** ifarkas has joined #tripleo | 07:46 | |
| jprovazn | well, you probably see TIE<->THT less tighten than me | 07:46 |
| lifeless | My impression is that 'when a new version is present, using its features is fine, but requiring a new version is problematic' | 07:46 |
| lifeless | OTOH | 07:47 |
| lifeless | we require new versions of e.g. pbr all the time | 07:47 |
| lifeless | and don't do semver bumps of the depending library | 07:47 |
| lifeless | I need to cook dinner, I'll be back shortly, will think on this more while I cook | 07:48 |
| jprovazn | OK, enjoy :) | 07:48 |
| * gfidente is also unsure about elements supporting two past releases as well as current | 07:49 | |
| *** eghobo has quit IRC | 07:50 | |
| *** jcoufal has quit IRC | 07:51 | |
| lifeless | gfidente: why? | 07:59 |
| lifeless | jprovazn: so back | 07:59 |
| jprovazn | lifeless: that was quick | 07:59 |
| lifeless | jprovazn: here's my argument for why non-patch-level dependency bumps are major version bumps, not new feature | 08:00 |
| lifeless | jprovazn: two eggs :) | 08:00 |
| lifeless | jprovazn: consider a web service | 08:00 |
| lifeless | that accepts as parameters the output from another service | 08:01 |
| lifeless | if you have to start using new urls on that other service | 08:01 |
| *** akrivoka has joined #tripleo | 08:01 | |
| lifeless | then your code has to change | 08:01 |
| lifeless | so my argument fails | 08:02 |
| lifeless | because if the dependency change is minor | 08:02 |
| lifeless | the other urls shouldn't be changing | 08:02 |
| lifeless | you'd stay on e.g. '1' and be getting their highest v1 objects | 08:02 |
| lifeless | so the question is then | 08:02 |
| lifeless | is tht a dependency of tie | 08:03 |
| *** jcoufal has joined #tripleo | 08:03 | |
| lifeless | or tie a dependency of tht | 08:03 |
| lifeless | or is it a tightly coupled thing | 08:03 |
| jprovazn | lifeless: is any of these components usable w/o each other? | 08:03 |
| loki184 | lifeless: hi, infra core jeremy has left a comment for you on this patch https://review.openstack.org/#/c/110562/ Please look into it. | 08:05 |
| lifeless | loki184: it needs rebasing | 08:05 |
| gfidente | lifeless, on why I'm sure about elements supporting three releases, it can affect basic install steps too, not only config steps, see this for example: https://review.openstack.org/#/c/94303/ | 08:06 |
| gfidente | *unsure | 08:06 |
| lifeless | loki184: oh, no that was an old revision | 08:06 |
| loki184 | lifeless: yup have uploaded a new patch set, patchset 11 he left a comment asking for your +1 | 08:07 |
| lifeless | gfidente: right, so we had to update our code to deal with an update upstream, and because we want to be compatible it wasn't just deleting the code | 08:07 |
| lifeless | loki184: I've +1d | 08:07 |
| lifeless | gfidente: so there is a cost there | 08:08 |
| loki184 | lifeless: Thank you :) | 08:08 |
| lifeless | jprovazn: To answer your question, I don't see why not - after all tuskar is roughly speaking a parallel implementation of THT | 08:11 |
| gfidente | lifeless, sort of tangent question, what is the update process supposed to be? assuming I am on J and want to update to K | 08:13 |
| gfidente | I update undercloud image, rebuild overcloud images, redeploy? | 08:14 |
| lifeless | build new images | 08:14 |
| lifeless | stack-update | 08:14 |
| lifeless | in principle | 08:14 |
| jprovazn | gfidente: redeploy is the problem - you want to avoid outages | 08:15 |
| gfidente | ok and that should happen for undercloud first? | 08:15 |
| gfidente | jprovazn, yep sorry, so redeploy isn't a loop of bringing down/up nodes but rather a stack-update | 08:16 |
| jprovazn | gfidente: aha, ok | 08:16 |
| lifeless | reality is though that openstack K may have new things that have to be calculated from the cluster | 08:16 |
| lifeless | so its likely that you'll need a new template no matter how much we try to avoid it | 08:17 |
| lifeless | in that situation we'll possibly need multiple stages | 08:17 |
| lifeless | I hope we can avoid that indefinitely :) | 08:17 |
| lifeless | jprovazn: so if the rule is that we have to update THT, tuskar, and any other templates first | 08:18 |
| lifeless | jprovazn: then that means that tie depends on tht | 08:18 |
| lifeless | jprovazn: does tht depend on tie? I think it does. | 08:18 |
| gfidente | jprovazn, on the stack-update thing, is it going to rebase the servers on the new images? | 08:18 |
| lifeless | jprovazn: because the signals that we send back | 08:18 |
| jprovazn | lifeless: it does because of callbacks/notifications | 08:19 |
| jprovazn | or signals or how it's called | 08:19 |
| jprovazn | yes | 08:19 |
| lifeless | so if we consider it as a unit | 08:19 |
| lifeless | never having to pass new parameters to the *template* *to deploy something you deployed before* | 08:20 |
| lifeless | seems like one of the things backwards compat should deliver | 08:20 |
| lifeless | deploying K instead of J - new thing, new parameters, thats fine | 08:20 |
| lifeless | and the argument for considering it a unit is because of the bidirectional dependencies. | 08:21 |
| lifeless | jprovazn: ^ does that make sense to you? | 08:21 |
| lifeless | jcoufal: are we meeting tonight? | 08:22 |
| gfidente | lifeless, FWIW, I am on board with that and would make that explicit in the spec | 08:22 |
| lifeless | jcoufal: (I recall you wanted to change the time) | 08:22 |
| *** rushiagr is now known as rushiagr_away | 08:22 | |
| *** jp_at_hp has joined #tripleo | 08:23 | |
| gfidente | lifeless, btw, when you say stack-update, to rebase servers to new image heat will have to stop/rebuild right? possibly not all of them at the same time | 08:23 |
| openstackgerrit | Gonéri Le Bouder proposed a change to openstack/tripleo-image-elements: PEP8 on the doc and elements files https://review.openstack.org/106319 | 08:23 |
| *** derekh has joined #tripleo | 08:23 | |
| lifeless | gfidente: there are several update strategies | 08:23 |
| lifeless | gfidente: offline REBUILD_WITH_EPHEMERAL is what we have working today | 08:24 |
| jprovazn | lifeless: yes, makes sense | 08:24 |
| lifeless | gfidente: online there is takeovernode | 08:24 |
| gfidente | I have yet to explore this, sounds like will stress HA ;P | 08:25 |
| lifeless | induitably | 08:25 |
| derekh | lifeless: pass rate upto 55% on hp1, looking at error now | 08:25 |
| *** regebro has joined #tripleo | 08:25 | |
| lifeless | derekh: I had a bit of a poke | 08:26 |
| lifeless | derekh: 65 timeouts | 08:26 |
| lifeless | derekh: and quite a few where the oc node is being deleted, which left me blinking | 08:26 |
| lifeless | let me poke at the spec shortly | 08:27 |
| *** lucasagomes has joined #tripleo | 08:28 | |
| derekh | lifeless: ok, I'm going to kick off 100 with no nova delete at the end so I can log into the failed instances afterwards and dig into the problems | 08:30 |
| lifeless | derekh: roger | 08:30 |
| jprovazn | lifeless: "never having to pass new parameters to the *template* *to deploy something you deployed before*" makes sense, though if I consider THT+TIE as a unit, then "Increasing the required metadata to use an element." should be fine (it's unit internal change), though this would require having to update THT,tuskar and others first | 08:33 |
| *** loki184 has quit IRC | 08:35 | |
| *** akrivoka has quit IRC | 08:35 | |
| *** loki184 has joined #tripleo | 08:35 | |
| *** athomas has joined #tripleo | 08:39 | |
| pblaho | Goneri: ping | 08:39 |
| *** jang has quit IRC | 08:43 | |
| *** jang has joined #tripleo | 08:45 | |
| *** pradeep has quit IRC | 08:48 | |
| *** rushiagr_away is now known as rushiagr | 08:48 | |
| pblaho | have anyone seen this "OSError: [Errno 13] Permission denied: '/opt/git/openstack/tripleo-incubator/openstack-tools'" in logs for failed check-tripleo-ironic-overcloud-precise-ha ? | 08:52 |
| lifeless | pblaho: thats new to me | 08:54 |
| pblaho | lifeless: http://logstash.openstack.org/#eyJzZWFyY2giOiJtZXNzYWdlOlwiT1NFcnJvcjogW0Vycm5vIDEzXSBQZXJtaXNzaW9uIGRlbmllZDogJy9vcHQvZ2l0L29wZW5zdGFjay90cmlwbGVvLWluY3ViYXRvci9vcGVuc3RhY2stdG9vbHMnXCIiLCJmaWVsZHMiOltdLCJvZmZzZXQiOjAsInRpbWVmcmFtZSI6IjYwNDgwMCIsImdyYXBobW9kZSI6ImNvdW50IiwidGltZSI6eyJ1c2VyX2ludGVydmFsIjowfSwic3RhbXAiOjE0MDg0MzgyMzk3MjF9 | 08:54 |
| pblaho | lifeless: ie here with this patch https://review.openstack.org/#/c/114948/ | 08:55 |
| *** yamahata has quit IRC | 08:57 | |
| *** pradeep has joined #tripleo | 08:59 | |
| *** untriaged-bot has joined #tripleo | 09:01 | |
| untriaged-bot | Untriaged bugs so far: | 09:01 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1356637 | 09:01 |
| uvirtbot | Launchpad bug 1356637 in tripleo "os-refresh-config fails in hosts element when update is different" [Undecided,New] | 09:01 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1355898 | 09:01 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1357346 | 09:01 |
| uvirtbot | Launchpad bug 1355898 in tripleo "mariadb doesn't start after an upgrade" [Undecided,New] | 09:01 |
| uvirtbot | Launchpad bug 1357346 in tripleo "undercloud failing to come up on devtest due to pacemaker in os-refresh-config" [Undecided,New] | 09:01 |
| *** untriaged-bot has quit IRC | 09:01 | |
| *** proffalken has quit IRC | 09:04 | |
| *** andreaf_ has quit IRC | 09:10 | |
| *** pelix has joined #tripleo | 09:10 | |
| *** proffalken has joined #tripleo | 09:11 | |
| *** ramishra has joined #tripleo | 09:14 | |
| openstackgerrit | A change was merged to openstack/os-cloud-config: Small fixes to support Python 3 https://review.openstack.org/111606 | 09:18 |
| *** ramishra has quit IRC | 09:18 | |
| *** proffalken has quit IRC | 09:24 | |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-specs: TripleO and backwards compatibility https://review.openstack.org/102081 | 09:24 |
| lifeless | jprovazn: gfidente: ^ | 09:24 |
| *** proffalken has joined #tripleo | 09:27 | |
| *** proffalken has quit IRC | 09:29 | |
| *** proffalken has joined #tripleo | 09:30 | |
| lifeless | ok, early night for me; see you all tomorrow :) | 09:34 |
| *** akrivoka has joined #tripleo | 09:38 | |
| *** andreaf has joined #tripleo | 09:42 | |
| *** Daviey has quit IRC | 09:44 | |
| *** rushiagr is now known as rushiagr_away | 09:47 | |
| *** rushiagr_away is now known as rushiagr | 09:50 | |
| *** pblaho is now known as pblaho|afk | 10:00 | |
| *** lazy_prince is now known as killer_prince | 10:08 | |
| *** Yanyanhu has quit IRC | 10:08 | |
| *** Daviey has joined #tripleo | 10:13 | |
| *** rlandy has joined #tripleo | 10:18 | |
| *** killer_prince is now known as lazy_prince | 10:19 | |
| *** pradipta is now known as pradipta_away | 10:29 | |
| *** pblaho|afk is now known as pblaho | 10:29 | |
| *** andreaf has quit IRC | 10:33 | |
| openstackgerrit | Ladislav Smola proposed a change to openstack/os-cloud-config: Make create tenant idempotent https://review.openstack.org/113852 | 10:41 |
| openstackgerrit | Ladislav Smola proposed a change to openstack/os-cloud-config: Allow setup services endpoints https://review.openstack.org/113882 | 10:41 |
| openstackgerrit | Ladislav Smola proposed a change to openstack/os-cloud-config: Make create keystone endpoint idempotent https://review.openstack.org/113883 | 10:41 |
| openstackgerrit | Ladislav Smola proposed a change to openstack/os-cloud-config: Make create roles idempotent https://review.openstack.org/113839 | 10:41 |
| openstackgerrit | Ladislav Smola proposed a change to openstack/os-cloud-config: Make setup endpoint idempotent https://review.openstack.org/112825 | 10:41 |
| *** pradipta_away is now known as pradipta | 10:44 | |
| rdopieralski | pblaho: what is the status of python-tuskarclient right now, can we use the v2 api for plan manipulation? | 10:47 |
| pblaho | rdopieralski: yes... all patches were merged... but API is not completed... https://review.openstack.org/#/c/113988/ | 10:50 |
| pblaho | rdopieralski: and by all patches I mean patches for python bindings.... not for command line interface | 10:51 |
| *** ifarkas has quit IRC | 10:55 | |
| *** ifarkas_ has joined #tripleo | 10:57 | |
| rdopieralski | pblaho: so I can already use the python-tuskarclient, but it won't work yet? | 10:58 |
| *** Azher has quit IRC | 10:59 | |
| d0ugal | rdopieralski: That's correct, AFAIK, the API responses are all mock data at the moment | 10:59 |
| rdopieralski | awesome, thanks | 10:59 |
| d0ugal | rdopieralski: but you could run jdobs review that is above, I think it's ready - just need another core. | 11:00 |
| d0ugal | lsmola: ^ :) | 11:00 |
| rdopieralski | hmmm | 11:01 |
| lsmola | rdopieralski: btw. I don't think tuskar client was released yet | 11:01 |
| lsmola | rdopieralski: you will have to install it locally | 11:01 |
| lsmola | d0ugal: looking | 11:01 |
| lsmola | d0ugal: this ^ should make tuskar work? | 11:01 |
| d0ugal | Oh yeah, I forgot about releasing. | 11:01 |
| d0ugal | lsmola: Yeah, it adds the "real" API. | 11:01 |
| lsmola | d0ugal: btw. afaik we don't have working overlcoud templates yet, right? | 11:02 |
| d0ugal | connects the API to the storage etc. | 11:02 |
| d0ugal | lsmola: not sure | 11:02 |
| lsmola | d0ugal: I mean nobody has tested it on any real deployment right, or did you? | 11:02 |
| d0ugal | lsmola: I've not, I don't know if jdob did. | 11:02 |
| lsmola | d0ugal: i saw some script for populating of templates into tuskar | 11:02 |
| *** pmyers has joined #tripleo | 11:02 | |
| d0ugal | lsmola: Yeah, that's probably my template loading review. | 11:02 |
| lsmola | d0ugal: right, so we will have tuskar, but we can't use it for deployment yet, so... | 11:04 |
| lsmola | d0ugal: does your script contains list of default templates? | 11:04 |
| lsmola | d0ugal: if it does, maybe we have the templates already | 11:04 |
| d0ugal | lsmola: No, it expects you point it to a directory of templates to load. | 11:05 |
| d0ugal | i.e. t-h-t | 11:05 |
| lsmola | d0ugal: I think there was at least compute and controller tempalte done | 11:05 |
| lsmola | d0ugal: so how do we create the default set of roles? | 11:05 |
| d0ugal | Look at the review, that's all I've got so far. | 11:05 |
| pblaho | lsmola: d0ugal's review https://review.openstack.org/#/c/113248/ | 11:07 |
| *** lucasagomes is now known as lucas-hungry | 11:09 | |
| lsmola | pblaho: d0ugal yeah I saw this one | 11:11 |
| lsmola | pblaho: d0ugal I was interested in the whole story, who is working on creating set of 4 default roles we will have in Juno? | 11:12 |
| lsmola | pblaho: d0ugal we will need set of working roles and working templates from TuskarAPI, is somebody working on that? Or is that done? | 11:12 |
| lsmola | pbelanyi: d0ugal that should be some init script afaik | 11:13 |
| d0ugal | lsmola: Sounds reasonable | 11:14 |
| d0ugal | lsmola: I think you can work on that if you want? otherwise something we should talk about later maybe | 11:15 |
| *** akrivoka has quit IRC | 11:15 | |
| lsmola | d0ugal: ah ok, lets chat about that later today | 11:15 |
| lsmola | d0ugal: I got lots of things on my ToDo list, but I will do that if no one else will :-) | 11:16 |
| d0ugal | Yeah, I don't have time this week and then I'm on PTO, so not gonna volunteer ;) | 11:16 |
| *** ifarkas_ has quit IRC | 11:18 | |
| *** andreaf has joined #tripleo | 11:19 | |
| *** ifarkas_ has joined #tripleo | 11:29 | |
| openstackgerrit | Ladislav Smola proposed a change to openstack/tripleo-incubator: Pass password for SNMPd correctly https://review.openstack.org/115249 | 11:30 |
| *** ifarkas_ has quit IRC | 11:32 | |
| *** ifarkas has joined #tripleo | 11:32 | |
| openstackgerrit | Ladislav Smola proposed a change to openstack/tripleo-incubator: Test if Undercloud Ceilometer monitors Overcloud https://review.openstack.org/101479 | 11:33 |
| *** loki184 has quit IRC | 11:37 | |
| lsmola | d0ugal: ok | 11:38 |
| *** dshulyak_ has joined #tripleo | 11:49 | |
| *** pblaho has quit IRC | 11:50 | |
| *** weshay has joined #tripleo | 11:57 | |
| *** akrivoka has joined #tripleo | 11:59 | |
| *** dprince has joined #tripleo | 12:00 | |
| *** weshay is now known as weshay_mtg | 12:00 | |
| *** bmahalakshmi has quit IRC | 12:04 | |
| SpamapS | tchaypo: I am | 12:06 |
| SpamapS | tchaypo: I am still in Raleigh I mean | 12:06 |
| *** morazi has joined #tripleo | 12:07 | |
| *** dshulyak_ has quit IRC | 12:10 | |
| *** pblaho has joined #tripleo | 12:14 | |
| *** ramishra has joined #tripleo | 12:16 | |
| *** ramishra has quit IRC | 12:21 | |
| *** pblaho has quit IRC | 12:24 | |
| *** ccrouch has joined #tripleo | 12:25 | |
| *** pblaho has joined #tripleo | 12:27 | |
| *** lucas-hungry is now known as lucasagomes | 12:30 | |
| *** pensu has quit IRC | 12:39 | |
| openstackgerrit | Dougal Matthews proposed a change to openstack/tuskar: Tuskar command to import a set of templates https://review.openstack.org/113248 | 12:42 |
| *** julim has joined #tripleo | 12:42 | |
| *** julim has quit IRC | 12:43 | |
| *** pblaho has quit IRC | 12:48 | |
| *** markmc has joined #tripleo | 12:49 | |
| *** pblaho has joined #tripleo | 12:56 | |
| *** nosnos has quit IRC | 12:56 | |
| jprovazn | slagle: Hi, I replied to your comment on https://review.openstack.org/#/c/114532/ - can you please recheck? | 13:00 |
| *** pblaho is now known as pblaho|meeting | 13:00 | |
| *** jpeeler has joined #tripleo | 13:01 | |
| *** jpeeler has joined #tripleo | 13:01 | |
| *** julim has joined #tripleo | 13:03 | |
| slagle | jprovazn: ok, yea i missed that we already had 8.8.8.8 in setup-neutron | 13:05 |
| slagle | jprovazn: so was that actually getting used? | 13:05 |
| jprovazn | slagle: what do you mean by getting used? If you set the option, then yes - this nameserver info is provided by dhcp, it's up to your instance if it respects it or not, IIRC I checked in cirros logs that 8.8.8.8 was set, but this would require double check | 13:08 |
| slagle | jprovazn: yea, that's what i meant. was just curious if it was working or not | 13:10 |
| slagle | trying to understand the point of the patch | 13:10 |
| slagle | i get that it's made to be configurable, that's good | 13:10 |
| jprovazn | slagle: so if you don't pass dns-nameserver when creating a neutron subnet, nameserver points to the neutron server (192.0.2.3 for OC nodes), if you run your own DNS for yours setup, you might want to use your own nameserver | 13:15 |
| openstackgerrit | A change was merged to openstack/tripleo-heat-templates: Set the MySQL root password https://review.openstack.org/102562 | 13:17 |
| *** pradipta is now known as pradipta_away | 13:24 | |
| slagle | jprovazn: ok, got it, i replied on the patch | 13:24 |
| jprovazn | slagle: Thanks. In that case I wonder if we shouldn't extend the TE_DATAFILE loading little bit - this file is overwritten in https://github.com/openstack/tripleo-incubator/blob/master/scripts/devtest_testenv.sh#L206 - so it's not optimal for custom user setting. Maybe we might merge it with an optional user-settings file if such file exists? | 13:31 |
| *** pblaho|meeting is now known as pblaho | 13:31 | |
| jprovazn | slagle: otherwise if I want add support for optional user settings, I have to always update devtest_testenv and put them here | 13:32 |
| *** pcrews has joined #tripleo | 13:33 | |
| slagle | jprovazn: devtest_testenv is only run once though | 13:33 |
| *** pcrews_ has joined #tripleo | 13:34 | |
| *** pblaho has quit IRC | 13:34 | |
| *** pcrews__ has joined #tripleo | 13:35 | |
| *** pblaho has joined #tripleo | 13:37 | |
| slagle | we write to $TE_DATAFILE in several places, boot-seed-vm for instance | 13:37 |
| slagle | the initial creation of it where it is overwritten if contained to devtest_testenv | 13:37 |
| slagle | *is contained | 13:37 |
| *** pcrews has quit IRC | 13:38 | |
| jprovazn | slagle: yes, my point is that this file is not suitable for user custom settings because it's being overwritten | 13:38 |
| gfidente | slagle, I tend to agree with jprovazn on the datafile not being easy to edit | 13:38 |
| gfidente | but I know there is general agreement on not adding more env VARS | 13:39 |
| jprovazn | slagle: this file is suitable for data you need to create and cary through devtest setup | 13:39 |
| *** pcrews_ has quit IRC | 13:39 | |
| openstackgerrit | Dougal Matthews proposed a change to openstack/tuskar: Migrate from common code to oslo.db https://review.openstack.org/104548 | 13:40 |
| gfidente | jprovazn, I think tchaypo in an email to openstack-dev also mentioned to have the passwords file location saved in $TE_DATAFILE | 13:40 |
| openstackgerrit | tom-howley proposed a change to openstack/tripleo-image-elements: Configure haproxy logging https://review.openstack.org/113827 | 13:40 |
| slagle | jprovazn: i'm not following the distinction around "user custom settings" | 13:40 |
| gfidente | but I think it is getting unclear what is good and what is not good to configure via env VAR then | 13:41 |
| slagle | that's what TE_DATAFILE has in it already, no? | 13:41 |
| slagle | POWER_USER, etc | 13:41 |
| *** pcrews__ has quit IRC | 13:41 | |
| jprovazn | slagle: this means editing devtest_Testenv file anytime I want to add a new setting | 13:42 |
| jprovazn | slagle: gfidente: so I think if this setting would be in a json file, we should check if there is some "custom.json" and merge content of the custom setting when writing TE_DATAFILE in devtest_Testenv | 13:42 |
| slagle | jprovazn: and you don't have to do that | 13:42 |
| *** pcrews has joined #tripleo | 13:43 | |
| jprovazn | slagle: what I don't have to do? | 13:44 |
| slagle | "editing devtest_Testenv file anytime I want to add a new setting" | 13:44 |
| slagle | devtest_testenv is only run once, when $TE_DATAFILE doesn't exist | 13:44 |
| slagle | yes, it will overwrite if you happen to already have one...but that's not typical | 13:45 |
| slagle | you're specifically running devtest_testenv to create a new test environment, you wouldn't be trying to reuse an existing $TE_DATAFILE | 13:45 |
| *** jcoufal has quit IRC | 13:45 | |
| *** jcoufal has joined #tripleo | 13:46 | |
| slagle | my point is really just that the values for these nameservers need to be saved somewhere | 13:48 |
| slagle | $TE_DATAFILE seemed like a reasonable place. since we really only have that or write-tripleorc | 13:50 |
| jprovazn | slagle: I agree. I was just pointing out that from my experience putting this into TA_DATAFILE was not ideal because my custom setting was gone on every new devtest run | 13:50 |
| jprovazn | nvm, thanks for your comment | 13:52 |
| jprovazn | I will put it into some json file | 13:52 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Run dbsync commands with service user https://review.openstack.org/115018 | 13:59 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Use neutron-rootwrap from /usr/bin https://review.openstack.org/115019 | 14:00 |
| *** ramishra has joined #tripleo | 14:01 | |
| *** bmahalakshmi has joined #tripleo | 14:02 | |
| *** spzala has joined #tripleo | 14:03 | |
| openstackgerrit | A change was merged to openstack/tuskar: Adds logic layer for handling plans and roles https://review.openstack.org/112997 | 14:04 |
| *** marun has joined #tripleo | 14:04 | |
| *** marun is now known as marun_afk | 14:05 | |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Remove cinder-volume from controller image. https://review.openstack.org/115191 | 14:05 |
| *** lblanchard has joined #tripleo | 14:05 | |
| *** ramishra has quit IRC | 14:06 | |
| *** ramishra has joined #tripleo | 14:07 | |
| openstackgerrit | Petr Blaho proposed a change to openstack/tuskar: Wires up REST API to underlying Tuskar logic https://review.openstack.org/113988 | 14:10 |
| openstackgerrit | Petr Blaho proposed a change to openstack/tuskar: Fix for roles with dashes in the name https://review.openstack.org/114004 | 14:10 |
| openstackgerrit | Petr Blaho proposed a change to openstack/tuskar: Add set value functionality to plans manager https://review.openstack.org/113986 | 14:10 |
| openstackgerrit | Petr Blaho proposed a change to openstack/tuskar: Fix for scaling default https://review.openstack.org/114002 | 14:10 |
| openstackgerrit | Petr Blaho proposed a change to openstack/tuskar: Simplify code returning transfer_plans https://review.openstack.org/115297 | 14:10 |
| d0ugal | pblaho: careful when you rebase - looks like you resent jdobs reviews | 14:12 |
| *** agordeev has quit IRC | 14:12 | |
| d0ugal | pblaho: so they just lost all their +A's :( | 14:13 |
| *** lazy_prince is now known as killer_prince | 14:13 | |
| pblaho | d0ugal: exactly :-( | 14:14 |
| pblaho | d0ugal: I discovered too late | 14:14 |
| d0ugal | pbelanyi: did you do `git rebaes -i master`? | 14:14 |
| pblaho | d0ugal: yeah, I did something like that | 14:14 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: fix: test_os_svc_daemon.py https://review.openstack.org/114948 | 14:14 |
| d0ugal | pblaho: `git rebase -i reviews/jaysbranch...` is safer :) | 14:15 |
| d0ugal | I made this error recently too | 14:15 |
| pblaho | d0ugal: the thing was that jdob's branch has failing tests.... and they are ok when rebased/merged with master.... so it works for jenkins... it did not worked for me.... so I rebased on top of master.... | 14:16 |
| d0ugal | ah | 14:16 |
| d0ugal | I see | 14:16 |
| *** jkraj has joined #tripleo | 14:25 | |
| *** marun_afk has quit IRC | 14:27 | |
| *** gary7ny has joined #tripleo | 14:27 | |
| *** ramishra has quit IRC | 14:30 | |
| openstackgerrit | Lennart Regebro proposed a change to openstack/tuskar-ui: Moving back to using python-tuskarclient instead of mocking it https://review.openstack.org/115302 | 14:32 |
| openstackgerrit | Ben Nemec proposed a change to openstack/tripleo-image-elements: Revert "Add workaround for reloading haproxy on Fedora" https://review.openstack.org/115303 | 14:33 |
| regebro | My change above is a work in progress, but feedback is very welcome. | 14:34 |
| *** rdopieralski has quit IRC | 14:37 | |
| *** mestery_afk is now known as mestery | 14:39 | |
| *** ramishra has joined #tripleo | 14:39 | |
| *** jprovazn has quit IRC | 14:40 | |
| *** pblaho is now known as pblaho|afk | 14:46 | |
| *** pblaho|afk is now known as pblaho | 14:47 | |
| *** ramishra has quit IRC | 14:51 | |
| *** julim has quit IRC | 14:59 | |
| openstackgerrit | Ana Krivokapic proposed a change to openstack/tuskar-ui: Disable Identity dashboard by default https://review.openstack.org/115309 | 14:59 |
| *** untriaged-bot has joined #tripleo | 15:00 | |
| untriaged-bot | Untriaged bugs so far: | 15:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1356637 | 15:00 |
| uvirtbot | Launchpad bug 1356637 in tripleo "os-refresh-config fails in hosts element when update is different" [Undecided,New] | 15:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1355898 | 15:00 |
| uvirtbot | Launchpad bug 1355898 in tripleo "mariadb doesn't start after an upgrade" [Undecided,New] | 15:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1357346 | 15:00 |
| uvirtbot | Launchpad bug 1357346 in tripleo "undercloud failing to come up on devtest due to pacemaker in os-refresh-config" [Undecided,New] | 15:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1351010 | 15:00 |
| uvirtbot | Launchpad bug 1351010 in ironic "Update to syslinux 6 breaks TFTPboot" [High,Confirmed] | 15:00 |
| *** untriaged-bot has quit IRC | 15:00 | |
| *** mestery is now known as mestery_afk | 15:01 | |
| *** pradeep has quit IRC | 15:02 | |
| *** ramishra has joined #tripleo | 15:04 | |
| *** marun has joined #tripleo | 15:08 | |
| *** marun has quit IRC | 15:09 | |
| *** marun has joined #tripleo | 15:09 | |
| *** marun has quit IRC | 15:09 | |
| *** marun has joined #tripleo | 15:10 | |
| *** marun has quit IRC | 15:10 | |
| *** ramishra has quit IRC | 15:12 | |
| *** marun has joined #tripleo | 15:14 | |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Get rid of overspecification in vm.xml. https://review.openstack.org/93976 | 15:18 |
| *** marun has quit IRC | 15:22 | |
| *** wuhg has quit IRC | 15:22 | |
| *** tzumainn has joined #tripleo | 15:28 | |
| *** sirushti has joined #tripleo | 15:29 | |
| *** mestery_afk has quit IRC | 15:39 | |
| openstackgerrit | Ana Krivokapic proposed a change to openstack/tuskar-ui: Change page header heading to H1 https://review.openstack.org/115325 | 15:42 |
| *** ramishra has joined #tripleo | 15:43 | |
| regebro | what the heck does || : mean in a bash script? | 15:44 |
| *** dtantsur is now known as dtantsur|brb | 15:45 | |
| *** ramishra has quit IRC | 15:47 | |
| bnemec | regebro: or true. It's a way to allow commands to optionally fail without set -e killing the script. | 15:49 |
| regebro | bnemec: I see. | 15:50 |
| *** IvanBerezovskiy has left #tripleo | 15:50 | |
| openstackgerrit | Julia Kreger proposed a change to openstack/tripleo-heat-templates: Add nova metadata for group https://review.openstack.org/113358 | 15:52 |
| *** derekh has quit IRC | 15:52 | |
| *** loki184 has joined #tripleo | 15:53 | |
| *** sdake has joined #tripleo | 15:56 | |
| *** sdake has quit IRC | 15:56 | |
| *** sdake has joined #tripleo | 15:56 | |
| *** rdopieralski has joined #tripleo | 16:09 | |
| *** pblaho has quit IRC | 16:15 | |
| *** sdake has quit IRC | 16:17 | |
| *** sdake has joined #tripleo | 16:17 | |
| *** penick has joined #tripleo | 16:19 | |
| *** sdake has quit IRC | 16:20 | |
| *** rushiagr is now known as rushiagr_away | 16:20 | |
| *** sdake has joined #tripleo | 16:20 | |
| *** marun has joined #tripleo | 16:21 | |
| *** rushiagr_away is now known as rushiagr | 16:22 | |
| *** marun has quit IRC | 16:23 | |
| *** marun has joined #tripleo | 16:24 | |
| *** marun has quit IRC | 16:25 | |
| *** eghobo has joined #tripleo | 16:26 | |
| *** julim has joined #tripleo | 16:27 | |
| *** cody-somerville has quit IRC | 16:28 | |
| *** marun has joined #tripleo | 16:32 | |
| *** dshulyak_ has joined #tripleo | 16:35 | |
| *** ifarkas has quit IRC | 16:40 | |
| openstackgerrit | A change was merged to openstack/tuskar: Add set value functionality to plans manager https://review.openstack.org/113986 | 16:41 |
| *** eghobo has quit IRC | 16:41 | |
| openstackgerrit | A change was merged to openstack/tuskar: Fix for scaling default https://review.openstack.org/114002 | 16:42 |
| openstackgerrit | A change was merged to openstack/tuskar: Fix for roles with dashes in the name https://review.openstack.org/114004 | 16:43 |
| openstackgerrit | A change was merged to openstack/tuskar: Wires up REST API to underlying Tuskar logic https://review.openstack.org/113988 | 16:45 |
| *** jkraj has quit IRC | 16:47 | |
| openstackgerrit | Alexis Lee proposed a change to openstack-infra/tripleo-ci: Add firewall traffic on port 27410 https://review.openstack.org/115341 | 16:48 |
| openstackgerrit | Alexis Lee proposed a change to openstack-infra/tripleo-ci: Add firewall traffic on port 27410 https://review.openstack.org/115341 | 16:52 |
| *** dsneddon has quit IRC | 16:56 | |
| *** dsneddon has joined #tripleo | 16:57 | |
| *** rdopieralski has quit IRC | 16:57 | |
| openstackgerrit | Alexis Lee proposed a change to openstack-infra/tripleo-ci: Add firewall traffic on port 27410 https://review.openstack.org/115341 | 16:59 |
| *** penick has quit IRC | 17:01 | |
| *** eghobo has joined #tripleo | 17:03 | |
| openstackgerrit | Jonathan Brownell proposed a change to openstack/diskimage-builder: Increase source-repositories support for tarballs https://review.openstack.org/104694 | 17:03 |
| *** penick has joined #tripleo | 17:04 | |
| openstackgerrit | Alexis Lee proposed a change to openstack-infra/tripleo-ci: Add firewall traffic on port 27410 https://review.openstack.org/115341 | 17:04 |
| *** akrivoka has quit IRC | 17:05 | |
| openstackgerrit | Alexis Lee proposed a change to openstack-infra/tripleo-ci: Add firewall traffic on port 27410 https://review.openstack.org/115341 | 17:06 |
| *** loki184 has quit IRC | 17:12 | |
| *** dtantsur|brb is now known as dtantsur | 17:14 | |
| *** jang1 has joined #tripleo | 17:15 | |
| *** loki184 has joined #tripleo | 17:26 | |
| *** eghobo has quit IRC | 17:26 | |
| *** eghobo has joined #tripleo | 17:27 | |
| *** pradeep has joined #tripleo | 17:32 | |
| *** killer_prince has quit IRC | 17:37 | |
| *** loki184 has quit IRC | 17:37 | |
| *** ChuckC has quit IRC | 17:37 | |
| *** pcrews has quit IRC | 17:41 | |
| *** eghobo has quit IRC | 17:41 | |
| *** pcrews has joined #tripleo | 17:41 | |
| *** loki184 has joined #tripleo | 17:44 | |
| *** loki184 has quit IRC | 17:55 | |
| *** panda has quit IRC | 17:55 | |
| *** panda has joined #tripleo | 17:56 | |
| *** ChuckC has joined #tripleo | 18:00 | |
| *** chuckC_ has joined #tripleo | 18:02 | |
| *** lucasagomes is now known as lucas-dinner | 18:03 | |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Setup MySQL root password https://review.openstack.org/103514 | 18:05 |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/os-cloud-config: Updated from global requirements https://review.openstack.org/114332 | 18:06 |
| *** pensu has joined #tripleo | 18:07 | |
| *** penick has quit IRC | 18:08 | |
| *** sirushti has left #tripleo | 18:08 | |
| *** pensu has quit IRC | 18:15 | |
| *** weshay_mtg has quit IRC | 18:21 | |
| *** weshay has joined #tripleo | 18:21 | |
| *** jp_at_hp has quit IRC | 18:22 | |
| *** dtantsur is now known as dtantsur|afk | 18:22 | |
| dsneddon | I have a TripleO undercloud set up, and I've used heat to import my overcloud nodes. When I click "deploy" in Horizon, I get the error "HTTPResponse instance has no attribute 'headers'". How do I troubleshoot that? | 18:22 |
| *** pradeep has quit IRC | 18:23 | |
| dsneddon | There are no useful HTTP errors in the Apache logs. A packet capture shows HTTP 200's, so I don't know where to look next. | 18:23 |
| *** penick has joined #tripleo | 18:24 | |
| *** bmahalakshmi has quit IRC | 18:26 | |
| *** pelix has quit IRC | 18:28 | |
| openstackgerrit | Jonathan Brownell proposed a change to openstack/diskimage-builder: Increase source-repositories support for tarballs https://review.openstack.org/104694 | 18:39 |
| openstackgerrit | Jonathan Brownell proposed a change to openstack/diskimage-builder: Increase source-repositories support for tarballs https://review.openstack.org/104694 | 18:44 |
| ccrouch | dsneddon: looks like maybe https://bugs.launchpad.net/tuskar/+bug/1313731? | 18:44 |
| uvirtbot | Launchpad bug 1313731 in python-tuskarclient "Error when using Tuskar to redeploy the overcloud" [Medium,Fix released] | 18:44 |
| ccrouch | you might have more luck in #tuskar | 18:44 |
| *** cwolferh has joined #tripleo | 18:45 | |
| dsneddon | ccrouch, I believe you are right. The change hasn't made it to RDO downstream. | 18:45 |
| *** cwolferh_ has quit IRC | 18:45 | |
| lifeless | morning | 18:48 |
| *** rushiagr is now known as rushiagr_away | 18:54 | |
| *** penick has quit IRC | 18:56 | |
| *** penick has joined #tripleo | 18:59 | |
| Ng | Is it almost meeting time this week? | 18:59 |
| *** pensu has joined #tripleo | 19:00 | |
| bnemec | Ng: I think this is the alternate week. | 19:05 |
| lifeless | Ng: 'no' but I'd like a timeslice after I get off this call | 19:09 |
| Ng | lifeless: sure. Approx eta? Just stuffing a bit of dinner in, atm | 19:13 |
| *** jprovazn has joined #tripleo | 19:13 | |
| *** ccrouch has quit IRC | 19:21 | |
| *** akuznetsov has quit IRC | 19:41 | |
| *** marun has quit IRC | 19:49 | |
| lifeless | Ng: top of the hour | 19:53 |
| *** ccrouch has joined #tripleo | 19:54 | |
| Ng | lifeless: k | 19:58 |
| *** rlandy has quit IRC | 19:59 | |
| *** stevebaker has quit IRC | 20:00 | |
| *** stevebaker has joined #tripleo | 20:00 | |
| lifeless | Ng: so hi | 20:04 |
| Ng | lifeless: such hey | 20:04 |
| lifeless | hangout? | 20:04 |
| Ng | lifeless: k, lemmie just get my headset ready | 20:05 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Allow to specify an empty package list in pkg_map https://review.openstack.org/113246 | 20:05 |
| bnemec | ERROR: The Parameter (MysqlRootPassword) was not defined in template. | 20:06 |
| *** pensu has quit IRC | 20:06 | |
| bnemec | :-( | 20:06 |
| bnemec | I think I may have done it again. | 20:06 |
| bnemec | Those two mysql changes both passed CI individually, but I'm thinking they may not work together. | 20:06 |
| bnemec | MysqlRootPassword is a resource in the template, but is being passed in as a parameter in the stack-create call. | 20:07 |
| *** andreaf_ has joined #tripleo | 20:11 | |
| *** andreaf_ has quit IRC | 20:12 | |
| *** andreaf_ has joined #tripleo | 20:12 | |
| *** andreaf_ has quit IRC | 20:12 | |
| *** andreaf_ has joined #tripleo | 20:13 | |
| lifeless | bnemec: propose the revert, I'll +2 :) | 20:14 |
| *** andreaf has quit IRC | 20:14 | |
| bnemec | lifeless: Okay, testing to make sure that fixes it locally but I'm pretty sure that's my problem. | 20:15 |
| lifeless | bnemec: https://review.openstack.org/#/c/103449/ please too :) | 20:15 |
| bnemec | lifeless: Yeah, I'm working my way through that one. My knowledge on the subject is...limited. ;-) | 20:15 |
| slagle | bnemec: i don't understand how 103514 ever passed CI | 20:20 |
| slagle | in fact, i see no mention of MysqlRootPassword in the CI logs at all | 20:20 |
| greghaynes | did it pass CI? | 20:20 |
| bnemec | slagle: Hmm, that's a good point. That should never have worked. | 20:20 |
| slagle | um, i have a bad feeling | 20:21 |
| *** markmc has quit IRC | 20:21 | |
| greghaynes | oh, p4 did pass | 20:21 |
| slagle | are we still testing what we think we're testing after this merged...https://review.openstack.org/#/c/89974/ | 20:21 |
| greghaynes | haha | 20:22 |
| openstackgerrit | Ben Nemec proposed a change to openstack/tripleo-incubator: Revert "Setup MySQL root password" https://review.openstack.org/115395 | 20:24 |
| *** pmyers has quit IRC | 20:26 | |
| openstackgerrit | Ben Nemec proposed a change to openstack/tripleo-incubator: Intentionally broken commit https://review.openstack.org/115396 | 20:31 |
| *** dprince has quit IRC | 20:34 | |
| *** eghobo has joined #tripleo | 20:34 | |
| *** eghobo has quit IRC | 20:35 | |
| *** athomas has quit IRC | 20:36 | |
| slagle | bnemec: looking at the successful run on https://review.openstack.org/#/c/115303/ | 20:38 |
| slagle | it looks like it tested it with the old change | 20:38 |
| slagle | err "the old code" | 20:39 |
| slagle | http://logs.openstack.org/03/115303/1/check-tripleo/check-tripleo-novabm-overcloud-f20-nonha/891e937/console.html#_2014-08-19_15_11_23_847 | 20:39 |
| bnemec | slagle: Yes it does. :-( | 20:40 |
| slagle | i think we need to revert that /opt/git merge | 20:40 |
| greghaynes | yeaaaa | 20:40 |
| greghaynes | :/ | 20:40 |
| greghaynes | push up the revert | 20:40 |
| bnemec | Looking that way. | 20:40 |
| openstackgerrit | James Slagle proposed a change to openstack-infra/tripleo-ci: Revert "Use /opt/git instead of /opt/stack/new" https://review.openstack.org/115399 | 20:41 |
| * slagle mashes gerrit revert button | 20:41 | |
| *** stevebaker has quit IRC | 20:41 | |
| *** stevebaker has joined #tripleo | 20:41 | |
| openstackgerrit | James Slagle proposed a change to openstack-infra/tripleo-ci: Revert "Use /opt/git instead of /opt/stack/new" https://review.openstack.org/115399 | 20:42 |
| greghaynes | clarkb: ^ | 20:42 |
| openstackgerrit | Jonathan Brownell proposed a change to openstack/diskimage-builder: Increase source-repositories support for tarballs https://review.openstack.org/104694 | 20:42 |
| greghaynes | clarkb: Save us clarkb, youre our only hope | 20:42 |
| clarkb | http://logs.openstack.org/03/115303/1/check-tripleo/check-tripleo-novabm-overcloud-f20-nonha/891e937/logs/devstack-gate-setup-workspace-new.txt.gz | 20:43 |
| clarkb | thats the interesting log | 20:43 |
| clarkb | what repo needs updating? | 20:43 |
| greghaynes | tripleo-incubator | 20:44 |
| greghaynes | I see it in the list of project in devstack-gate | 20:44 |
| clarkb | http://logs.openstack.org/03/115303/1/check-tripleo/check-tripleo-novabm-overcloud-f20-nonha/891e937/logs/devstack-gate-setup-workspace-new.txt.gz#_2014-08-19_14_34_51_905 | 20:45 |
| clarkb | so it updated. make sure you are running out of that dir? | 20:45 |
| lifeless | ok breakfast | 20:46 |
| bnemec | clarkb: I think we're not. | 20:46 |
| bnemec | http://logs.openstack.org/03/115303/1/check-tripleo/check-tripleo-novabm-overcloud-f20-nonha/891e937/logs/devstack-gate-setup-workspace-new.txt.gz#_2014-08-19_14_34_50_807 | 20:46 |
| lifeless | then I'm going to nag reviewers mercilessly to review this vlan patch so I can unblock dprince's network work | 20:46 |
| bnemec | Oh, wiat. | 20:46 |
| bnemec | *wait | 20:46 |
| clarkb | d-g is definitely doing the correct thing | 20:46 |
| clarkb | I am saying make sure your thing actually uses the correct path | 20:46 |
| greghaynes | ttp://logs.openstack.org/03/115303/1/check-tripleo/check-tripleo-novabm-overcloud-f20-nonha/891e937/console.html#_2014-08-19_14_38_20_928 | 20:47 |
| greghaynes | http://logs.openstack.org/03/115303/1/check-tripleo/check-tripleo-novabm-overcloud-f20-nonha/891e937/console.html#_2014-08-19_14_38_20_928 | 20:48 |
| greghaynes | so looks like we export correct tripleo_root | 20:48 |
| bnemec | Oh, hold on. 115303 is not an incubator change. | 20:48 |
| bnemec | It's t-i-e. | 20:48 |
| clarkb | bnemec: doesn't matter | 20:48 |
| clarkb | bnemec: the mechanism is the same | 20:48 |
| clarkb | unless you meant that you expected that job log to show a proposed update to incubator | 20:48 |
| clarkb | greghaynes: where does that export happen as the sudoing in d-g may throw it away | 20:49 |
| greghaynes | clarkb: nice. It happens in toci_devtest.sh | 20:49 |
| greghaynes | right before we devtest_seed | 20:49 |
| *** mestery has joined #tripleo | 20:49 | |
| greghaynes | so I dont think there can be any sudes between then | 20:49 |
| greghaynes | s/sudes/sudos | 20:49 |
| clarkb | greghaynes: ya sounds right | 20:49 |
| *** mestery has quit IRC | 20:49 | |
| *** mestery has joined #tripleo | 20:50 | |
| greghaynes | ok, maybe we wait for smoketest to confirm this is correct before getting too far ahead of ourselves | 20:50 |
| greghaynes | Time to watch https://jenkins05.openstack.org/job/check-tripleo-novabm-undercloud-precise-nonha/230/console patiently ;) | 20:51 |
| bnemec | So, we pull down the proposed change, but we're in /opt/stack/new when we do it. I think. | 20:53 |
| bnemec | logs.openstack.org/03/115303/1/check-tripleo/check-tripleo-novabm-overcloud-f20-nonha/891e937/logs/devstack-gate-setup-workspace-new.txt.gz#_2014-08-19_14_34_50_807/ | 20:53 |
| bnemec | Argh, copy-paste fail. | 20:54 |
| *** andreaf_ has quit IRC | 20:54 | |
| bnemec | http://logs.openstack.org/03/115303/1/check-tripleo/check-tripleo-novabm-overcloud-f20-nonha/891e937/logs/devstack-gate-setup-workspace-new.txt.gz#_2014-08-19_14_34_50_807/ | 20:54 |
| *** andreaf_ has joined #tripleo | 20:54 | |
| bnemec | That didn't work right either. :-( | 20:54 |
| *** jprovazn has quit IRC | 20:54 | |
| *** andreaf_ has quit IRC | 20:54 | |
| *** andreaf_ has joined #tripleo | 20:55 | |
| *** sdake has quit IRC | 20:55 | |
| *** andreaf_ has quit IRC | 20:55 | |
| lifeless | why do we think the /opt/git thing is broken? | 20:55 |
| *** andreaf_ has joined #tripleo | 20:55 | |
| slagle | specifically, i'm not sure | 20:56 |
| slagle | generally, our CI jobs are not testing the proposed changes | 20:56 |
| greghaynes | The change which adds MysqlRootPassword as a param to heat env - I was unable to find it being passed in | 20:56 |
| slagle | that seemed like a likely culprit | 20:57 |
| bnemec | I think we're checking out the proposed change to /opt/stack/new, then running out of /opt/git. | 20:57 |
| *** jtomasek has quit IRC | 20:57 | |
| lifeless | ah | 20:59 |
| lifeless | that would be a hugish problem | 20:59 |
| greghaynes | http://logs.openstack.org/14/103514/4/check-tripleo/check-tripleo-ironic-undercloud-precise-nonha/95c4ee2/console.html | 20:59 |
| greghaynes | We should be able to see MysqlRootPassword in the Parameters | 21:00 |
| greghaynes | at http://logs.openstack.org/14/103514/4/check-tripleo/check-tripleo-ironic-undercloud-precise-nonha/95c4ee2/console.html#_2014-08-19_15_41_24_901 | 21:00 |
| *** untriaged-bot has joined #tripleo | 21:00 | |
| untriaged-bot | Untriaged bugs so far: | 21:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1356637 | 21:00 |
| uvirtbot | Launchpad bug 1356637 in tripleo "os-refresh-config fails in hosts element when update is different" [Undecided,New] | 21:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1355898 | 21:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1357346 | 21:00 |
| uvirtbot | Launchpad bug 1355898 in tripleo "mariadb doesn't start after an upgrade" [Undecided,New] | 21:00 |
| uvirtbot | Launchpad bug 1357346 in tripleo "undercloud failing to come up on devtest due to pacemaker in os-refresh-config" [Undecided,New] | 21:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1351010 | 21:00 |
| uvirtbot | Launchpad bug 1351010 in ironic "Update to syslinux 6 breaks TFTPboot" [High,Confirmed] | 21:00 |
| *** untriaged-bot has quit IRC | 21:00 | |
| greghaynes | https://jenkins05.openstack.org/job/check-tripleo-novabm-undercloud-precise-nonha/230/console did not exit 1 | 21:01 |
| greghaynes | bnemec: ^ | 21:01 |
| bnemec | Yeah, I didn't think it would. | 21:01 |
| *** weshay has quit IRC | 21:01 | |
| lifeless | ok | 21:03 |
| lifeless | so lets roll that back :( | 21:03 |
| greghaynes | haha | 21:03 |
| greghaynes | yeaaa, might be a good idea :) | 21:03 |
| bnemec | lifeless: https://review.openstack.org/115399 | 21:03 |
| greghaynes | Do we want to even wait for CI on that one? | 21:04 |
| lifeless | no | 21:04 |
| slagle | nah, let's +A it | 21:04 |
| lifeless | if it breaks it just tells us trunk is broken | 21:04 |
| slagle | then recheck the intentionally broken commit | 21:04 |
| openstackgerrit | A change was merged to openstack-infra/tripleo-ci: Revert "Use /opt/git instead of /opt/stack/new" https://review.openstack.org/115399 | 21:04 |
| greghaynes | +A | 21:04 |
| slagle | make sure it fails this time :) | 21:04 |
| openstackgerrit | Ben Nemec proposed a change to openstack/tripleo-incubator: Intentionally broken commit https://review.openstack.org/115396 | 21:05 |
| *** cody-somerville has joined #tripleo | 21:07 | |
| *** cody-somerville has joined #tripleo | 21:07 | |
| *** cody-somerville_ has joined #tripleo | 21:10 | |
| *** julim has quit IRC | 21:11 | |
| lifeless | maybe we should delete /opt/stack/new | 21:13 |
| lifeless | e.g. mv /opt/stack/{new,obsolete} | 21:13 |
| lifeless | to ensure that we're really using the new location | 21:13 |
| *** lblanchard has quit IRC | 21:13 | |
| lifeless | clarkb: is it possible to have an inverted job? | 21:14 |
| clarkb | no | 21:14 |
| clarkb | I am pretty sure you all are using the wrong path | 21:14 |
| lifeless | we could do it in 'userspace' I guess | 21:14 |
| *** cody-somerville has quit IRC | 21:14 | |
| clarkb | why would you move? | 21:14 |
| clarkb | everything in /opt/stack/new is up to date | 21:14 |
| lifeless | clarkb: mordred put the patch up | 21:14 |
| *** dshulyak_ has quit IRC | 21:14 | |
| lifeless | clarkb: we assumed he knew what he was doing | 21:14 |
| clarkb | mordred: that change looks completely wrong to me | 21:16 |
| *** cody-somerville_ has quit IRC | 21:16 | |
| clarkb | lifeless: the way d-g works is it copies from /opt/git to /opt/stack/new then updates in /opt/stack/new | 21:16 |
| clarkb | if you use /opt/git as a source of truth you will break because it is just a cache | 21:16 |
| lifeless | clarkb: I believe morded is working on not needing the copy to /opt/stack/new to improve performance | 21:17 |
| clarkb | lifeless: that change should happen in devstack-gate not in toci | 21:17 |
| lifeless | clarkb: possibly in both; anyhow - argue with mordred :) | 21:18 |
| bnemec | Hey look, we exited 1: https://jenkins01.openstack.org/job/check-tripleo-novabm-undercloud-precise-nonha/204/console | 21:19 |
| bnemec | So CI should be sane again. | 21:19 |
| greghaynes | \O/ | 21:22 |
| greghaynes | so, we should probably recheck things | 21:22 |
| clarkb | lifeless: I may have missed it but do you know if mordred was doing this as a general thing for other projects too? | 21:23 |
| *** penick has quit IRC | 21:24 | |
| *** penick has joined #tripleo | 21:31 | |
| *** penick has quit IRC | 21:35 | |
| *** mestery_ has joined #tripleo | 21:36 | |
| lifeless | clarkb: there was a #TODO somewhere at one point | 21:37 |
| *** mestery has quit IRC | 21:37 | |
| lifeless | clarkb: see commit 2978e47c0350da1db71704717cb04e7a302a0899 | 21:37 |
| lifeless | clarkb: in d-g | 21:38 |
| lifeless | clarkb: which I think was/is related | 21:38 |
| clarkb | lifeless: all that does is use /opt/git as the cache | 21:38 |
| clarkb | lifeless: it does not use it was the working repos | 21:38 |
| clarkb | which is why the tripleo change broke because morded switched to using that as the working repo | 21:39 |
| lifeless | clarkb: so put this down to me being confused and lets ask mordred when he undeserts. | 21:39 |
| clarkb | wfm | 21:39 |
| *** penick has joined #tripleo | 21:39 | |
| *** mestery_ is now known as mestery | 21:40 | |
| *** jang1 has quit IRC | 21:44 | |
| *** mestery has quit IRC | 21:45 | |
| *** mestery has joined #tripleo | 21:46 | |
| openstackgerrit | Julia Kreger proposed a change to openstack/os-collect-config: CORE-1270 Allow using non-default collectors https://review.openstack.org/114116 | 21:47 |
| lifeless | TheJulia: whats the CORE-%d for? Thts irrelevant upstream :) | 21:49 |
| openstackgerrit | Julia Kreger proposed a change to openstack/os-collect-config: CORE-1270 Allow using non-default collectors https://review.openstack.org/114116 | 21:50 |
| bnemec | + exit 1 | 21:54 |
| bnemec | Sigh. I forgot to undo the broken change before I re-ran devtest. | 21:54 |
| ccrouch | TheJulia: I think CORE-1270 resisted your attempts to remove it | 21:55 |
| *** mestery has quit IRC | 22:01 | |
| *** sdake has joined #tripleo | 22:07 | |
| *** sdake has quit IRC | 22:07 | |
| *** sdake has joined #tripleo | 22:07 | |
| TheJulia | borked .git/config | 22:23 |
| openstackgerrit | Julia Kreger proposed a change to openstack/os-collect-config: Allow using non-default collectors https://review.openstack.org/114116 | 22:24 |
| jroll | is tripleo CI ok? I have a patch that has hit this three times in a row: https://bugs.launchpad.net/openstack-ci/+bug/1344050 | 22:31 |
| uvirtbot | Launchpad bug 1344050 in openstack-ci "When installing horizon dependencies: fatal error: pcre.h: No such file or directory" [Undecided,New] | 22:31 |
| jroll | said patch: https://review.openstack.org/#/c/115148/ | 22:31 |
| *** sdake has quit IRC | 22:34 | |
| bnemec | jroll: Your last failure was actually http://logs.openstack.org/48/115148/7/check-tripleo/check-tripleo-ironic-undercloud-precise-nonha/0f5fe5f/console.html#_2014-08-19_22_19_35_898 | 22:39 |
| bnemec | A broken patch made it in because our CI wasn't testing new changes properly for a while. | 22:39 |
| bnemec | We're reverting the bad change in https://review.openstack.org/#/c/115395/ | 22:39 |
| jroll | oh huh | 22:40 |
| jroll | thanks bnemec | 22:40 |
| *** penick has quit IRC | 22:50 | |
| *** mestery has joined #tripleo | 22:53 | |
| *** mestery_ has joined #tripleo | 23:02 | |
| *** mestery has quit IRC | 23:02 | |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-image-elements: Remove hardcoded version. https://review.openstack.org/115426 | 23:03 |
| *** mestery_ has quit IRC | 23:07 | |
| *** mestery has joined #tripleo | 23:07 | |
| *** mestery has quit IRC | 23:12 | |
| *** mestery has joined #tripleo | 23:13 | |
| *** spzala has quit IRC | 23:29 | |
| *** jpeeler has quit IRC | 23:30 | |
| *** ccrouch has quit IRC | 23:37 | |
| *** mestery has quit IRC | 23:41 | |
| *** mestery has joined #tripleo | 23:44 | |
| *** mestery has quit IRC | 23:45 | |
| *** yamahata has joined #tripleo | 23:53 | |
| *** marun has joined #tripleo | 23:56 | |
| *** markmc has joined #tripleo | 23:57 | |
| *** sdake has joined #tripleo | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!