| *** openstackgerrit has quit IRC | 00:16 | |
| *** openstackgerrit has joined #openstack-release | 00:17 | |
| *** jhesketh has joined #openstack-release | 01:15 | |
| *** jhesketh has quit IRC | 01:16 | |
| openstackgerrit | Kirill Zaitsev proposed openstack/releases: Update murano deliverables with 1.0.1 version https://review.openstack.org/248969 | 01:29 |
|---|---|---|
| *** jgriffith_away is now known as jgriffith | 01:30 | |
| kzaitsev_mb | ok, I think th reason it was not included — was that we tagged 1.0.0 incorrectly =/ So probably this caused some confustion. | 01:31 |
| kzaitsev_mb | Wonder if I should remove 1.0.0 (I believe I shouldn't) or somehow mark, that it was incorrectly tagged and should not be used. | 01:32 |
| *** kzaitsev_mb has quit IRC | 01:45 | |
| *** bswartz has joined #openstack-release | 01:52 | |
| *** kzaitsev_mb has joined #openstack-release | 01:56 | |
| *** jhesketh has joined #openstack-release | 01:58 | |
| *** jhesketh has quit IRC | 02:03 | |
| *** jhesketh has joined #openstack-release | 02:06 | |
| *** amotoki has joined #openstack-release | 02:13 | |
| *** kzaitsev_mb has quit IRC | 02:15 | |
| *** david-ly_ has quit IRC | 02:36 | |
| *** dims has joined #openstack-release | 02:40 | |
| dims | tonyb : can you please fix this merge conflict? https://review.openstack.org/#/c/248530/ | 02:41 |
| tonyb | dims: sure. | 02:43 |
| openstackgerrit | Tony Breeds proposed openstack/releases: Add a warning for new mitaka releases of oslo.messaging https://review.openstack.org/248530 | 02:51 |
| dims | thanks tonyb +2A | 02:52 |
| tonyb | dims: this time without addind EOF to the file ;p | 02:52 |
| dims | tonyb even better :) | 03:12 |
| tonyb | dims: vim does it by default can't work out how to turn it off :( I can use git add -p to work around it | 03:13 |
| *** dims has quit IRC | 03:17 | |
| *** dims has joined #openstack-release | 03:22 | |
| *** dims has quit IRC | 03:35 | |
| *** david-lyle has joined #openstack-release | 04:08 | |
| *** aukhan has joined #openstack-release | 05:51 | |
| *** ifat_afek has joined #openstack-release | 07:05 | |
| *** jgriffith is now known as jgriffith_away | 07:15 | |
| openstackgerrit | Merged openstack/reno: force the right return type for mocked urandom https://review.openstack.org/246583 | 08:06 |
| *** bdemers has quit IRC | 08:12 | |
| *** bdemers has joined #openstack-release | 08:14 | |
| *** ifat_afek has quit IRC | 09:06 | |
| *** mattoliverau has quit IRC | 09:10 | |
| *** matt6434 has joined #openstack-release | 09:11 | |
| *** openstackgerrit has quit IRC | 09:16 | |
| *** dtantsur|afk is now known as dtantsur | 09:17 | |
| *** openstackgerrit has joined #openstack-release | 09:17 | |
| *** ifat_afek has joined #openstack-release | 09:46 | |
| *** kzaitsev_mb has joined #openstack-release | 10:20 | |
| *** prometheanfire has quit IRC | 10:27 | |
| *** prometheanfire has joined #openstack-release | 10:29 | |
| openstackgerrit | Merged openstack/reno: add verbose and quiet options to command line for debugging https://review.openstack.org/246582 | 10:43 |
| openstackgerrit | Merged openstack/reno: scan all changes and filter on files in reno not git log https://review.openstack.org/246584 | 10:43 |
| *** aukhan has quit IRC | 10:48 | |
| *** dtantsur is now known as dtantsur|brb | 10:50 | |
| *** matt6434 has quit IRC | 10:54 | |
| *** matt6434 has joined #openstack-release | 10:55 | |
| *** bswartz has quit IRC | 10:56 | |
| *** bswartz has joined #openstack-release | 11:00 | |
| *** amotoki has quit IRC | 11:01 | |
| *** kzaitsev_mb has quit IRC | 11:01 | |
| *** dims has joined #openstack-release | 11:06 | |
| *** kzaitsev_mb has joined #openstack-release | 11:22 | |
| *** dtantsur|brb is now known as dtantsur | 11:51 | |
| *** bdemers has quit IRC | 12:55 | |
| *** gordc has joined #openstack-release | 12:59 | |
| *** jhesketh has quit IRC | 13:06 | |
| *** jhesketh has joined #openstack-release | 13:07 | |
| *** josh6627 has joined #openstack-release | 13:29 | |
| *** josh6627 has quit IRC | 13:29 | |
| *** doug-fish has joined #openstack-release | 13:30 | |
| *** jhesketh has quit IRC | 13:30 | |
| *** jhesketh has joined #openstack-release | 14:02 | |
| *** jhesketh has quit IRC | 14:03 | |
| *** ifat_afek_ has joined #openstack-release | 14:03 | |
| *** jhesketh has joined #openstack-release | 14:04 | |
| *** ifat_afek has quit IRC | 14:05 | |
| *** ifat_afek has joined #openstack-release | 14:09 | |
| *** ifat_afek_ has quit IRC | 14:12 | |
| openstackgerrit | Graham Hayes proposed openstack/reno: Add deprecations section to release notes https://review.openstack.org/247704 | 14:27 |
| ttx | oops | 14:28 |
| mugsie | ? | 14:29 |
| ttx | dhellmann, dims: I just accidentally pushed a 3.0.1 tag for oslo.rootwrap, during new release tooling testing | 14:31 |
| dims | ttx gerrit master tip? | 14:31 |
| ttx | yeah, over 3.0.0 | 14:31 |
| ttx | I can kill it really fast | 14:31 |
| dims | should be fine | 14:31 |
| ttx | or just let it live | 14:31 |
| *** amotoki has joined #openstack-release | 14:32 | |
| dims | https://github.com/openstack/oslo.rootwrap/compare/3.0.0...3.0.1 | 14:32 |
| dims | seems identical | 14:32 |
| ttx | yeah, same commit | 14:32 |
| dims | no worries | 14:32 |
| dims | we should be ok | 14:32 |
| *** mriedem_away is now known as mriedem | 14:33 | |
| openstackgerrit | Thierry Carrez proposed openstack-infra/release-tools: New tooling for service releases https://review.openstack.org/249241 | 14:37 |
| openstackgerrit | Thierry Carrez proposed openstack-infra/release-tools: [WIP] New tooling for service releases https://review.openstack.org/249241 | 14:38 |
| ttx | This is the thing I was testing ^ | 14:38 |
| ttx | dhellmann: still WIP ^ ping me when around to discuss what we actually want there | 14:39 |
| dhellmann | ttx: at least you didn't accidentally send like 100 emails about it #beenthere | 14:42 |
| dhellmann | ttx: do you think we'll be able to use release.sh for libraries, too? or do we need to make release_from_yaml.py decide whether to call that or release_postversion.sh for libs? | 14:43 |
| ttx | dhellmann: so that approach I took is taking release_postversion and strip it from the Launchpaddy things | 14:43 |
| ttx | and remove the release notes stuff since we may want to evolve that | 14:44 |
| dhellmann | ok, that's what I expected but your commit message called out service projects | 14:44 |
| dhellmann | ah, yeah, we need to decide what to do there. I think we still want release notes for libs to announce the new release | 14:44 |
| ttx | Ideally we would have a single thing that we could plug behind the tag jobs | 14:44 |
| dhellmann | exactly | 14:44 |
| ttx | To simplify the discussion let's discuss what we want to do at mitaka-1 | 14:44 |
| ttx | We said a number of things, I want to make sure we are on the same line | 14:45 |
| dhellmann | ++ | 14:45 |
| ttx | For mitaka-1 we'll tag 13.0.0.0b1 things | 14:45 |
| ttx | we said those would be requested through openstack/releases changes | 14:45 |
| ttx | but that we would somehow not make thoise appear on docs.o.o/releases | 14:46 |
| ttx | since they are technically not releases | 14:46 |
| ttx | Currently the release.sh script does not distinguish | 14:46 |
| ttx | between milestones and real releases | 14:46 |
| ttx | the only place where I considered detecting that was on the tag message | 14:47 |
| dhellmann | we said that d.o.o/releases would only show pre-releases if those were the most recent release | 14:47 |
| ttx | dhellmann: oh right, that | 14:47 |
| dhellmann | so that work will all be done inside the releases repo, and it can wait until after the milestone | 14:47 |
| *** jgriffith_away is now known as jgriffith | 14:47 | |
| *** kzaitsev_mb has quit IRC | 14:47 | |
| ttx | So the tag is pushed, that generates a tarball, the script actually doesn't wait for that | 14:47 |
| ttx | we should probably watch for tarball job failure instead of having the script wait | 14:48 |
| dhellmann | we used to wait for the tarball so we could download it and upload it to launchpad, right? | 14:48 |
| ttx | yes | 14:48 |
| dhellmann | ok. we may not need it for m1, but at some point we need to upload signatures to the tarballs server, right? is the thing building the tarball going to do that step? | 14:48 |
| ttx | then the script just posts messages on Launchpad bugs that happen to be mentioned between the last two tags | 14:49 |
| ttx | dhellmann: I suspect the automation will do all that ? | 14:49 |
| dhellmann | that would make more sense | 14:49 |
| ttx | since it will sign it can upload the signature | 14:49 |
| ttx | we already trust the bot at that stage | 14:49 |
| *** bdemers has joined #openstack-release | 14:49 | |
| dhellmann | right | 14:50 |
| ttx | but yeah, there is a part of the automation that will be triggered by the openstack/releases merge (basically what the release.sh script does) | 14:50 |
| ttx | and another that will be triggered by the tag push | 14:50 |
| dhellmann | ok, the script is good for services, but we need to decide where we're going to send the release note email -- maybe that's a separate job or a separate publisher for library releases? | 14:50 |
| ttx | (tarball generation and upload) | 14:50 |
| dhellmann | right | 14:51 |
| ttx | That leaves us with announcement, yes | 14:51 |
| ttx | What do we want to do there ? | 14:51 |
| dhellmann | I'm personally fine with including those emails for services as well as libraries, but that's not what we've done in the past | 14:51 |
| ttx | We have 3 types of announcements now | 14:52 |
| ttx | Manual ones like we did for liberty-3 | 14:52 |
| ttx | "We are thrilled to announce" type emails like we do for libs | 14:53 |
| ttx | and something we could build on top of reno | 14:53 |
| dhellmann | yeah, it would be interesting to see if we could convert the lib announcements to reno, too | 14:53 |
| ttx | do we need to keep "We are thrilled to announce" type emails if we start using reno ? | 14:54 |
| dhellmann | although one aspect of the existing announcements reno wouldn't give us is the requirements change | 14:54 |
| ttx | sounds like release announcements could just parrot the reno content ? | 14:54 |
| dhellmann | I would prefer to have 1 system for those emails. If we think reno can do it, we don't need all of what we have now | 14:54 |
| dhellmann | right | 14:54 |
| dhellmann | reno already has a way to produce the report for a given version as restructuredtext to the console, and rst2txt will format that nicely as the body of an email, so it won't be hard | 14:55 |
| dhellmann | the trick is that so far mostly services and not libs have adopted reno | 14:55 |
| ttx | so we might need a tool that builds the release announcement email for us to send after the fact, as a third step | 14:55 |
| ttx | until that can be rolled in the post-tag automation with the rest | 14:55 |
| dhellmann | that sounds right | 14:56 |
| dhellmann | I'll think about that some today | 14:56 |
| ttx | One grey area: the signatures -- should we do anything about that between now that we drop LP and when we'll have teh upload automation/signing set up ? | 14:56 |
| dhellmann | maybe that's something to build into reno itself, with a thin wrapper to pass it some arguments | 14:56 |
| dhellmann | we should talk to fungi about the signing part -- I haven't had a chance to catch up with him about my spec, yet, either. Maybe we can grab him today. | 14:57 |
| stevemar | ttx: thanks for replying to my questions on ML | 14:57 |
| ttx | So ideally we would build two new scripts. One to run on when the openstack/releases change is merged, ultimately to run as post-merge hook on that repo (let's call that one release.sh) and one to run when the tarball is built, ultimately to run as part of the tarball creation/upload job (let's call that one tarball.sh) | 14:59 |
| *** kzaitsev_mb has joined #openstack-release | 14:59 | |
| ttx | release.sh needs to push the tag and add mentions to the LP bugs | 14:59 |
| ttx | tarball.sh needs to generate signature and upload it and prepare release announcement | 15:00 |
| ttx | we may want to skip on the sig right now | 15:00 |
| dhellmann | rather than writing a new tarball.sh, we could also just add some steps to the existing script that is building the tarballs in the first place | 15:00 |
| ttx | sure, that would likely be in the same script in the end. But for mitaka-1 we might want it as a standalone tool to run manually | 15:01 |
| fungi | i do have a spec for hosting the tarball signatures, and also uploading them to pypi | 15:01 |
| dhellmann | ttx: sure | 15:01 |
| ttx | I'm just double-checking we need two tools/jobs. I think we do | 15:01 |
| dhellmann | fungi : that's right, I keep forgetting that's a separate spec | 15:02 |
| ttx | since those are about two separate repos | 15:02 |
| dhellmann | ttx: right | 15:02 |
| fungi | the tarball/wheel signing will happen on a trusted worker as an intermediate step between building the tarball and uploading it to pypi | 15:02 |
| ttx | If we agree, release.sh as I just proposed it is probably not very far of what we need for the first script. | 15:03 |
| dhellmann | fungi : we also want signatures for things not uploaded to pypi | 15:03 |
| dhellmann | ttx: yep, modulo the one comment I left about that read call | 15:03 |
| ttx | I'd say it should be even less aggressive validating input (that should run as an openstack/releases check job instead) | 15:03 |
| dhellmann | I think I agree with that | 15:04 |
| dhellmann | it should be safe to have it run with a tag that already exists, for example | 15:04 |
| fungi | 1. normal worker builds tarball/wheel and stores them on tarballs.o.o, 2. signing worker retrieves tarball/wheel and generates signatures which are then stored adjacent to them on tarballs.o.o, 3. upload worker retrieves tarball/wheel/signatures and validates them before uploading to pypi/maven central/npm/wherever | 15:04 |
| ttx | Are you fine working on the other script ? The "let's prepare a release announcement" one to run as part of the tarball automation | 15:04 |
| dhellmann | fungi : ok, I was worried about part of step 2 there and it sounds like that's covered | 15:05 |
| fungi | yeah, we'll host our own copies of the signatures | 15:05 |
| dhellmann | ttx: yeah, I'll work on an announce.sh | 15:05 |
| fungi | i've also set tarballs.o.o working via https since a few weeks ago | 15:05 |
| dhellmann | fungi : ok, good | 15:05 |
| dhellmann | ooh, I should update the releases repo links if they aren't already https | 15:05 |
| ttx | dhellmann: detecting that the Launchpad messages were already pushed would be a bit unreliable though. So the script would just repost the message | 15:05 |
| ttx | harmless, but not idempotent :) | 15:06 |
| dhellmann | yeah, I agree there | 15:06 |
| fungi | anyway, i'm hoping to get started on the signing work in the next few weeks, emergencies permitting | 15:06 |
| dhellmann | fungi : have you had a chance to review https://review.openstack.org/#/c/245907/ ? | 15:06 |
| ttx | dhellmann: I'll rewrite the tool so that it takes YAML as input | 15:07 |
| ttx | (release.sh) | 15:07 |
| dhellmann | fungi : feel like it's probably incomplete | 15:07 |
| fungi | likely not, i was out of town with no internet access for the past week+ | 15:07 |
| fungi | looking now | 15:07 |
| dhellmann | ttx: nah, we'll have release_from_yaml.py call release.sh | 15:07 |
| ttx | dhellmann: ah, ok | 15:07 |
| dhellmann | ttx: that way release_from_yaml can make multiple calls to it, and release.sh only needs to deal with one set of inputs | 15:07 |
| dhellmann | fungi : yeah, I thought probably not | 15:07 |
| fungi | yeah, looks like you uploaded that after i went on vacation | 15:07 |
| fungi | i'll stick it on my priority pile | 15:08 |
| dhellmann | ttx: do you want to update release_from_yaml.py, too? | 15:08 |
| dhellmann | fungi : thanks! | 15:08 |
| *** dims_ has joined #openstack-release | 15:08 | |
| ttx | dhellmann: I'll look into it yes | 15:08 |
| dhellmann | ok, thanks | 15:09 |
| *** dims has quit IRC | 15:09 | |
| dhellmann | I'll talk with dims about getting oslo converted over to use reno | 15:09 |
| dhellmann | I should probably make a more general announcement that we will want to use reno for release note emails for all projects, including libs | 15:10 |
| *** kzaitsev_mb has quit IRC | 15:10 | |
| dhellmann | ttx: since the libraries are developer-focused, what do you think about including the release notes in their regular documentation? | 15:15 |
| ttx | dhellmann: let's call the second script announce.sh and have a corresponding announce_from_yaml.py ? | 15:16 |
| dhellmann | instead of under d.o.o/releasenotes | 15:16 |
| ttx | dhellmann: sounds good | 15:16 |
| * ttx is about to gut release_from_yaml from announcy stuff | 15:16 | |
| dhellmann | ttx: sure, or a mode flag on release_from_yaml.py | 15:16 |
| dhellmann | k | 15:16 |
| * dhellmann ducks out briefly for some pre-holiday errands | 15:18 | |
| openstackgerrit | Thierry Carrez proposed openstack-infra/release-tools: Release tagging automation https://review.openstack.org/249241 | 15:40 |
| *** dims_ has quit IRC | 16:30 | |
| * dhellmann returns | 16:37 | |
| *** dims has joined #openstack-release | 16:46 | |
| dhellmann | ttx: if you'll approve https://review.openstack.org/#/c/248861/ I'll cut a reno release today | 16:49 |
| *** bnemec has quit IRC | 16:50 | |
| ttx | dhellmann: on it | 17:00 |
| ttx | done | 17:01 |
| * ttx runs to doctors appointment | 17:01 | |
| dhellmann | ttx: ty | 17:02 |
| openstackgerrit | Merged openstack/reno: add complex formatting example https://review.openstack.org/248861 | 17:03 |
| openstackgerrit | Doug Hellmann proposed openstack/releases: reno 1.1.0 https://review.openstack.org/249332 | 17:13 |
| *** dims has quit IRC | 17:18 | |
| *** dims_ has joined #openstack-release | 17:18 | |
| *** david-ly_ has joined #openstack-release | 17:19 | |
| *** david-l__ has joined #openstack-release | 17:23 | |
| *** david-ly_ has quit IRC | 17:23 | |
| *** david-lyle has quit IRC | 17:23 | |
| *** dtantsur is now known as dtantsur|afk | 17:28 | |
| *** david-l__ has quit IRC | 17:30 | |
| *** david-lyle has joined #openstack-release | 17:44 | |
| *** david-lyle has quit IRC | 18:03 | |
| *** david-lyle has joined #openstack-release | 18:11 | |
| *** mriedem has quit IRC | 18:17 | |
| *** ifat_afek has quit IRC | 18:20 | |
| *** mriedem has joined #openstack-release | 18:21 | |
| *** kzaitsev_mb has joined #openstack-release | 18:41 | |
| kzaitsev_mb | someone said murano here. =) | 18:41 |
| kzaitsev_mb | oh no. it was on infra, sotrry | 18:42 |
| *** bnemec has joined #openstack-release | 18:47 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/releases: New release for pylockfile https://review.openstack.org/249385 | 19:11 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/releases: New release for pylockfile https://review.openstack.org/249385 | 19:14 |
| *** kzaitsev_mb has quit IRC | 19:20 | |
| dims_ | kzaitsev_ws : murano | 19:21 |
| openstackgerrit | Merged openstack/releases: keystonemiddleware 2.3.2 (liberty) https://review.openstack.org/248926 | 19:23 |
| dims_ | dhellmann : switched to using the .1 release for the pylockfile ^^ | 19:24 |
| dhellmann | dims_ : ack, just a sec | 19:24 |
| dims_ | no hurries | 19:24 |
| *** kzaitsev_mb has joined #openstack-release | 19:30 | |
| dhellmann | lifeless : I'm going to be writing up instructions for liaisons to remove the version line from all setup.cfg files as part of the M1 tagging process. I want to confirm with you that I'm giving the instructions in the right order, which is that we should tag, and then remove the version line. I'm basing that on this experiment: http://paste.openstack.org/show/479801/ Does that match what you would expect? | 19:30 |
| kzaitsev_mb | yeah, dims_ it works like that =) I get a push notififcation when someone mentions me or the m word =) But right now I have terrible connectivity problems. (sorry for off-topic =)) | 19:31 |
| dims_ | haha, yes i do the same for oslo as well :) | 19:32 |
| openstackgerrit | Merged openstack/releases: reno 1.1.0 https://review.openstack.org/249332 | 19:32 |
| lifeless | dhellmann: yes | 19:33 |
| lifeless | dhellmann: that is what I'd expect | 19:33 |
| dhellmann | lifeless : perfect, thank you | 19:34 |
| lifeless | dhellmann: we should document this in the pbr docs | 19:34 |
| dhellmann | lifeless : indeed, I'll make a note to do that | 19:34 |
| *** kzaitsev_mb has quit IRC | 19:36 | |
| *** kzaitsev_mb has joined #openstack-release | 19:41 | |
| *** kzaitsev_mb has quit IRC | 19:50 | |
| *** kzaitsev_mb has joined #openstack-release | 19:52 | |
| *** dims has joined #openstack-release | 19:53 | |
| *** dims_ has quit IRC | 19:56 | |
| openstackgerrit | Doug Hellmann proposed openstack-infra/release-tools: move add_comment.py to a console script https://review.openstack.org/249396 | 19:58 |
| *** bdemers has quit IRC | 20:04 | |
| *** bdemers has joined #openstack-release | 20:05 | |
| *** matt6434 is now known as mattoliverau | 20:52 | |
| openstackgerrit | Merged openstack/releases: Add a warning for new mitaka releases of oslo.messaging https://review.openstack.org/248530 | 20:53 |
| *** kzaitsev_mb has quit IRC | 21:10 | |
| *** kzaitsev_mb has joined #openstack-release | 21:13 | |
| openstackgerrit | Doug Hellmann proposed openstack-infra/release-tools: remove debug statement in release_from_yaml.py https://review.openstack.org/249426 | 21:21 |
| openstackgerrit | Doug Hellmann proposed openstack-infra/release-tools: fix error handling and bug reporting in release script https://review.openstack.org/249427 | 21:21 |
| openstackgerrit | Doug Hellmann proposed openstack-infra/release-tools: make release_from_yaml.py figure out the deliverable files for us https://review.openstack.org/249428 | 21:21 |
| *** kzaitsev_mb has quit IRC | 21:21 | |
| dhellmann | ttx: next I'll refactor release_from_yaml.sh to support releasing and announcing ^^ | 21:22 |
| *** dims has quit IRC | 21:37 | |
| *** dims has joined #openstack-release | 21:38 | |
| *** gordc has quit IRC | 21:45 | |
| *** bdemers has quit IRC | 22:10 | |
| *** kzaitsev_mb has joined #openstack-release | 22:21 | |
| *** kzaitsev_mb has quit IRC | 22:26 | |
| *** mriedem has quit IRC | 22:38 | |
| *** doug-fis_ has joined #openstack-release | 22:49 | |
| *** doug-fish has quit IRC | 22:53 | |
| *** doug-fis_ has quit IRC | 22:53 | |
| *** mestery has quit IRC | 23:04 | |
| *** doug-fish has joined #openstack-release | 23:04 | |
| *** mestery has joined #openstack-release | 23:05 | |
| *** doug-fish has quit IRC | 23:08 | |
| *** dims_ has joined #openstack-release | 23:28 | |
| *** dims has quit IRC | 23:29 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!