| *** dims has joined #openstack-relmgr-office | 00:01 | |
| *** dims has quit IRC | 00:50 | |
| *** dims has joined #openstack-relmgr-office | 00:58 | |
| *** dims has quit IRC | 03:39 | |
| *** dims has joined #openstack-relmgr-office | 04:40 | |
| *** dims_ has joined #openstack-relmgr-office | 04:41 | |
| *** dims has quit IRC | 04:45 | |
| *** dims_ has quit IRC | 04:46 | |
| *** ig0r_ has joined #openstack-relmgr-office | 05:51 | |
| *** ig0r__ has quit IRC | 05:55 | |
| *** dims has joined #openstack-relmgr-office | 07:43 | |
| *** dims has quit IRC | 07:48 | |
| ttx | Office hours start, yay! grabbing coffee | 08:06 |
|---|---|---|
| *** dims has joined #openstack-relmgr-office | 09:44 | |
| *** dims_ has joined #openstack-relmgr-office | 09:45 | |
| *** dims__ has joined #openstack-relmgr-office | 09:46 | |
| *** dims has quit IRC | 09:48 | |
| *** dims_ has quit IRC | 09:50 | |
| *** dims__ has quit IRC | 09:50 | |
| *** dims has joined #openstack-relmgr-office | 10:46 | |
| *** dims has quit IRC | 10:51 | |
| *** dims has joined #openstack-relmgr-office | 12:02 | |
| *** dims has quit IRC | 12:07 | |
| *** dims has joined #openstack-relmgr-office | 13:19 | |
| dhellmann | dims: sorry, missed your note yesterday, but it looks like a nice pile of releases went out | 13:37 |
| dhellmann | ttx: I'm probably going to have to miss the cross-project meeting today | 13:37 |
| ttx | dhellmann: sure, np | 13:38 |
| ttx | dhellmann: let me know when you have 10 minutes, I have a few topics to discuss with you | 13:38 |
| dhellmann | ttx: now is good | 13:38 |
| dhellmann | did you talk with mestery about the release ACL change for the neutron libs, or is that his initiative? | 13:38 |
| dims | dhellmann: i accidently gave a big boost to tooz, taskflow and oslo.vmware. though we did not break anything, we will have to cut 2.x for them next (instead of 1.x) | 13:39 |
| ttx | dhellmann: it is his initiative. One of the topics I wanted to discuss with you actually | 13:39 |
| dhellmann | dims: perhaps we should re-evaluate how we make the proposed breaking changes instead of automatically incrementing the version | 13:39 |
| dhellmann | ttx: ok | 13:39 |
| ttx | in particular in which release model bucket we'd like them in | 13:40 |
| ttx | All those things are release:independent | 13:40 |
| ttx | so my understanding is that there is no "final" release at the end of the cycle | 13:41 |
| dhellmann | I suppose his change implies a release:managed change in governance. Maybe we should ask for that before the ACL change, or as part of it. | 13:41 |
| dhellmann | hmm, so no stable branches? | 13:41 |
| ttx | so far we only accepted to "manage" things that followed the cycle (with miletsones or with intermediary) | 13:41 |
| ttx | maybe it's an oversight, due to the weirdness of the current tags | 13:42 |
| dhellmann | that's a good point | 13:42 |
| ttx | maybe it's not | 13:42 |
| * dhellmann thinks about whether there's any reason to have managed independent projects | 13:42 | |
| ttx | My take is we shouldn't manage things that are purely "independent" | 13:42 |
| ttx | our added value there is limited, better focus on things that have cycle adherence | 13:43 |
| dhellmann | it's probably too early for mestery to be online to talk to him directly | 13:43 |
| ttx | so I guess I should clarify the release model with him there | 13:43 |
| dhellmann | I agree, but I'm thinking about whether there are *any* exceptions -- these libs would not be exceptions | 13:44 |
| ttx | if they end up being more like libraries (cycle-with-intermediary) then we could definitely manage them | 13:44 |
| dhellmann | right | 13:44 |
| dhellmann | ok, I'll go change my vote on that patch | 13:44 |
| ttx | Next topic is connected | 13:45 |
| ttx | New release model tags @ https://review.openstack.org/#/c/198789/ | 13:45 |
| ttx | I'd like some early review on the WIP from you before I start trying to edit projects.yaml to match | 13:46 |
| ttx | since that's likely to be a painful patch to push | 13:46 |
| dhellmann | ok, I'll look at those during my office hours today | 13:46 |
| ttx | Last topic is the Tokyo CFP and if we'd propose a Liberty release management talk | 13:47 |
| dhellmann | yes, I saw the reminder for that yesterday afternoon | 13:47 |
| ttx | I think we can try, so I'll do an etherpad to collect all the things we'd like to communicate | 13:47 |
| ttx | and see if there is enough to justify a talk there | 13:47 |
| ttx | I'll pm you the link | 13:48 |
| dhellmann | yes, I was thinking about your point that there won't be any time not overlapping with the design summit, but I think we should do something more to publicize the release model changes | 13:48 |
| ttx | ideally we'd brainstorm on that befoire the end of week | 13:48 |
| dhellmann | sounds good | 13:48 |
| ttx | and come to yes/no call early next week, in time to submit | 13:48 |
| ttx | alright! that is all the urgent stuff I had :) | 13:48 |
| dhellmann | my schedule this week is a bit of a mess because of the travel, esp. in my mornings, but we can do some async brainstorming | 13:49 |
| ttx | dhellmann: mine is a mess early next week. I'll take Monday and most of Tuesday off | 13:49 |
| ttx | (Bastille day) | 13:50 |
| mestery | dhellmann ttx: Here | 13:50 |
| ttx | mestery: hi! | 13:50 |
| ttx | mestery: about the ACL patch | 13:50 |
| dhellmann | ttx: I'll be home some time tomorrow, but then drive to atlanta on thursday so I won't be online until a few hours later than usual | 13:50 |
| mestery | ttx: yes | 13:50 |
| dhellmann | I should be able to chat real-time early tomorrow if we need it | 13:50 |
| ttx | mestery: the issue there is the release model for those things in the stadium | 13:50 |
| mestery | Ok | 13:50 |
| ttx | mestery: the yaml says that they are released independently, without following the cycle at all | 13:51 |
| mestery | ttx: yes, correct | 13:51 |
| ttx | i.e. no stable branch, no "final" release etc | 13:51 |
| mestery | I've seen some of them want stable branches already | 13:51 |
| ttx | they can't have stable branches if they don't do a final cycle release though | 13:51 |
| mestery | Does "release:independent" mean NO stable branches, or that it just doesn't participate in stable branches? | 13:51 |
| ttx | that is what stable branches are created *from* | 13:51 |
| dhellmann | mestery: it means "we don't keep up with what they do, buyer beware" | 13:51 |
| mestery | Interesting | 13:52 |
| mestery | dhellmann: :) | 13:52 |
| ttx | so no cycle adherence, no stable branch | 13:52 |
| mestery | OK, we'll need to talk on this | 13:52 |
| mestery | Because osme of them want stable branches | 13:52 |
| ttx | mestery: what would it be called ? | 13:52 |
| ttx | obviously not stable/kilo since it doesn't recignize kilo | 13:52 |
| ttx | stable/0.5.0 ? | 13:52 |
| dhellmann | why wouldn't they use the same stable names as everyone else? | 13:53 |
| ttx | (that is fine, but the stable maint team and the release team will likely prefer to ignore it then) | 13:53 |
| ttx | dhellmann: because they don't do a "end of cycle release" | 13:53 |
| dhellmann | that name will actually test them against master, which is probably not what they want | 13:53 |
| ttx | so they choose to ignore the cycle completely | 13:53 |
| dhellmann | ttx: right now they don't, but don't we want them to? | 13:54 |
| mestery | FYI, on a call, going async | 13:54 |
| dhellmann | mestery: ack | 13:54 |
| ttx | dhellmann: they should want to, I agree. But apparently they don't ? | 13:54 |
| ttx | It may be a misunderstanding of what cycle-with-intermediary means | 13:55 |
| ttx | basically my point is... you can't have stable branches and ignore development cycles completely. | 13:56 |
| dhellmann | ttx: ok, I think we agree. release:managed and release:independent are mutually exclusive, so they have to pick one of the other release models | 13:56 |
| ttx | and if you don't have stable branches and ignore development cycles, I'd rather not manage you | 13:56 |
| dhellmann | and I think based on our previous discussions of stable, we don't want branches using that name created outside of the control of the stable-maint team | 13:56 |
| ttx | dhellmann: well the new release:independent is. The old one could be combined with "release:at-6mo-cycle-end | 13:56 |
| dhellmann | so we could let them reuse the bug branches if they want to backport | 13:57 |
| dhellmann | ttx: right, I'm thinking of the new one | 13:57 |
| ttx | (which again proves that the current one is unsuable :) | 13:57 |
| ttx | unusable* | 13:57 |
| ttx | dhellmann: how would bug branches be different from cutting random stable/X.Y.Z branches (like gnocchi does) ? | 13:58 |
| dhellmann | ttx: branding | 13:58 |
| dhellmann | also, technically, stable/x.y.z is tested against master because there are no other stable branches with the same name | 13:58 |
| ttx | oh, you mean we should disallow stable/X.Y.Z altogether | 13:59 |
| dhellmann | right | 13:59 |
| dhellmann | call those something else | 13:59 |
| dhellmann | "stable" is a reserved name | 13:59 |
| ttx | hmm, we'd have to see how common that is | 13:59 |
| dhellmann | just like we don't want ops and tc "tags" using the same name for different meanings... | 13:59 |
| ttx | I agree with you, I just fear that ship has sailed | 14:00 |
| ttx | especially with stackforge imports that went wild with branch naming | 14:00 |
| ttx | would we require a cleanup ? | 14:00 |
| dhellmann | ah, I didn't realize it was that widespread | 14:01 |
| ttx | I guess we'd have to look at which projects did stable/random and ask them why they did it | 14:01 |
| ttx | if it's long-lived and common, find another name for it | 14:01 |
| dhellmann | yeah | 14:01 |
| ttx | if it's short-lived and exceptional, direct them to bug/x.y.z | 14:02 |
| ttx | I'd go for backports/2.3.4 | 14:02 |
| ttx | that does not imply stability | 14:02 |
| ttx | or rules. Just the fact that it's a backport. | 14:03 |
| ttx | because you're right, calling them stable/x.y.z will break testing in unfun ways | 14:03 |
| dhellmann | so with the others, but the expectations are lower there | 14:04 |
| dhellmann | we could add support for backports/ branches, but I didn't originally intend for bug/ branches to be short-lived so I'm also ok with loosening the restrictions we ended up applying | 14:05 |
| ttx | dhellmann: I thought bug branches would live just enough for one backport, tag, and then dissappear ? | 14:05 |
| ttx | anyway, that's a parallel discussion | 14:06 |
| dhellmann | ttx: well, I didn't expect them to be deleted | 14:06 |
| dhellmann | yeah | 14:06 |
| dhellmann | I expected them to be lighter weight than stable, and to explicitly be tested against master (vs. the implicit thing we get when stable/foo doesn't match any other projects) | 14:07 |
| ttx | for the neutron stadium, they have to pick between "independent, ad-hoc backport branches, tag yourself" and "cycle-with-intermediary, stable branches, tag by RelMgt team" | 14:07 |
| ttx | ideally they would all pick the same model | 14:08 |
| dhellmann | yes, I think we should encourage (require?) the same model | 14:08 |
| ttx | both options are definitely possible though. | 14:08 |
| ttx | My understanding was that they cherished their independence and so I was surprised that they would ask us to tag. The benefit for those seems pretty limited | 14:09 |
| dhellmann | it sounds from mestery's email that he sees the second option as preferred | 14:09 |
| ttx | He may have to get back to his team and discuss the two options. | 14:09 |
| dhellmann | right, I think they may need to have more discussion | 14:10 |
| ttx | Just wanted to check if we were on the same line before answering on that review | 14:10 |
| ttx | and we are | 14:10 |
| dhellmann | yes, I think so | 14:10 |
| dhellmann | fwiw, I also assumed the teams wanted the change, but I guess that's not a safe assumption | 14:11 |
| dhellmann | ttx: the doc tools might be an example of a case where we have something release:managed, not on a cycle boundary, and without stable branches | 14:45 |
| dhellmann | ttx: https://review.openstack.org/#/c/197297/1 | 14:45 |
| ttx | I gues sthe question becomes, why is it a snowflake | 14:57 |
| dhellmann | ttx: it's not a production tool, so why would they need stable releases? maybe they do, for stable doc builds? | 15:11 |
| mestery | ttx dhellmann: I see the two options above, I' | 15:13 |
| mestery | I'll present those to the team and see what people think | 15:13 |
| mestery | thanks for the counsel here! | 15:13 |
| *** dims has quit IRC | 15:16 | |
| dhellmann | mestery: ++, let us know if there are concerns either way | 15:18 |
| mestery | dhellmann: sure, thanks for that! I think it may make sense for these releases to be done by the neutron team instead of library-release, I'll consult and let you know. | 15:18 |
| mestery | Thanks for all the help! | 15:18 |
| *** dims has joined #openstack-relmgr-office | 15:43 | |
| * dhellmann prepares for office hours | 17:01 | |
| *** david-lyle has quit IRC | 17:19 | |
| *** bnemec has joined #openstack-relmgr-office | 17:20 | |
| bnemec | dhellmann: Howdy | 17:20 |
| dhellmann | bnemec: bonjour | 17:21 |
| dhellmann | bnemec: so, tripleo releases? | 17:22 |
| bnemec | dhellmann: Yeah, I think you released os-cloud-config a while back and said there was a missing permission in the launchpad project that made it not work right with the release scripts. | 17:22 |
| bnemec | I'm guessing all of the tripleo projects need that fixed. | 17:22 |
| dhellmann | checking... | 17:23 |
| dhellmann | ok, looking at https://launchpad.net/os-cloud-config I see the driver is SpamapS and the maintainer is "tripleo" | 17:23 |
| dhellmann | we need to get the release team into the permissions somewhere, let me look at another project to see how we did that | 17:24 |
| dhellmann | for os-brick, the "openstack administrators" group owns the "cinder drivers" team https://launchpad.net/~cinder-drivers | 17:25 |
| dhellmann | and cinder-drivers is listed as the drivers for os-brick | 17:25 |
| dhellmann | with openstack administrators the maintainer | 17:25 |
| dhellmann | I'm not sure which of those gives us the right permissions | 17:25 |
| dhellmann | I suspect if we change the owner of tripleo (https://launchpad.net/~tripleo) to "openstack administrators" that would help, but might be insufficient | 17:26 |
| bnemec | That might do it. oslo.concurrency just has Oslo Drivers for both: https://launchpad.net/oslo.concurrency/ | 17:26 |
| dhellmann | since spamaps is still listed as the driver, and I think the driver has permissions to create milestones | 17:26 |
| dhellmann | yeah, so I think you want "tripleo" as both driver and maintainer for all of the projects, and "openstack administrators" as owner of "tripleo" | 17:27 |
| bnemec | dhellmann: Okay, I'll see about getting that set up. Thanks. | 17:28 |
| dhellmann | bnemec: we should also add "Openstack release team" as a member of tripleo | 17:29 |
| dhellmann | bnemec: I think that's the real thing we need, but having the ownership set up right will make sure we can administer the team later | 17:29 |
| bnemec | Yeah, I actually don't have administration permission, so that's the first thing I need to resolve. :-) | 17:30 |
| dhellmann | ok :-) | 17:30 |
| *** dims has quit IRC | 18:53 | |
| *** dims has joined #openstack-relmgr-office | 18:58 | |
| *** Rockyg has joined #openstack-relmgr-office | 19:06 | |
| *** dims has quit IRC | 19:48 | |
| *** david-lyle has joined #openstack-relmgr-office | 19:50 | |
| morganfainberg | dhellmann: i think we need to release a new keystone auth (should now go to the keystoneauth1 package | 19:51 |
| morganfainberg | dhellmann: it should just be head and version should be 0.<next> | 19:52 |
| morganfainberg | lmk if we have issues. | 19:52 |
| dhellmann | morganfainberg: I'm headed into the tc meeting, so it's going to need to wait a bit | 19:53 |
| morganfainberg | Anytime today works | 19:53 |
| morganfainberg | Just ping me if we end up with any issues on publishing etf | 19:54 |
| morganfainberg | Etc* | 19:54 |
| morganfainberg | and/or when youre looking at it | 19:54 |
| dhellmann | morganfainberg: let me look quickly... | 19:54 |
| dhellmann | morganfainberg: http://paste.openstack.org/show/352912 | 19:54 |
| morganfainberg | Lgtm | 19:55 |
| dhellmann | so 0.3.0? | 19:55 |
| dhellmann | morganfainberg: is anything using keystoneauth yet that's going to have issues with the namespace move? | 19:55 |
| morganfainberg | I think i have everything in order for publishing keystoneauth1 from there. If it fails we'll try after meetings to resolve | 19:55 |
| morganfainberg | Nothing should use it yet | 19:55 |
| morganfainberg | If they do, they were warned in big text in the rwadme and package info | 19:56 |
| morganfainberg | This will break until we do 1.0. | 19:56 |
| morganfainberg | 0.3 is good | 19:56 |
| dhellmann | ok | 19:56 |
| dhellmann | morganfainberg: tagged | 19:58 |
| morganfainberg | Thanks. Ill watch it. | 19:59 |
| morganfainberg | If it doesnt work as expected... Ill circle up with you later today/tomorrow | 19:59 |
| morganfainberg | And we can poke it with a stick. Or something | 19:59 |
| dhellmann | morganfainberg: I'm going to be unavailable for a while this evening, but dims, lifeless, and ttx can all tag again if needed | 20:00 |
| *** wshao has joined #openstack-relmgr-office | 20:00 | |
| morganfainberg | dhellmann: ack | 20:00 |
| lifeless | \o/ | 20:08 |
| morganfainberg | dhellmann: published successfully. Yay. Will bug lifeless and others as needed if other things crop up. Thnx again. | 20:08 |
| nikhil_k | lifeless: I think cutting a release should be fine today given the patches may or may not be merged | 20:27 |
| lifeless | ok, I'll cut a release after the tc meeting | 20:37 |
| lifeless | dhellmann: could I beg a walk-through of the full tooling for that? I want to be sure I follow your process | 20:38 |
| dhellmann | lifeless: for a library the release_postprocess.sh script does the whole thing, except actually sending the email (it writes the body to a file in relnotes) | 20:46 |
| dhellmann | lifeless: I'm going to have to move locations after the tc meeting (the place I'm working is closing) but I should be able to help you when I get back online shortly | 20:47 |
| lifeless | kk | 20:49 |
| lifeless | I'll read the code and give it a spin to see | 20:49 |
| lifeless | does it do the tag too ? | 20:49 |
| lifeless | [I'll find out :))] | 20:49 |
| *** wshao has quit IRC | 21:03 | |
| *** dims has joined #openstack-relmgr-office | 21:13 | |
| *** wshao has joined #openstack-relmgr-office | 21:13 | |
| dhellmann | lifeless: I'm back online if you want to go through it -- there's a little setup you have to do for the launchpad scripting credentials | 21:23 |
| *** dims has quit IRC | 21:30 | |
| ttx | dhellmann: the oslo.concurrency 1.8.1 release announcement doesn't mention "kilo" -- is that a bug or a feature ? | 21:34 |
| dhellmann | ttx: that may have been due to how dims ran the release notes script, or it may have been a bug. I did fix a bug with the --stable flag in that script, let me see if it merged | 21:35 |
| lifeless | dhellmann: sec, arguing about client APIs and use of TLS or callbacks or more functional structures | 21:35 |
| *** dims has joined #openstack-relmgr-office | 21:35 | |
| dhellmann | ttx: https://review.openstack.org/196712 is blocked on another review, I think | 21:35 |
| *** dims has quit IRC | 21:35 | |
| *** Guest7393 has joined #openstack-relmgr-office | 21:35 | |
| dhellmann | ttx: yeah, it's at the end of a chain where you made a suggestion on the first patch and I didn't address it yet | 21:36 |
| dhellmann | lifeless: np | 21:36 |
| ttx | ack | 21:43 |
| dhellmann | ttx: do you want to merge those and let me fix that in a follow-up or was it a blocking comment? | 21:49 |
| ttx | dhellmann: looking | 21:50 |
| ttx | dhellmann: let's merge it now | 21:50 |
| *** wshao has quit IRC | 21:51 | |
| openstackgerrit | Merged openstack-infra/release-tools: Close milestones for post-version releases https://review.openstack.org/195130 | 21:54 |
| openstackgerrit | Merged openstack-infra/release-tools: Handle more errors when updating bugs https://review.openstack.org/195242 | 21:54 |
| openstackgerrit | Merged openstack-infra/release-tools: Catch errors checking for zuul status https://review.openstack.org/195319 | 21:56 |
| dhellmann | lifeless: before you tag anything, we should wait for the rest of this patch series to land ^^ | 21:57 |
| Rockyg | dhellmann: Thierry had a great idea to put the Product WG as a project under the User Committee. I'm going to abandon the original TC patch and create a new one for doing it that way | 21:58 |
| dhellmann | Rockyg: sounds good | 21:58 |
| Rockyg | I've already started the conversation on the user-committee ML and Tom Fifeld has agreed to drive it there. | 21:58 |
| Rockyg | It's probably easiest at this point to abandon the infra patch and resubmit as a new with this info, too. | 21:59 |
| Rockyg | Product group did have one question/request. | 21:59 |
| Rockyg | rather than openstack/openstack-user-stories, they hope we can just go for openstack/user-stories for the repo. | 22:00 |
| Rockyg | Your thoughts? | 22:00 |
| openstackgerrit | Merged openstack-infra/release-tools: Fix --stable flag https://review.openstack.org/196712 | 22:01 |
| dhellmann | lifeless: ok, you should be able to refresh your release-tools sandbox and get a sane version of the script | 22:02 |
| Rockyg | dhellmann: ^^ oh, and the name changed from usecases to user-stories | 22:02 |
| dhellmann | Rockyg: do we expect to have more than one group trying to write user stories? | 22:02 |
| dhellmann | with the specs we have a prefix on the name to indicate the scope | 22:03 |
| dhellmann | applying that here seems reasonable, if we might have more than one repo with them | 22:03 |
| *** wshao has joined #openstack-relmgr-office | 22:05 | |
| Rockyg | So, there are other WGs that are generating user stories. But the Product_WG wants to collect the cross wg stories. The directory structure and focus would be a bit different. We would link to the vertical's user stories | 22:05 |
| Rockyg | Or so the thinking goes. ;-) | 22:05 |
| Rockyg | but, if the plans don't pan out, we could subdir the stories and up-publish to a single point even so. | 22:07 |
| Rockyg | I find that one of the hardest parts of starting a repo is deciding on the structure, and most people don't understand the difficulties until they try to use what they have "designed". Which is why source code is relatively easy, since it's a solved problem. | 22:08 |
| dhellmann | Rockyg: if you think there will be others, then I would keep the "openstack" prefix to indicate its cross-project nature | 22:11 |
| dhellmann | lifeless: did you still want to do that run through? it's approaching dinner time here, so I'll be dropping off in a bit | 22:12 |
| Rockyg | dhellmann: lemme check how telco is doing it.... | 22:14 |
| lifeless | dhellmann: yes please | 22:15 |
| lifeless | dhellmann: I'm ready now | 22:15 |
| lifeless | dhellmann: updating ma sandbox | 22:15 |
| dhellmann | lifeless: cool, so as I said your main entry point is that release_postversion.sh script | 22:15 |
| dhellmann | it does all of the tagging and launchpad magic | 22:15 |
| lifeless | nikhil_k: what version # do you want ? | 22:16 |
| dhellmann | if you're not already set up to run scripts for launchpad, the first time you run that it will set up the credentials for you | 22:16 |
| dhellmann | I have, at times in the past, had trouble with lynx getting those credentials properly, but it worked eventually so I was able to run the scripts on my dev vm rather than directly on my laptop | 22:17 |
| dhellmann | and you obviously need gpg keys for the signature for the new tag | 22:17 |
| dhellmann | you'll need space locally to check out the git repo to make that tag, but that shouldn't be an issue for most of the projects | 22:18 |
| dhellmann | lifeless: which project are you tagging? | 22:19 |
| lifeless | glance_store | 22:19 |
| lifeless | there are non-bugfix changes | 22:19 |
| lifeless | so I'm thinking 0.7.0 | 22:19 |
| dhellmann | ok, the list_unreleased_changes.sh script is useful for finding the pending changes | 22:19 |
| dhellmann | yeah, that version # makes sense | 22:20 |
| *** wshao has quit IRC | 22:20 | |
| lifeless | ok, no reply from nikil or sigmavirus | 22:24 |
| lifeless | should I abort? | 22:24 |
| dhellmann | lifeless: I would definitely wait until someone from that team is around and active so they can deal with fallout from the release | 22:24 |
| lifeless | ack | 22:25 |
| lifeless | though, constraints. yay. | 22:25 |
| lifeless | anyhow | 22:25 |
| dhellmann | I like to get them to sign off on the list of changes, too, so they're not surprised with what's going to happen | 22:25 |
| dhellmann | that part will be explicit when we get the releases repo up and going | 22:25 |
| lifeless | I'm expecting to run ./release_postversion.sh liberty 0.7.0 glance-store | 22:25 |
| lifeless | ? | 22:26 |
| dhellmann | you need a sha in there | 22:26 |
| lifeless | ohh right | 22:26 |
| lifeless | HEAD | 22:26 |
| lifeless | but other than that looks ok? | 22:26 |
| dhellmann | yes, that's correct | 22:26 |
| dhellmann | I've started using the actual shas from the output of list_unreleased_changes.sh to be pedantic about what we agreed I should tag and what was actually tagged, but for a slow moving repo that's not as important | 22:26 |
| *** Rockyg has quit IRC | 22:57 | |
| *** sigmavirus24 has joined #openstack-relmgr-office | 23:55 | |
| sigmavirus24 | Sorry about that. Was cooking & eating dinner. A release of glance_store isn't really a priority at the moment. It should be a problem but it will definitely be less confusing if all of the drivers work on Python 3 instead of "all but 2" working | 23:56 |
| lifeless | shouldn't or should ? | 23:57 |
| sigmavirus24 | I'm hoping we can get some more cores to look at the last two patches from haypo | 23:57 |
| sigmavirus24 | *shouldn't | 23:57 |
| sigmavirus24 | Sorry | 23:57 |
| lifeless | so | 23:57 |
| lifeless | its like this. I can hit enter on the command line I have here | 23:57 |
| lifeless | and we do it | 23:58 |
| lifeless | or I can not :) | 23:58 |
| lifeless | the constraints are - if it blows up, we fix it | 23:58 |
| lifeless | at least until bed - so if you're going to be aroundish for say 90m | 23:58 |
| lifeless | then I think we're pretty good to go | 23:58 |
| sigmavirus24 | Let me look at g-r quickly for the different branches | 23:59 |
| thingee | lifeless, dhellmann looking back at the instructions here http://lists.openstack.org/pipermail/openstack-dev/2015-June/066346.html it doesn't mention where I request tags, but I would like to release 1.3.0 of python-cinderclient ed2b133d4e5ac91c7a5719f53e666e870f9fd546 | 23:59 |
| lifeless | thingee: here | 23:59 |
| lifeless | thingee: ok | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!