- Thursday, October 21, 2021

release-team@conference.openafs.org

Thursday, October 21, 2021< ^ >

kaduk@jabber.openafs.org/barnowl has set the subject to: openafs release team

Room Configuration

Room Occupants

GMT+0
[11:59:32] mbarbosa joins the room
[13:48:58] meffie joins the room
[13:56:14] Cheyenne joins the room
[15:56:48] wiesand joins the room
[16:02:40] <meffie> hello all.
[16:03:40] <kaduk@jabber.openafs.org/barnowl> greetings
[16:03:45] <wiesand> Hello
[16:03:49] <kaduk@jabber.openafs.org/barnowl> caught up in typing review comments, so missed the top of the hour :)
[16:04:08] <Cheyenne> hello all
[16:04:36] <meffie> me too ;)
[16:05:19] <wiesand> Sadly, I have no progress to report today. Major disruption (positive one, though) beginning of the week, and today I've been under the weather.
[16:05:58] <kaduk@jabber.openafs.org/barnowl> Hopefully you will get well soon
[16:06:32] <wiesand> That's likely, thanks.
[16:08:02] <kaduk@jabber.openafs.org/barnowl> I didn't get a whole lot done, either -- it was a busy week with the
IETF for me. Possibly even just the review comment on 14766 that made
me late for the meeting; I forget about earlier in the week
[16:08:05] <wiesand> 14831 looks pretty innocent, but I still have to make up my mind whether we should make that a clean cherry-pick by pulling in more outstanding changes. Sorry.
[16:10:26] <Cheyenne> I have a ps of 14831 that was rebased on top of 14825 which I think cleaned up the one minor conflict
[16:12:24] <meffie> thank you for the review comments on 14766.
[16:13:11] <wiesand> Thanks. It's entirely possible that there's no real problem - I'd just like to verify that. The sad state of 1.6.x is the result of me making the assumption that it's gonna be alright…
[16:15:14] <meffie> well, also we made changes in 1.6.x that were intentionally left out of 1.8.x, hence a divergence
[16:16:28] <wiesand> The fundamental problem is that "stable" series are dragging on for much too long
[16:16:59] <meffie> hmm, yes.
[16:17:22] <kaduk@jabber.openafs.org/barnowl> A little too stable, as it were
[16:17:33] <meffie> lol
[16:17:33] <kaduk@jabber.openafs.org/barnowl> well, maybe not exactly like that
[16:20:05] <wiesand> There's an inbalance between three variables - churn rate on master, rate of picking up changes from master into stable, and time between major stable releases.
[16:20:25] <wiesand> Of course I'm the one to blame most for that…
[16:20:49] <kaduk@jabber.openafs.org/barnowl> or am I to blame for most of that?
[16:21:50] <meffie> it's a very old code base, with lots of technical debt, so no one is to blame.
[16:23:07] <wiesand> 1.8.0 was announce 3.5 years ago. And 2.0 will take a few more years. My feeling is that avoiding more code (and even feature) skew is very important at this point.
[16:23:39] <wiesand> or we'll slow down more and more
[16:24:05] <meffie> or, accept that the stable series is a different branch
[16:24:14] <wiesand> That's why I'd like to take my time with seemingly simple stuff like 14831.
[16:24:34] <meffie> with backports from upstream branches
[16:26:02] <meffie> but yes it makes sense to take time with stable patches!
[16:26:05] <wiesand> Mike: that's an alternative view, but frankly I don't like it.
[16:26:21] <meffie> but it is the truth :)
[16:27:34] <Cheyenne> one thing is to many use the topic field better to label the type of change: bugfix, ossupport, cleanup, newfeature
[16:27:39] <wiesand> Hmm, maybe. But that's the kind of thought I'd rather ponder when I'm on top of thing again.
[16:28:43] <Cheyenne> s/many/maybe/
[16:30:02] <kaduk@jabber.openafs.org/barnowl> Oh hey, I can probably actually remove kdump today
[16:30:18] <meffie> yay!
[16:30:46] <wiesand> Nice. Should that change be conveyed to stable?
[16:30:59] <kaduk@jabber.openafs.org/barnowl> (just skimming the top of the list open changes as sorted by activity)
[16:31:06] <Cheyenne> I also have a pending commit that removes src/JAVA
[16:31:20] <Cheyenne> (sitting in github at the moment)
[16:31:27] <kaduk@jabber.openafs.org/barnowl> I'm inclined to say it's not needed on stable, but we could always
revisit if we get some merge conflicts in the future
[16:31:51] <meffie> that makes sense to me, it is a cleanup change.
[16:32:10] <meffie> and no one is using it, so could go into stable.
[16:35:06] <kaduk@jabber.openafs.org/barnowl> I'm not seeing a whole lot else at the top of
https://gerrit.openafs.org/#/q/project:openafs%20is:open%20branch:master
worth discussing today ... any other topics?
[16:35:56] <Cheyenne> still getting clean amd64 linux builds on linux-5.15-rc6+ with the 5.15 patch (both master and 1.8.x)
[16:36:03] <meffie> thanks Cheyenne
[16:36:12] <meffie> i have nothing.
[16:37:57] <meffie> btw, i released OpenAFS Ansible Collection version 1.2 on ansible galaxy and github. the new release added fedora support plus a bunch of fixes and some contributed changes from github.
[16:38:22] <meffie> some sites are using it now in production for their cells :)
[16:38:44] <kaduk@jabber.openafs.org/barnowl> Thanks for maintaining that!
[16:39:07] <meffie> i forget how to install openafs now :)
[16:39:41] <meffie> that's all i got.
[16:40:17] <kaduk@jabber.openafs.org/barnowl> Okay. Motion to adjourn?
[16:40:36] <meffie> second
[16:40:42] <Cheyenne> sounds fine
[16:40:52] <wiesand> Ok, let's adjourn. Thanks a lot folks!
[16:41:06] <meffie> have a safe day all. and a good week.
[16:41:25] <kaduk@jabber.openafs.org/barnowl> Thanks everyone, and have a good week!
[16:41:29] <kaduk@jabber.openafs.org/barnowl> Get well soon, wiesand!
[16:42:23] meffie leaves the room
[17:26:36] wiesand leaves the room
[18:48:18] meffie joins the room
[19:49:48] meffie leaves the room
[21:07:59] mbarbosa leaves the room
[21:08:04] mbarbosa joins the room
[21:52:57] mbarbosa leaves the room
[22:32:51] meffie joins the room
[22:42:56] meffie leaves the room