- Friday, January 22, 2021

release-team@conference.openafs.org

Friday, January 22, 2021< ^ >

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

Room Configuration

Room Occupants

GMT+0
[12:02:44] mbarbosa joins the room
[13:03:55] mbarbosa leaves the room
[13:05:52] mbarbosa joins the room
[16:24:41] mbarbosa leaves the room
[16:24:52] mbarbosa joins the room
[16:48:07] yadayada joins the room
[16:59:05] meffie joins the room
[17:00:12] <meffie> good day.
[17:00:18] <kaduk@jabber.openafs.org/barnowl> greetings
[17:00:30] wiesand joins the room
[17:00:43] <yadayada> hello
[17:01:00] <wiesand> Hi
[17:01:13] cwills joins the room
[17:01:22] <meffie> welcome yadayada.
[17:01:22] <cwills> Hi
[17:03:48] <wiesand> I hope everyone recovered from the rx/0x6000… fallout
[17:04:09] <kaduk@jabber.openafs.org/barnowl> debian stable is still, well, overly stable in its package
composition.
[17:04:19] <kaduk@jabber.openafs.org/barnowl> But my systems are happy now, at least.
[17:05:04] <meffie> thank you for the debian maintenance ben.
[17:06:50] <wiesand> Seconded.
[17:06:56] <meffie> we will be setting up tests with the clock set to the future to help catch this class of bug.
[17:07:41] <kaduk@jabber.openafs.org/barnowl> I assume we break somewhere in 2038, at least, but that sounds like
good coverage to have.
[17:07:58] <meffie> yes, 2038 looms.
[17:08:10] <wiesand> Nice idea. Though one would hope this was the last such bug before the 2038 apocalypse.
[17:08:40] <kaduk@jabber.openafs.org/barnowl> One would hope so, yes, but I think such testing is probably general
best practice.
[17:09:09] <wiesand> While it's a crying shame that SL7 is still on 1.6, it made things easy this time ;-)
[17:09:42] <meffie> agreed
[17:10:00] <wiesand> Re 1.8.8, I'm in the process of merging away the aklog changes.
[17:10:25] <yadayada> Thanks wiesand for merging aklog fixes
[17:10:44] <meffie> do you want andrew to review again before those are merged?
[17:11:04] <meffie> he was not added as a reviewer
[17:11:29] <yadayada> we can ask Andrew to review it once more, initial commits were reviewed by him
[17:12:01] <meffie> yes
[17:13:07] <wiesand> Andrew reviewed 14476 and 14490, which are merged. The other ones are much simpler/smaller, and I was willing to merge them w/o more review. But if you feel that's in orde, I'll just hold my horses.
[17:13:55] <meffie> ok
[17:14:26] <meffie> i added him as a reviewer, so i think it would not take long since he reviewed the ones on master.
[17:15:51] <wiesand> The remaining ones touch nothing except aklog.c, so keeping them in the queue will not cause any path conflict issues.
[17:16:05] <meffie> excellent. thank you.
[17:17:57] <wiesand> Loose 1.8.8 ends are FBSD, 10831, the rx/restart-hang topic Mike brought up a fortnight ago (waiting for the change numbers to pull up), Big Sur and Linux 5.11. And of course whatever elese you'll bring up.
[17:18:50] <meffie> btw, marcio is looking into the report today on -info about the openafs macos client. mark and marcio dont see the same results yet, so well reach out to the reporter for info.
[17:19:22] <meffie> yadayada, did you see that message?
[17:19:22] <kaduk@jabber.openafs.org/barnowl> > waiting for the change numbers to pull up
https://gerrit.openafs.org/#/q/topic:rx/restart-hang shows those as
merged on master
[17:19:36] <cwills> I re-submitted the Linux 5.11 patches to master with changes as per Andrew's review.  Actual fix is pretty simple.
[17:19:37] <kaduk@jabber.openafs.org/barnowl> Thanks for the heads-up; I haven't actually read that message on -info
yet.
[17:20:33] <meffie> speculation is it has to do with -dynroot settings or some such.
[17:21:38] <meffie> (sorry for speaking out of turn.)
[17:22:56] <yadayada> mike for Mac I do see for each minor update we need to create new OpenAFS version for BigSur which is not required. I saw Marcio comments will think more on it, how we can handle it, but I am in opinion that for each minor release new version should not be created.
[17:23:18] <wiesand> yes. dynroot/afsdb came to mind  - fallout of the "no mor in-kernel networking for you" change?
[17:23:30] <meffie> thanks.
[17:23:35] <kaduk@jabber.openafs.org/barnowl> I think the version numbers and release notes worked pretty well
overall with doing them on the dedicated branch for 1.8.7 and merging
into 1.8.x -- the only churn was that we had gotten rid of
LINUX_PKGREL on 1.8.x since 1.8.6, so there is a tiny bit of
resolution in the merge commit.
[17:24:58] <wiesand> While I still deeply despise those merge commits, it's not so bad this time because we had very few changes in the queue.
[17:26:31] <wiesand> 14264 was adjusted, and I think we're back to normal.
[17:27:07] <meffie> excellent news.
[17:29:37] <wiesand> Anything else to discuss regarding 1.8.8?
[17:29:39] <kaduk@jabber.openafs.org/barnowl> Is there more for 1.8.x this week?
[17:29:52] <kaduk@jabber.openafs.org/barnowl> Haha, messages crossed in flight
[17:30:11] <meffie> ben, will the change for buster be called 1.8.2-2 ?
[17:30:18] <meffie> just curious.
[17:31:09] <kaduk@jabber.openafs.org/barnowl> I think it will be more like a debXuY name suffix
[17:31:46] <kaduk@jabber.openafs.org/barnowl> (I have nothing else for 1.8.8 myself, to be clear.)
[17:32:01] <wiesand> (nor do I)
[17:32:05] <meffie> thanks
[17:32:34] <meffie> i have nothing else for 1.8.8pre1
[17:32:47] <wiesand> On to 1.9/master then I guess.
[17:33:06] <kaduk@jabber.openafs.org/barnowl> Not a huge amount to report, though I did make some progress.
[17:33:27] <kaduk@jabber.openafs.org/barnowl> I made a pass through the kstat change that (IIRC) Mark had asked
about, with only minor changes left.
[17:33:30] <wiesand> There was a request on -info to add the CID fixes to 1.9.0?
[17:33:54] <cwills> Yes there was
[17:33:57] <kaduk@jabber.openafs.org/barnowl> The CID fixes are already on master; I don't know if they thought they
wanted a 1.9.1 or something
[17:34:22] <kaduk@jabber.openafs.org/barnowl> But I'm not terribly inclined to do a 1.9.1 just for the CID fixes,
since the release came with prominent disclaimers to not actually run
it in production...
[17:35:41] <wiesand> There seems to be a misconception out there though, since 1.9.0 was called a "release candidate" in that request…
[17:35:42] <kaduk@jabber.openafs.org/barnowl> I opened the linux 5.11 changes early this morning before Cheyenne had
posted the updates, so I didn't look at them yet.
[17:35:42] <meffie> i was thinking we just issue a 1.9.1 that is the current master tip?
[17:35:54] <meffie> i thought that was the plan for the 1.9 series.
[17:36:08] <kaduk@jabber.openafs.org/barnowl> Yes, 1.9.1 will just be the tip of master at whatever point we do it.
[17:36:18] <cwills> Only change from yesterday was an updated commit message and relocating an include
[17:36:45] <kaduk@jabber.openafs.org/barnowl> Where are we with the bigsur stuff, though?
[17:37:09] <kaduk@jabber.openafs.org/barnowl> I put a -1 on one change for a minor rebase resolution thing, but
Yadav has a -1 on the middle one
[17:37:17] <kaduk@jabber.openafs.org/barnowl> How can that be resolved?
[17:37:42] <meffie> are we waiting from an update from Marcio?
[17:37:59] <kaduk@jabber.openafs.org/barnowl> That's my understanding, but I wasn't sure if we all think so.
[17:38:11] <meffie> if so,i can ping him after this meeting.
[17:38:19] <kaduk@jabber.openafs.org/barnowl> Okay, thanks.
[17:38:44] <kaduk@jabber.openafs.org/barnowl> The SOCKPROXY_ENV one is the biggest change there, and I left some
comments earlier this morning even though I didn't do a full review.
[17:39:03] <meffie> thank you ben.
[17:39:38] <kaduk@jabber.openafs.org/barnowl> I've been going through the indentation cleanups from Cheyenne during
the meeting (that lead up to the LINUX_ENV cleanup) in the background;
good stuff.
[17:40:15] <kaduk@jabber.openafs.org/barnowl> IIRC the macos bigsur and linux 5.11 are the most pressing topics.  I
still have audit named-pipe and FBSD open in tabs to get back to as
well.
[17:40:49] <kaduk@jabber.openafs.org/barnowl> Is there other stuff I should be prioritizing?
[17:41:01] <meffie> not that i know of.
[17:41:05] <meffie> thank you.
[17:41:48] <cwills> 14038 (just a minor clean up)
[17:42:51] <kaduk@jabber.openafs.org/barnowl> Ah, I see
[17:43:41] <kaduk@jabber.openafs.org/barnowl> I should be able to do that today
[17:43:46] <kaduk@jabber.openafs.org/barnowl> Other topics for today?
[17:45:20] <meffie> just to mention, cwills has been testing the improvement to the openafs ansible collection. hope to have a tag soon for yadav and others.
[17:46:08] <yadayada> sure Mike, I have couple of setups where I use OpenAFS anisble scripts.
[17:46:11] <meffie> i've added a number of "install methods" for all sort of scenarios. and we will be using ansible molecule to spin up cells for testing.
[17:46:43] <meffie> thats all.
[17:46:56] <kaduk@jabber.openafs.org/barnowl> Sounds useful; thanks!
[17:48:06] <meffie> (i've resisted making any changes to openafs code to make this easier so we can test any version ;)
[17:48:36] <kaduk@jabber.openafs.org/barnowl> Your restraint is appreciated :)
[17:48:52] <meffie> heh. thank you.
[17:48:56] <kaduk@jabber.openafs.org/barnowl> But it sounds like we may be ready for a motion to adjourn
[17:49:06] <meffie> second.
[17:49:09] <yadayada> second
[17:49:15] <wiesand> Sustained
[17:49:19] <cwills> Sounds good.  Everyone have a safe week
[17:49:22] <kaduk@jabber.openafs.org/barnowl> Thanks everyone!
[17:49:26] <yadayada> Thanks All
[17:49:28] <wiesand> Thanks everybody!
[17:50:18] cwills leaves the room
[17:50:28] wiesand leaves the room
[17:51:17] mvita2 leaves the room
[18:06:34] yadayada leaves the room
[18:10:56] cwills joins the room
[18:12:05] meffie leaves the room
[20:55:13] cwills leaves the room
[22:30:11] mbarbosa leaves the room