Home
release-team@conference.openafs.org
Friday, April 12, 2019< ^ >
Room Configuration
Room Occupants

GMT+0
[13:00:44] meffie joins the room
[13:00:52] <meffie> good day
[13:00:53] wiesand joins the room
[13:00:57] <wiesand> Hello
[13:01:41] <kaduk@jabber.openafs.org/barnowl> greetings
[13:02:08] <mvita> hi
[13:03:22] <wiesand> Despite all planning, this week has been slow here again. Very sorry for that.
[13:03:44] <kaduk@jabber.openafs.org/barnowl> I know the feeling...
[13:04:06] <wiesand> Does anyone understand Andre's comment on 13544?
[13:04:50] <wiesand> I don't get the "fixes the same build error as mentioned in the commit message for 1.8.3pre1" part.
[13:05:17] <kaduk@jabber.openafs.org/barnowl> You can't build 32-bit RPMs on amd64 EL5 host
[13:05:55] meffie joins the room
[13:05:58] <wiesand> oh what a pity
[13:07:01] <mvita> looking
[13:07:21] meffie leaves the room
[13:07:34] <kaduk@jabber.openafs.org/barnowl> (OTOH, if you're still using EL5, that's kind of a pity, too...)
[13:07:55] <mvita> (ouch)
[13:08:28] <kaduk@jabber.openafs.org/barnowl> openafs.mit.edu is on EL6 and even that's getting a bit painful [learn
more later in this meeting]
[13:09:09] <meffie> (i think that was wiesand's point :)
[13:09:15] <wiesand> I'm not so sure I'd want to delay 1.8.3 for that feature…
[13:09:47] <wiesand> And I still don't understand what the "commit message for 1.8.3pre1" is
[13:10:03] <kaduk@jabber.openafs.org/barnowl> You're grouping the words wrong
[13:10:26] <kaduk@jabber.openafs.org/barnowl> (fixes (the same build error as mentioned in the commit message)) (for
1.8.3pre1)
[13:12:18] <wiesand> Ah, *now* I got it. Thanks <:-)
[13:13:12] <kaduk@jabber.openafs.org/barnowl> happy to help :)
[13:13:15] <wiesand> So, is it completely safe to include in 1.8.3 w/o a pre2 or further testing?
[13:13:52] <kaduk@jabber.openafs.org/barnowl> Well, the mechanics of how the change is done make it more risky than
just the initial description would seem.  But ISTR I did a pretty
careful review before taking it on master.
[13:14:42] <kaduk@jabber.openafs.org/barnowl> And it's the sort of thing that can be "mechanically" verified -- log
the build commands before and after the change and confirm that the
compiler invocation is the same.  repeat on all OSes you care about...
[13:16:15] <wiesand> Fine, but still… any objections to leaving the change for 1.8.4pre1?
[13:16:19] <kaduk@jabber.openafs.org/barnowl> I am not sure that I myself would put that time in, though; deferring
to 1.8.4 is the path of less resistance
[13:16:58] <wiesand> I don't like making Andrew sad, but that's my preference right now.
[13:17:48] <meffie> yes it would be nice to have in 1.8.3, it must have been and oversight to getting into the the p1.
[13:18:27] <meffie> i hate to delay 1.8.3 any longer tho.
[13:19:17] <mvita> I'm fine w/ delay until 1.8.4
[13:19:44] <meffie> maybe we can have a 1.8.4pre1 in not so long, and until then sites running el5 will need this patch for their rpmbuild.
[13:21:02] <wiesand> or whip up a VM, if they really still have 32bit EL5 systems but none that's suitable for running rpmbuild - pretty special IMO
[13:21:54] <wiesand> Ok, I put my foot down now - deferred to 1.8.4pre1.
[13:22:02] <meffie> ok
[13:22:29] <wiesand> And I really plan to get that out fairly swiftly.
[13:24:34] <wiesand> I'd like to prepare the 1.8.3 release later today. We may not get it out before the weekend then, but monday should be doable.
[13:24:44] <mvita> +1
[13:25:43] <kaduk@jabber.openafs.org/barnowl> +1 here as well
[13:27:15] <wiesand> I'd even leave the trivial ones in the queue to 1.8.4. So it's really just removing the "in progress" from NEWS and doing the versions strings change. With a few +1s on those, we should at least have tarballs for smoke testing on g.c.o. and a shot at the web page later today.
[13:28:24] <wiesand> I'll start ~30 minutes after the meeting ends.
[13:28:53] <kaduk@jabber.openafs.org/barnowl> yay
[13:29:29] <wiesand> On to other matters then? Master? Buildbot?
[13:30:27] <kaduk@jabber.openafs.org/barnowl> Most of my AFS time this week/end was doing server maintenance
[13:30:28] <meffie> thank you Ben for fixing the webproxy thing for the buildbot.
[13:30:45] <wiesand> +1
[13:30:57] <kaduk@jabber.openafs.org/barnowl> But I finished doing code review on the random websocket proxy code
from a stackoverflow post and got that deployed
[13:31:13] <meffie> the new macos builders faltered last night. going have marcio look at that.
[13:31:19] <kaduk@jabber.openafs.org/barnowl> (It actually looks quite similar to what landed in upstream apache
2.4, not very surprisingly, so the code review was not too bad)
[13:32:04] <kaduk@jabber.openafs.org/barnowl> And I also got a new version of certbot running so that we could renew
our certs; I had to get python27 installed for that to work
[13:32:29] <kaduk@jabber.openafs.org/barnowl> 4-ish days left was probably cutting it a bit close...
[13:32:45] <mvita> thank you so much Ben!
[13:32:58] <kaduk@jabber.openafs.org/barnowl> I've been trying to look at some stuff for master this morning but not
a whole lot to report from my end.
[13:33:18] <kaduk@jabber.openafs.org/barnowl> I think there's been some activity in gerrit that I haven't looked at
yet, so I don't know how interesting/urgent it is
[13:33:57] <meffie> just old stuff. i made some progress on pat's static analysis fixes.
[13:34:11] <kaduk@jabber.openafs.org/barnowl> Ah.  I feel kind of bad that we haven't gotten those in yet
[13:34:45] <meffie> and i rebased some old commits this week, for clean up things. (vnode-type-consistency)
[13:35:16] <meffie> i feel bad too. hope pat does not feel snubbed.
[13:36:36] <kaduk@jabber.openafs.org/barnowl> I don't think there's anything else from me.
Well, I will be in pacific time next week and 6am might be a stretch
[13:37:26] <meffie> also, did some cleanups for wiki.openafs, and have some clean up commits for openafs-web (no content changes, just html fixes).
[13:37:41] <meffie> i've not heard back from Chaskiel on that front.
[13:37:55] <wiesand> We can shift next week's meeting from my point of view, but I have no idea whether that would work for Mark & Mike
[13:38:28] <meffie> next friday may be a holiday (or at least 1/2 day)
[13:38:50] <mvita> depends on the shift - and yes, holiday (in the original sense)
[13:38:53] <wiesand> it will be here :)
[13:39:02] <kaduk@jabber.openafs.org/barnowl> :)
[13:40:05] <wiesand> how about +2h?
[13:40:21] <mvita> yes
[13:40:30] <meffie> oh sorry. i missed something on the 1.8.x topic...
[13:41:12] <kaduk@jabber.openafs.org/barnowl> go ahead
[13:41:15] <wiesand> ok…
[13:41:16] <meffie> i wanted to mention it looks like we hit another incident due to the bug fixed by gerrit 13517
[13:41:43] <meffie> so it would be goodness to have a 1.8.4pre1 soon.
[13:42:18] <meffie> and this incident seemed a bit more severe if i understand. mvita knows the details.
[13:42:44] <mvita> yes - it wedged a fileserver
[13:43:02] <kaduk@jabber.openafs.org/barnowl> ouch
[13:43:41] <mvita> at least with respect to accessing  a particular volume
[13:43:42] <meffie> some how the leaked connections this fixes leads to sadness.
[13:44:11] <mvita> if you leak all 4 calls on a given conn, and that thread is holding locks, then yes sadness ensues
[13:44:12] <meffie> just wanted to call that out, since it surprised me.
[13:44:43] <kaduk@jabber.openafs.org/barnowl> It is good to know about, yes
[13:44:44] <wiesand> I'll do what I can to make 1.8.4pre1 happen a.s.a.p.
[13:45:36] <meffie> thanks, just wanted to let everyone know that fix is more important than at least i thought.
[13:45:50] <wiesand> The two extra days off around the easter weekend may help…
[13:46:02] <wiesand> Understood, thanks.
[13:46:12] <kaduk@jabber.openafs.org/barnowl> *nods*
[13:46:32] <meffie> thanks, that's all i had on that topic.
[13:46:39] <wiesand> Ben, Mike, how about shifting next weeks meeting by 2 hours?
[13:47:04] <kaduk@jabber.openafs.org/barnowl> I can't guarantee anything, but that would make it more likely I'd be
able to attend
[13:47:11] <meffie> fine by me. an early start is helpful :)
[13:47:47] <kaduk@jabber.openafs.org/barnowl> Er, it would be a later start, no?
11am for you, IIUC
[13:47:54] <wiesand> We're talking about starting 2 hours *later* than usual
[13:47:58] <meffie> oh, yes!
[13:48:09] <meffie> yes, west coast, d'oh.
[13:48:24] <mvita> MOAR COFFEE
[13:48:30] <meffie> i can sleep in. yes, +2 hours is fine.
[13:49:00] <kaduk@jabber.openafs.org/barnowl> Sounds like a plan, then
[13:49:08] <wiesand> Agreed then. Thanks!
[13:49:12] <meffie> thanks.
[13:49:51] <wiesand> Snow…
[13:50:33] <wiesand> More to discuss today?
[13:51:08] <meffie> i have nothing else. have a good weekend.
[13:51:09] <kaduk@jabber.openafs.org/barnowl> not from me
[13:51:42] <wiesand> Let's adjourn then. Thanks a lot everyone!
[13:51:49] <kaduk@jabber.openafs.org/barnowl> Thanks everyone!
[13:52:00] <mvita> bye
[13:52:26] <wiesand> Looks like the weekend here will be too cold to work in the garden - should help catching up…
[13:52:42] wiesand leaves the room
[20:53:01] meffie leaves the room