- Wednesday, April 6, 2016

release-team@conference.openafs.org

Wednesday, April 6, 2016< ^ >

Room Configuration

Room Occupants

GMT+0
[00:02:54] kadukoafs@gmail.com/barnowl15A61B68 leaves the room
[00:03:05] kadukoafs@gmail.com/barnowl15A61B68 joins the room
[06:54:37] Jeffrey Altman leaves the room
[07:09:49] Jeffrey Altman joins the room
[07:52:22] Jeffrey Altman leaves the room
[07:58:38] Jeffrey Altman joins the room
[12:57:28] meffie joins the room
[12:59:16] wiesand joins the room
[14:00:38] <wiesand> Hello
[14:02:03] <meffie> Hello
[14:02:47] <mvita> hi, sorry I won't be able to pay much attention today but I'll try to keep up
[14:03:32] <wiesand> I doubt we have much to discuss
[14:03:49] <wiesand> Mike: convince me to hit submit on 12089?
[14:03:50] <kadukoafs@gmail.com/barnowl15A61B68> I'm at IETF this week, so haven't been doing much.
[14:04:03] <meffie> ah, cool.
[14:04:41] <wiesand> Any news on 11654?
[14:05:08] <kadukoafs@gmail.com/barnowl15A61B68> I'll take another look at it now.
[14:05:25] <meffie> andrew said yesterday he would review 11654 and it's dependent gerrit
[14:05:42] <wiesand> 12206?
[14:06:15] <kadukoafs@gmail.com/barnowl15A61B68> Do you think we should wait for Andrew's review before merging?
[14:06:37] <kadukoafs@gmail.com/barnowl15A61B68> (It seems that maybe we would only want to wait before putting it in a
1.6 release, and putting it on master is less harmful.)
[14:07:34] <meffie> i need to finish my review of 12214
[14:07:43] <wiesand> If we wait for Andrew, we can just as well wait with putting it into master.
[14:08:35] <wiesand> Yes, this needs review - or to wait for the next regular release.
[14:09:00] <kadukoafs@gmail.com/barnowl15A61B68> Yeah, 12214 was fun to review [for master]
[14:09:41] <wiesand> Re there doubts whether it belongs into 1.6?
[14:09:46] <wiesand> (Are)
[14:11:31] <meffie> well, it definitely is a fix. i just want to review it again.
[14:12:37] <wiesand> That’s fine.
[14:12:37] <meffie> re 12086, i think that should be in 1.6. it looks less scary in unified diff mode.
[14:13:04] <kadukoafs@gmail.com/barnowl15A61B68> I'm happy with 12214 going into 1.6.
(But did not explicitly re-review the pullup.)
[14:13:21] <meffie> it fixes a problem where someone uses vos changeaddr and then you need to use vldb_check to fix your vldb
[14:13:55] <wiesand> Mike: 12086 or 12089?
[14:14:03] <meffie> oh, hm.
[14:14:33] <meffie> i meant 12089
[14:14:36] <meffie> (sorry)
[14:16:19] <wiesand> I don’t object. It’s just one of those where I’d feel better with another +1 - or if someone spells out to me why it’s low-risk ;-)
[14:16:31] <meffie> it's not critical, since the bug has been there forever, but it's a pretty straightforward fix, and only affects vos changeaddr -newaddr, which people should not be doing anyway.
[14:17:15] <meffie> yes, i agree, another reviewer would be good. i'll dig one up :)
[14:17:57] <meffie> (i had someone that knows the afs code base agree to help review 1.6.x patches just this morning)
[14:18:18] <wiesand> That would help!
[14:18:26] <meffie> excellent.
[14:19:24] <meffie> i'll ping him on 12089.
[14:20:13] <wiesand> So, we wait some more for review of 12089, 11654 (+12206), and the dcache-32bit-overflow stack? Another week?
[14:20:21] <meffie> yes, please.
[14:21:09] <wiesand> Good. Let’s pray that no security issue comes up anytime soon...
[14:21:15] <meffie> i'll also run some smoke tests.
[14:21:52] <wiesand> I have EL5/6/7 clients and a fileserver running with all these included. No db server though.
[14:22:01] shadow joins the room
[14:23:03] <meffie> btw, i didnt release 12166 when out in 1.6.17.
[14:23:12] <meffie> i went looking for it for 1.6.18 :)
[14:23:22] <kadukoafs@gmail.com/barnowl15A61B68> Ah, "realize"
[14:23:29] <meffie> s/release/realize/
[14:24:04] <wiesand> ah.
[14:24:16] <kadukoafs@gmail.com/barnowl15A61B68> I decided that I was comfortable doing the security release directly
on the stable branch, since the changes already there were small fixes
that even potentially could be considered security-related.
[14:24:19] <meffie> i cant words
[14:25:01] <meffie> yes, it's ok. i'm just glad it's in.
[14:25:30] <wiesand> I actually proposed doing it like this.
[14:26:30] <wiesand> That’s also why I like to wait for some "critical mass" before I start merging patches with a bit more risk or churn.
[14:26:38] <meffie> do we need bins for 1.6.17? i have rpms for rhel5/6/7.
[14:26:56] <wiesand> We have bins for EL5/6.
[14:27:09] <meffie> and i can make solaris bins too.
[14:27:11] <wiesand> And a decision not to provide bins for EL7+
[14:27:24] <wiesand> Solaris ones would make sense.
[14:27:57] <meffie> ok, i'll make dest style tarballs for solaris
[14:28:25] <wiesand> Let me know where to find them, and I’ll upload them.
[14:28:33] <meffie> will do.
[14:28:50] <wiesand> Anything else on 1.6.x today?
[14:29:27] <wiesand> Ben, anything on 1.8 then?
[14:29:42] <wiesand> ETA is still Christmas? ;-)
[14:30:49] <wiesand> Looks like we’re done for today?
[14:31:21] <meffie> i think we are in the home stretch for the externalize-log-rotation topic.
[14:31:36] <kadukoafs@gmail.com/barnowl15A61B68> Sure, Christmas, yeah.
[14:31:48] <meffie> lol
[14:32:17] <kadukoafs@gmail.com/barnowl15A61B68> I saw Mike's patch to make a configure option for the
read-while-writing ubik behavior, but it looks like I'll need to go
understand some of the previous commits in order to understand what
it's doing.
[14:32:45] <kadukoafs@gmail.com/barnowl15A61B68> And yes, we are probably in the home stretch for
externalize-log-rotation
[14:33:08] <kadukoafs@gmail.com/barnowl15A61B68> But I don't think I have anything else for today.
[14:33:13] <kadukoafs@gmail.com/barnowl15A61B68> Well, hmm.
[14:33:24] <meffie> ben i head off on rebasing after removing the sizeof(char). just say when, and i'll rebase the topic.
[14:33:33] <meffie> er held off
[14:33:41] <kadukoafs@gmail.com/barnowl15A61B68> Still need a +1 on akeyconvert.
[14:34:03] <meffie> ok. i'll ask some other people to review and test.
[14:34:21] <kadukoafs@gmail.com/barnowl15A61B68> meffie: we may be able to get away without doing a bulk rebase; if
things actually look good, I can just start merging and rebase only as
one-offs as needed.
[14:34:25] <kadukoafs@gmail.com/barnowl15A61B68> Thanks!
[14:35:06] <meffie> since i missed the leaks in akeyconvert, i'm useless.
[14:35:34] <meffie> i think ben put them there on purpose to see if i'd spot them.
[14:36:08] <kadukoafs@gmail.com/barnowl15A61B68> :(
[14:36:16] <kadukoafs@gmail.com/barnowl15A61B68> I think I was just sloppy.
[14:36:37] <kadukoafs@gmail.com/barnowl15A61B68> I think I reviewed that patch like five times, each time thinking it
was ready to merge and each time finding some new bug.
[14:37:49] <meffie> well, it works.
[14:38:38] <meffie> ok, well, have fun at ietf.
[14:38:47] <kadukoafs@gmail.com/barnowl15A61B68> Thanks!
[14:38:56] <kadukoafs@gmail.com/barnowl15A61B68> Catch you next week, if not before.
[14:39:04] <meffie> bye.
[14:39:23] <wiesand> Fine, let’s adjourn. Thanks for being here today!
[14:39:31] wiesand leaves the room
[15:22:04] shadow leaves the room
[16:07:07] shadow joins the room
[16:41:53] shadow leaves the room
[18:01:53] meffie leaves the room
[18:09:55] shadow joins the room
[18:39:57] mvita leaves the room
[20:35:03] shadow leaves the room