- Wednesday, February 4, 2015

release-team@conference.openafs.org

Wednesday, February 4, 2015< ^ >

Room Configuration

Room Occupants

GMT+0
[00:25:03] shadow joins the room
[00:25:41] shadow leaves the room
[03:31:53] shadow joins the room
[11:25:16] shadow@gmail.com/barnowlE5B64A04 leaves the room
[11:25:28] shadow@gmail.com/barnowlE5B64A04 joins the room
[11:33:49] shadow@gmail.com/barnowlE5B64A04 leaves the room
[11:33:58] shadow@gmail.com/barnowlE5B64A04 joins the room
[12:48:16] wiesand joins the room
[14:04:13] meffie joins the room
[14:30:26] kaduk joins the room
[14:38:24] Marc Dionne joins the room
[15:00:14] <kaduk> $timeofday
[15:01:05] <meffie> hi kaduk
[15:01:07] <wiesand> HI
[15:01:14] <meffie> hi wiesand
[15:01:35] <Marc Dionne> howdy
[15:02:11] <wiesand> Ah, Marc :)
[15:02:19] <wiesand> What are the Linux folks up to lately?
[15:02:40] <Marc Dionne> looks like 3.19 should be released this weekend
[15:03:16] <wiesand> And will we be good?
[15:03:26] <Marc Dionne> still good afaict
[15:04:12] <wiesand> Great, thanks.
[15:04:13] <kaduk> Debian finally pulled in a kernel change that broke 1.6.10, so I'm working on an update to unstable.
[15:04:42] <Marc Dionne> that d_alias change got backported to several stable releases
[15:04:46] <wiesand> Ubuntu pushed a kernel update for 12.04 (!) that needs 11659...
[15:05:14] <wiesand> It’s really about time to get out 1.6.11.
[15:05:36] <Marc Dionne> then there's that bit of breakage specific to the oracle unbreakable kernel
[15:05:57] <meffie> there's the new issue andrew found on solaris 11
[15:06:15] <meffie> data corruption on cm on zfs
[15:06:36] <wiesand> I wouldn’t want to delay 1.6.11 for the unbearable kernel.
[15:07:10] <wiesand> Is there a change on the pipeline for the solaris data corruption problem?
[15:07:23] <wiesand> haven’t checked RT lately...
[15:07:35] <meffie> i am talking about 11705
[15:07:36] <kaduk> I thought something hit gerrit for master, but could be misremembering
[15:07:51] <meffie> yes, it hit master over the weekend
[15:09:01] <Jeffrey Altman> the solaris fixes were merged on master.   they require pullups to 1.6
[15:09:26] <meffie> yes
[15:10:10] <wiesand> 11705 clashes with 11616 11694 11707
[15:10:49] <meffie> i'm just mentioning for wiesand
[15:11:08] <wiesand> So if it’s pulled up, those four should be (re)-pushed together.
[15:11:25] <wiesand> Or wait until the others are merged.
[15:11:36] <wiesand> I guess we consider this a blocker for pre2?
[15:11:51] <wiesand> Is it a “new” problem?
[15:12:20] <meffie> not a new problem, andrew found the root cause late last week.
[15:14:03] <wiesand> Looks like it’s low risk?
[15:14:18] <Jeffrey Altman> I believe it should be included in 1.6.11
[15:14:35] <shadow> i agree
[15:15:30] <wiesand> Fine, agreed.
[15:16:14] <kaduk> (I have a local patch to resolve my comments on 11694; build-testing in progress)
[15:16:39] <wiesand> Great, that would remove one of the blockers.
[15:16:57] <wiesand> How about 11669 -> 11616 ?
[15:17:15] <wiesand> (though that’s not a blocker, I think)
[15:17:19] <kaduk> I pushed the +2 button on 11669, I think.
[15:18:20] <wiesand> Will it be merged?
[15:19:49] <shadow> needs rebase along with parent
[15:20:30] <kaduk> 11710 to fix the mandatoryness of the enable-extra-iput configure option
[15:21:07] <kaduk> I will push a rebase unless you tell me not to
[15:21:23] <shadow> already did
[15:21:31] <kaduk> yay
[15:22:32] <wiesand> acinclude.m4 again... so 11710 joins the group 11705 11694 11616 11707 ... and 11700 too
[15:22:39] <kaduk> :-/
[15:23:50] <wiesand> Just more rebases. I can do that.
[15:24:05] <kaduk> slightly off-topic, what needs to be done to get the QuickStartGuide on the website rebuilt?  Is it built from master or stable?
[15:25:48] <shadow@gmail.com/barnowlE5B64A04> fiik. i knew at some point
[15:26:09] <wiesand> Any objections to 11700 for 1.6.11?
[15:27:06] <kaduk> Not from me :)
[15:27:09] <wiesand> Any objections to 11706 for 1.6.11?
[15:27:44] <kaduk> Do we think the docs on the website are served from AFS?  (Yes, yes, I should just go check.)
[15:28:04] <wiesand> And what about 11707?
[15:28:13] <meffie> no objections on 11706
[15:29:05] <Marc Dionne> i pushed 11707 because we have someone who actually ran into it.  it eventually leads to an assert
[15:29:35] <wiesand> Sounds like we want it. What’s the risk?
[15:30:53] <kaduk> 11707 is plausible, but I need to look at more context
[15:30:55] <Jeffrey Altman> The QuickStartGuide is served from grand.central.org/www/docs.openafs.org/
[15:30:56] <meffie> you risk people not getting asserts?
[15:31:39] <wiesand> I’m willing to take that one ;) But what about potential side effects?
[15:32:05] <meffie> ;) sorry, i know.
[15:32:12] <Marc Dionne> it only affects the case where you have a orrupt dir (which should not be common), and in that case the exiting code is obviously wrong
[15:33:03] <kaduk> Jeffrey: is there a machine to be remctl'd that updates that tree, or is it done by hand?
[15:33:24] <Jeffrey Altman> just copy the files in
[15:34:10] <wiesand> Marc: Thanks. Fine then.
[15:34:17] <Jeffrey Altman> at some point I wanted to create a versioned tree of the docs for each release but never got around to it
[15:35:19] <kaduk> > versioned tree
[15:35:56] <kaduk> > versioned tree
*nods*
there's a guy in #openafs who could benefit from the updated docs that include rxkad.keytab
[15:35:58] <Jeffrey Altman> a large part of it is because doing so requires building the docs for each release
[15:36:20] <Jeffrey Altman> of course, that text is different for master and 1.6 isn't it
[15:36:43] <shadow> yes
[15:38:22] <kaduk> Yeah, so I mentioned both ways.
[15:39:01] <wiesand> So for 1.6.11pre2 we want 11616 (after 11669), 11694 + pullup of 11710, 11700, 11706, 11707 ?
[15:39:49] <kaduk> seems likely
[15:39:56] <shadow> sounds good
[15:40:10] <wiesand> + pullup of 11705
[15:40:34] <wiesand> + 11709, and a NEWS update.
[15:42:23] <wiesand> We may manage to get it out Friday....
[15:44:34] <wiesand> And we’ll need two different SRPMs at least for Fedora20. Fun.
[15:44:42] <kaduk> I need to update 11710; should I try to order it after some other changes to reduce the path conflicts on acinclude.m4?  (which one(s)?)
[15:46:28] <wiesand> I have that list for 1.6.x only. I think there’s nothing else we want to pull up for 1.6.11 that would clash on master.
[15:47:12] <kaduk> Okay, thanks.
[15:49:37] <wiesand> I think we’re done discussing 1.6.11 then?
[15:50:28] <wiesand> Anything else to discuss today?
[15:51:30] <Jeffrey Altman> I'm quite distracted today
[15:51:45] <wiesand> I can handle the pullups and rebases. All I need is a merged 11669 and 11710.
[15:51:50] <kaduk> I wanted to ask Daria about 7896
[15:51:58] <wiesand> Jeff: Thanks for being here anyway.
[15:52:20] <kaduk> And maybe Jeff could look at the WINNT parts of 10784 if he has a moment, since I was not sure enough about those to +2 it.
[15:52:44] <kaduk> Oops, was in scrollback.  If Jeff's distracted, don't worry about 10784.
[15:53:57] <shadow> yeah. removed my -1 but i decline to +1. i still prefer a log of some sort. but not strongly enough to block
[15:54:11] <Jeffrey Altman> 10784 is fine.  its just removing all references to RX_CALL_BUSY which no longer exists
[15:55:34] <kaduk> Thanks.
[15:59:48] <kaduk> I don't think I'm prepared to have further discussion of 1.8 today.
"Please review things in gerrit."
[16:01:05] <wiesand> Fine. Thanks a lot for being here today despite the short notice, and for all the work.
[16:02:44] <wiesand> About to walk home now. Bye.
[16:02:56] wiesand leaves the room
[16:14:21] kaduk leaves the room
[16:15:03] meffie leaves the room
[17:25:38] kaduk joins the room
[18:07:54] shadow leaves the room
[18:08:59] shadow joins the room
[18:12:52] shadow leaves the room
[21:04:18] Marc Dionne leaves the room
[21:05:16] Marc Dionne joins the room
[21:15:55] Marc Dionne leaves the room
[23:42:20] kaduk leaves the room