Home
release-team@conference.openafs.org
Wednesday, May 28, 2014< ^ >
Room Configuration
Room Occupants

GMT+0
[06:02:31] Jeffrey Altman leaves the room
[06:02:31] Jeffrey Altman joins the room
[12:25:00] wiesand joins the room
[12:25:21] <wiesand> test
[13:14:51] meffie joins the room
[13:44:28] kaduk joins the room
[14:00:38] <wiesand> Hello
[14:00:47] <kaduk> g'mornin'.
[14:02:22] <meffie> hello
[14:02:39] Marc Dionne joins the room
[14:02:51] <shadow@gmail.com/barnowlC1E71073> hi
[14:03:09] <wiesand> Great. Let's start.
[14:03:18] <wiesand> Marc, any Linux news ? :-)
[14:03:27] <Marc Dionne> still ok for current rc7
[14:04:01] <Marc Dionne> i won't push anything for 3.16 until the parts that need changes hit mainline
[14:04:11] <wiesand> Good, thanks.
[14:04:50] <wiesand> So, any objections to resuming merging on 1_6_x now?
[14:05:12] <kaduk> Negative objections here.
[14:05:23] deason joins the room
[14:05:41] <shadow@gmail.com/barnowlC1E71073> doit
[14:06:07] <wiesand> Ok. If we do need a 1.6.8.1 for some reason, we can still create a branch. Worked fine last time IMO.
[14:06:32] <wiesand> Now I have to confess that I haven't spent much time in gerrit lately.
[14:06:45] <wiesand> Is there anything you'd consider ready at this point?
[14:07:21] <kaduk> I have mostly been looking at master, not 1.6 :-/
[14:07:59] <wiesand> There are a few trivial ones, like 11126.
[14:08:03] <deason> 10945, hah :)
[14:08:40] <wiesand> Yeah, stuff like that.
[14:08:58] <kaduk> We could probably start pulling in the stack reduction and coverity bits.  Feel free to add me as a reviewer if you need to.
[14:09:17] <wiesand> Tomorrow's a public holiday here. Afterwards, the trivial ones should be out of the way at least.
[14:10:17] <wiesand> I guess there's general consensus on the stack reduction ones?
[14:10:38] <kaduk> I didn't look at the 1.6 pullup specifically, but the general idea is fine.
[14:11:00] gorgo joins the room
[14:11:14] <wiesand> Some more review of those would be helpful.
[14:11:26] <meffie> the 1.6.x backport is being deployed on a large number of servers this week...
[14:11:44] <meffie> for testing
[14:12:09] <gorgo> I reviewed the backport and running on a nmber of clients
[14:12:33] <meffie> (by servers, i meant clients. sorry)
[14:12:51] <gorgo> not yet deployed on any ofthe heavy reproducers though
[14:13:56] <wiesand> Fine.
[14:15:09] <deason> do we have enough fixes yet for parallel-build that it can be actually fixed on 1.6?
[14:15:14] <wiesand> Regarding the coverity stack, it would be most helpful to get review for the lowest numbered ones first.
[14:15:47] <wiesand> Unfortunately, that's not easy the way gerrit sorts, at least by default.
[14:16:12] <meffie> deason: yes
[14:16:38] <deason> a bunch of them haven't been reported on by buildbot for some reason
[14:17:13] <deason> so it may be easiest to just resubmit; unless you want to just build assume they build and see if the end of stack builds okay
[14:17:14] <meffie> well, they still need to be merged on master, then i'll submit them on 1.6
[14:17:31] <meffie> i just rebased them again, last week (?)
[14:17:37] <deason> if we could get the parallel builds in there, it seems like those would be helpful to get in early
[14:17:54] <deason> since other gerrits will be 'testing' it whenever buildbot is triggered
[14:18:37] <meffie> the tip is 10370
[14:18:43] <kaduk> I think that the venus/tests thing with a freebsd-local patch is probably not visible on linux, because IIRC the mtime on directories gets propagated down the tree.
[14:19:35] <meffie> yes, i think it would be helpful to have the parallel-make fixes, since some of the buildbots use -j4, and fail from time to time.
[14:19:49] <wiesand> Cooling problem here :-(  I may be distracted for a few minutes...
[14:20:05] <kaduk> "from time to time" feels like "every other time", for a couple of them, sometimes.
[14:20:38] <meffie> yeah. sometimes, i just submit things to gerrit on top of 10370, just so it will build :)
[14:21:27] <kaduk> Maybe some gatekeeper wants to push the button for master?
[14:23:47] <shadow@gmail.com/barnowlC1E71073> on which? 10730 and up?
[14:24:06] <kaduk> Yeah.
[14:24:14] <shadow@gmail.com/barnowlC1E71073> "wants to"
[14:24:16] <kaduk> Chas and I +1'd that stack, and buildbot verified
[14:24:41] <wiesand> False alarm.
[14:27:46] <kaduk> Yay
[14:28:11] <meffie> yay
[14:28:34] <wiesand> One of the older ones is 10984 (raise free space reporting).
[14:28:46] <wiesand> Any objections to having this in a stable release?
[14:29:11] <kaduk> Did I not +1 that?  I thought I looked at it and it was fine.
[14:29:32] <wiesand> Not on 1.6.x. Noone.
[14:36:14] <kaduk> And now a flurry of activity.
[14:36:34] <shadow@gmail.com/barnowlC1E71073> i'd like to give you 11125 but we need to decide what to do about
11124
[14:36:36] <wiesand> Yes, I'm just watching it :)
[14:39:35] <gorgo> btw 11123 needs some attention, it's a huge performance gain
[14:40:43] <wiesand> These are all earmarked for 1.6.9 if they make it.
[14:45:02] <wiesand> What's the state of "Heimdal/UNIX"? Will we want 11075?
[14:46:48] <deason> to satisfy me, I need to look at it and submit something that avoids libkrb5 calls altogether if rxkad.keytab does not exist; and call that krb5 am-i-thread-safe function if it exists
[14:46:55] <shadow@gmail.com/barnowlC1E71073> probably we'd have to redo it. but i think we don't care
[14:47:13] <deason> but if I can't do that, what's there is good enough to satisfy the [0,1] people who currently care
[14:47:35] <deason> (er, probably good enough; I haven't looked in a while)
[14:48:22] <wiesand> Ok. Can wait, but go into 1.6.9 if nothing better is available.
[14:51:49] <wiesand> Ok. Thanks for the reviews (please feel free to continue) and other info. I think I have something to go on to start clearing the backlog.
[14:52:20] <wiesand> It will take a while though.
[14:53:16] <wiesand> I'll compile a todo list for the minutes.
[14:53:26] <wiesand> Anything else to discuss today?
[14:53:38] <wiesand> BTW, thanks for the extra space on grand.
[14:54:05] <jhutz@jis.mit.edu/owl> You can thank jweiss.  We'll pass that along.
[14:54:14] <kaduk> I just pushed 11177 for the venus/tests thing
[14:54:36] <kaduk> Sorry I didn't get it done before the meeting; fixing web edits for the wiki took longer than I expected.
[14:56:10] <wiesand> Now buildbot can test it on top of the parallel-build ones.
[14:57:10] <kaduk> Hmm, except that I didn't re-fetch before submitting.  Oh well.
[15:01:51] <wiesand> Ok, if there's nothing else for today, I have to run. Thanks a lot everyone! And my apologies for not doing the due amount of homework before the meeting. I hope that to improve before the next one.
[15:02:38] <wiesand> Bye.
[15:02:39] wiesand leaves the room
[15:03:47] meffie leaves the room
[15:11:12] gorgo leaves the room
[16:02:05] Jeffrey Altman leaves the room
[16:02:14] Jeffrey Altman joins the room
[16:07:47] Marc Dionne leaves the room
[18:05:57] Jeffrey Altman leaves the room
[18:05:59] Jeffrey Altman joins the room
[18:06:30] Jeffrey Altman leaves the room
[18:06:36] Jeffrey Altman joins the room
[19:34:15] deason leaves the room
[19:34:16] deason joins the room
[20:50:00] kaduk leaves the room
[21:46:55] shadow@gmail.com/barnowlC1E71073 leaves the room
[21:47:08] shadow@gmail.com/barnowlC1E71073 joins the room
[23:01:46] kaduk joins the room
[23:05:15] kaduk leaves the room
[23:32:21] deason leaves the room