[00:16:53] --- Jeffrey Altman has left: Disconnected [05:09:20] --- meffie has left [05:09:55] --- meffie has become available [05:16:15] --- Jeffrey Altman has become available [05:58:38] --- stephan.wiesand has become available [05:58:50] --- stephan.wiesand has left [06:46:19] --- stephan.wiesand has become available [06:55:33] test... [06:56:28] --- Marc Dionne has become available [06:57:32] msg received [06:57:50] pass [06:58:57] Thanks. I had to convince google once more. Same country, just a few hundred km from yesterday's location... [07:00:23] I knew my food would arrive the very moment we should start... [07:01:38] what are we having? [07:01:54] Fried charr. [07:02:34] Anyway, let's start. [07:03:23] I see the Linux d_count pullup is available now. Thanks. [07:04:02] --- shadow@gmail.com/barnowlC1401FC0 has become available [07:04:05] That's all we need for 1.6.5.1, right? [07:04:09] yeah [07:04:15] no problem. the other commits are needed when building with enable-checking here, but maybe not strictly necessary [07:04:35] --- deason has left [07:04:38] --- deason has become available [07:05:12] They look like straightforward fixes though. [07:05:24] But 1.6.5.1 should be kept minimal, I think. [07:06:04] Is 0c5afac the right point to branch? [07:08:43] Ok, I hhope Simon will sanity check when I ask him for the branch. [07:08:58] On to open bugs? [07:09:13] I think 0c5afac is correct [07:10:02] thanks. [07:10:22] Any thoughts on rt #131725? [07:13:33] Taking this as a no... [07:13:47] Any news on RT #131716? [07:14:09] sorry, I've got multiple things going on this morning. [07:14:14] let me look [07:14:58] 131716 is more clear about what it is, at least [07:15:06] re 131716... i did pullup the unrelated to ..716 aklog fix. it is on gerrit 10229 [07:15:30] I don't think it's worth holding up things for, but I can try building on netbsd and test changes [07:16:14] 10229 fixes the missing error translations on aklog. for 716, i started setting up a netbsd env to test. [07:16:24] 131725 is the AIX issue. No one from YFS has looked at it yet. I wouldn't hold a release for it. If we have a fix before 1.6.6 then it can be incorporated. It could be added as a .1 fix if it is platform specific [07:16:58] 131725; sorry, I was looking at this and hadn't seen it before today, we can work on getting some more info from that guy or the guy that reported soemthing similar on -devel [07:17:17] Ok. 1.6.6 shouldn't block on either, even though fixes like these are desirable IMO. [07:17:29] yes [07:17:31] 131716 shouldn't require netbsd to test. I wouldn't block 1.6.6 for that either. [07:18:52] Next one: getting reviews for 10165 seems to be very hard. Why? [07:19:02] I've pushed a number of commits to gerrit master which are intended to address a problem seen on Solaris in 1.6.x where the afs_CacheTruncateDaemon thread goes into an infinite loop. This is not a regression. [07:19:55] [the charr was excellent] [07:20:17] 10165 isn't the simplest thing, that's probably why [07:21:03] 10165 requires more than ten minutes to think about and most folks qualified to review it have not had many extended slots recently [07:21:04] Let's keep trying then. [07:21:28] Jeff:which ones are those? [07:22:16] topic is 'cache-truncation' [07:22:37] Stephen, are you asking who are the qualified reviewers for 10165? [07:23:22] Jeff, I was actually asking for the Solaris related changes. [07:23:32] But that's a good question too ;-) [07:24:12] Nothing new on 10179/10194. Let's see whether they make it. [07:24:54] Bringing us to the last topic I jotted down yesterday: The backlog. [07:25:08] unfortunately, 10165 is the salvager so there are very few that have extensive experience in the code. Tom and Jeff H. are qualified but not likely to be available. Derrick, Andrew, Simon. Less so myself. [07:26:14] 10165: i was worried a bit, i just went and convinced myself not to [07:26:33] Andrew, what was wrong with the list I sent? (This is actually related to 10165 since those are about salvager fixes too) [07:26:37] (i was ruminating on it and forgot to revisit the next morning) [07:27:00] The Cache Truncation patches are http://gerrit.openafs.org/#q,status:open+project:openafs+branch:master+topic:cache-truncation,n,z [07:27:20] Derrick: Thanks. [07:27:52] stephan: the "use the IH_CREATE_INIT function" commit was in there, but not the other commits in the ih_create_init topic [07:28:05] Jeff: Thanks. [07:28:18] the other commits in that topic were the ones that introduce that function; they wouldn't have appeared to touch the same files, if that's what you were going on [07:28:25] actualy numbers I need a moment to find... [07:28:28] Andrew: because according to my bookeeping they don't have to be rebased with htat stack since they touch different files. [07:28:59] 10178 needed 10175-10177, that was it [07:29:21] Does it really need them to build? [07:30:17] yes, you need "Create the IH_CREATE_INIT function" (10177) before "Use the new IH_CREATE_INIT function" (10178) :) [07:31:11] Ah, ok, sorry for that. In such cases, please feel free to add to the stack. [07:32:00] 10178 was in there because the files it touches overlap with 9409 and 9477. [07:32:16] The backlog is a tangled mess. [07:32:54] Another possibility is to leave out 10178 for now and rebase that one later. [07:33:13] In general, the ordering in which patches were submitted to master would be a good ordering for submitting them to 1.6.x [07:33:41] if the whole corpus are being pulled up, keeping relative order is good [07:34:47] And in general, I'd like to merge them in the order they were pulled (when feasible) [07:36:48] Shortening the backlog seems the best strategy to me at this point. In the future, we shouldn't allow it to grow that long. [07:38:22] can you have some separate topic branches, for staging the stable branch? [07:38:47] "brother, can you spare a topic branch?" [07:39:15] After the current set of rebases, I should be able to merge a lot of it (the backlog). [07:39:30] meffie: I'm not sure I understand the question. topic branches are used when patches are reviewed on a branch. you can't change topics after the patch is merged. [07:41:12] yes, i mean somehow before they are merged on to the stable branch. [07:41:12] I'm not sure topic branches would help much. [07:41:55] Using them for "topics" does help. [07:42:19] i'll retract the question, lets move on ;) [07:42:47] On to... anything else to discuss today> [07:43:04] s/\>/?/ [07:45:41] Ok. I'll drive back to my base camp. By the time I'm there it will be raining cats and dogs. Perfect time for writing minutes and merging changes. [07:45:51] good luck [07:45:52] we need the change from 10230 on the 1.6.x branch. it's my bug, i found on master, and see it is needed on 1.6.x [07:47:54] Which was merged a week ago. Ok, noted. [07:48:04] Anything else? [07:48:50] If not, thanks a lot everyone. [07:48:57] i'd like to see the compile_et makefile fixes on 1.6.x, since currently 'make clean', 'make' can fail. [07:49:07] have a good evening. [07:50:18] Mike, at least that's nothing I'd have earmarked not merge for 1.6.6. [07:50:33] So if it's there, it should go in. [07:51:07] Anything else? [07:51:53] If so, please feel free to add it here. I'll read the log later. [07:51:59] Bye everyone. [07:52:08] --- stephan.wiesand has left [07:59:39] --- Marc Dionne has left [09:10:39] --- meffie has left [09:10:40] --- meffie has become available [09:36:28] --- meffie has left [09:41:09] --- meffie has become available [09:43:08] --- meffie has left [10:18:20] --- meffie has become available [10:51:46] --- meffie has left [10:51:46] --- meffie has become available [10:52:45] --- meffie has left [10:52:46] --- meffie has become available [10:53:12] --- meffie has left [10:53:32] --- meffie has become available [10:54:10] --- meffie has left [10:54:11] --- meffie has become available [10:56:18] --- meffie has left [10:56:19] --- meffie has become available [10:57:13] --- meffie has left [10:57:14] --- meffie has become available [10:58:21] --- meffie has left [10:58:22] --- meffie has become available [14:44:17] --- meffie has left [15:51:02] --- deason has left [21:42:03] --- Jeffrey Altman has left: Disconnected [22:02:53] --- Jeffrey Altman has become available