[06:00:20] --- meffie has become available [06:31:09] --- Jeffrey Altman has become available [06:53:27] --- edgester has become available [06:57:51] --- wiesand has become available [07:00:57] Hello [07:01:04] mornin'. [07:01:47] Hi Ben. Who else is actually present? [07:02:20] edgester: who are you? [07:02:56] That would be Jason, I assume. [07:03:31] Ah, makes sense. [07:03:41] Anyway, we seem to be alone. [07:03:48] I'm here [07:04:16] Welcome. You're Jason? [07:04:20] yes [07:05:28] RT #131779 is for me :-( [07:06:29] I somehow thought that 131778 (previous ticket) had been reported a couple months ago. [07:06:43] (morning) [07:06:48] Wasn't that buserver? [07:06:54] Hello Mike [07:07:16] It might have been buserver, yes. [07:07:22] My memory is not very clear. [07:08:06] release notes say buserver, RT #131706, gerrit 10166 [07:08:18] Okay then. [07:09:13] May these be related? [07:10:47] Only to the extent that they are little-used and their implementations still date back to old, sloppy code, I suspect. [07:12:36] #131777: could this klogin still be in use anywhere? [07:12:52] It's not even built by default, right? [07:13:55] It looks like it's not built by default, yes. [07:14:45] Also, attempting "coordinated disclosure" is moot since the bug tracker is public. [07:15:34] Right, but he probably just doesn't know mails to bugs@openafs.org end up there. [07:15:39] None of these are showstoppers for 1.6.6. [07:15:53] hello [07:16:02] I still have to get around to ask Nicolas for more info. [07:16:11] Hello Jeff. From Nashville? [07:16:15] yes [07:17:43] I think gerrit 10472 is good to go. [07:17:44] I will reply to the klogin issue. In my opinion this is dead code that should just be deleted from the tree [07:17:58] I agree. Thanks. [07:18:13] Ok. [07:18:21] Ben, do you want to submit the patch that does so? [07:18:41] I can take a crack at it, sure. [07:19:23] gerrit 10518 and 10519 need review [07:19:29] it should be "git rm klogin.c" and remove the remaining references from the Makefile [07:19:51] Yeah, there's klogin and klogin.krb but it looks pretty simple. [07:20:10] That's not for 1.6 though? [07:20:26] I don't think there's any hurry to get it to 1.6. [07:20:30] I believe that 131778 is fixed by the same patch as 131706 [07:21:19] The very same one? (same file?) [07:21:52] I haven't read the code yet. [07:23:06] The 131706 patch only touches budb, I don't think the patch would be quite the same. [07:23:23] (I just glanced at the patch and vos.c now.) [07:27:51] --- deason has become available [07:28:11] Hi Andrew. [07:28:12] Hi Andrew [07:28:59] (klogin removal is 10522) [07:29:30] thank you [07:30:13] How about aiming for a pre2 next week? [07:35:58] Andrew, have you read the problem reports by Nicolas? [07:38:07] https://lists.openafs.org/pipermail/openafs-info/2013-November/040249.html [07:38:12] which one(s) is/are those? [07:39:50] I've seen them, but I haven't been able to look at anything from last week [07:39:52] yet [07:40:37] As said above, I still have to ask him for details. But if you'd like to... fine. [07:41:12] And there's a report against 1.6.1 in https://lists.openafs.org/pipermail/openafs-info/2013-November/040261.html [07:41:56] That's a NULL opcode in UFSOpen, right? There don't seem to be many (any?) function pointer dereferences staring out at me. [07:42:57] ecgarris's message confused me for a while because the "apache and samba sitting on top of openafs" was sort of only thrown in in passing. [07:43:48] nicolas'? no, it's just afs_get_dentry_from_fh failing, from the trace he reported [07:44:35] Ah. [07:45:45] (I guess it's clear that I don't look at linux panics very often.) [07:45:56] One of those GetDCache problems no longer ignored? [07:47:01] no [07:48:11] If it's a new bug in 1.6.6, what could be the cause? [07:50:00] something with the linux vfs integration, though I'm not sure how any recent changes could do it [07:50:16] or possibly us screwing up something during cache init [07:50:32] let's not forget it's zfs [07:50:46] but I don't really know what's going on, so I probably can't say much about it until later this week [07:52:29] There was some buildbot work you wanted to carry out after pre1? [07:54:30] that's up to jason; as far as I know, he's waiting on nothing else from me on it [07:55:51] deason: Were the custom buildbot mods with the newer version addressed? [07:57:37] I'm not sure what you mean, just gerrit integration? [07:57:44] yes [07:57:44] I sent you a patchset with the gerrit plugins [07:57:52] and with the config modifications [07:57:56] ok, I need to review that [07:58:25] if it works with the newer versions, then I'm fine with it [07:59:51] edgester: newer versions of buildbot or gerrit? [08:00:00] builtbot [08:00:53] do you have a schedule in mind? like, around thanksgiving? [08:01:26] that would be tomorrow [08:01:39] maybe around christmas [08:02:15] deason: did you get access to the Stanford box from Russ? [08:02:25] the new buildmaster? [08:03:51] no...? I had no idea IW as to ask russ for access [08:04:13] and yes, "around thanksgiving" would be this weekend; I thought since this stuff was effectively done like a month ago, we could implement it soon [08:04:40] I've dropped the ball [08:05:24] no no, timing, doesn't really matter, I guess [08:05:48] around christmas is fine; I don't think I'd be able to do anything else related to buildbot until after then anyway [08:05:49] I think that I need to coordinate with Russ [08:06:12] openafs.stanford.edu is a stanford managed system. anyone with access needs a stanford.edu Kerberos account [08:06:52] I need to request access for Andrew & Mike, and get a group/user for buildbot [08:07:31] Was there anything left for today other than buildbot discussions? [08:09:13] I don't think so. [08:09:16] Anyone? [08:09:25] nothing here [08:09:45] nope [08:10:11] Fine then. I wish you all a great thanksgiving. [08:10:28] Thanks & Bye. [08:10:33] ok, thanks [08:10:33] --- wiesand has left [08:11:44] ok, so I need to review/test Andrew's buildbot changes and request access for the Stanford box. anything else? [08:13:11] coordinate dns changes with jhutz [08:13:16] edgester: is the current buildbot config still in github? [08:13:52] yes [08:14:01] https://github.com/edgester/afsbotcfg [08:14:34] excellent, thank you [08:17:36] yup, the buildmaster config matches whats in githutb [08:21:44] would anyone object to having doubled buildbot 'verified' comments in gerrit? [08:21:59] I might split off a buildslave to test the new master [08:23:53] please don't. if you must use the production gerrit for testing, please request a test project from Simon and use that. [08:24:28] ok, I'll stand up test gerrit or request a test project [08:25:13] do the build numbers and links need to match up? in other words, do I need to sync the data from the old master before going live? [08:25:47] it would be preferable for the data to be present [08:25:56] ok [08:31:15] Is there any other buildbot stuff that needs to be discussed? [08:31:47] If not, then I wish happy holidays to everyone. [08:41:44] --- edgester has left [09:13:58] --- Jeffrey Altman has left: Disconnected [10:47:13] --- meffie has left [11:20:23] --- deason has left [13:15:27] --- Jeffrey Altman has become available