[00:06:00] --- Derrick Brashear has become available [00:15:49] --- Derrick Brashear has left [00:32:45] --- dev-zero@jabber.org has left [00:45:43] --- Derrick Brashear has become available [00:55:53] --- Derrick Brashear has left [01:25:47] --- Derrick Brashear has become available [01:35:25] --- Derrick Brashear has left [01:40:33] --- Derrick Brashear has become available [02:26:00] --- dev-zero@jabber.org has become available [05:59:20] --- mvitale has become available [06:03:22] --- Brandon Allbery has become available [07:16:56] --- dev-zero@jabber.org has left [07:33:16] --- Derrick Brashear has left [07:33:21] --- Derrick Brashear has become available [07:40:12] Anyone available to moderate afs3-stds? I may want to resend without the attachment... [07:40:35] I wonder if I can. Hang on. [07:43:18] i know i can. [07:43:59] I have [07:44:22] Thanks! [07:44:31] well. me too. was there anything left? i approved one and killed 3. [07:45:21] I approved Bens, and defered the others. [07:48:05] --- deason has become available [08:10:35] --- ktdreyer has become available [08:13:55] cudave: it looks like Gerrit is not routing your commit to the proper change number [08:14:25] even though it did the initial commit correctly to a new change number [08:14:40] right, I wonder if something happened during that rebase [08:14:55] --- kula has left [08:15:08] --- kula has become available [08:15:16] I think you might have needed to pull first before rebasing? At any rate, can you paste the log of your local commit there, and we can see what change-ID you have in there? [08:15:16] The usual reason for this happening is that the Change-Id has been removed from the commit message, or is no longer in the final paragraph of the message [08:15:21] right [08:15:49] git log shows the change-id being there [08:15:59] oh, I think I know what's happening... your local client doesn't see that 8284 has been merged upstream [08:16:09] so you need to fetch first [08:16:13] git pull --rebase [08:16:23] wasn't a rebase supposed to resync me? [08:16:41] "rebase" is a local operation... try pulling as jaltman recommended [08:16:50] kaduk: It looks like I entirely missed Tom's conference call log on rxgk, but caught Andrews on rxgk-afs [08:17:39] Simon: I suspected something like that. At least we see it now... [08:18:13] I haven't looked very much at the rxgk-afs discussions, yet :) [08:19:04] looks like the new git pull --rebase worked [08:19:31] cool, submitting now [08:20:06] The main challenge with rxgk-afs is callback channel protection [08:20:30] That's also the only place where any of this will actually change the implemented protocol [08:22:12] Yup, that will be an interesting change. [08:23:22] I'm trying to compile a list of all of the rxgk precursors that are already in OpenAFS master - so at least you'll have something to start from. [08:23:45] Essentially, up until my paid work on rxgk ending, everything that didn't require standardisation went into OpenAFS. [08:24:27] That will be quite helpful, thanks you. [08:25:15] I hope this morning's mail is not too much of an inundation for you. I tried to make tiny commits with easily reviewable changes, at least... [08:25:37] The most time consuming stuff isn't actually going to write the security class - it's stuff like having to remove LWP from anything that you want to use rxgk with. [08:26:45] Yeah, the security class is pretty self-contained, but integrating it will be exciting. [08:27:29] If OpenAFS still cares about things like the backup suite, then interesting isn't the half of it :) [08:29:00] --- meffie has become available [08:32:11] kaduk@mit.edu/barnowl: are you still missing the mail between dec 2010 and feb 2012? [08:35:24] meffie: looks like they're up on the lists.openafs.org/pipermail archive, now. [08:36:25] Also, deason sent me an mbox extracted from gmane, so I should be all set. [08:36:39] ok, i grabbed the mbox file a few weeks ago, and i see them. [08:36:55] I also forgot about the mbox on lists.openafs.org... iirc those are more likely to contain everything [08:37:28] yes. [08:40:02] so, if somone wants to release the website for the updated newsletter index page... [08:40:14] --- Derrick Brashear has left [08:40:20] --- Derrick Brashear has become available [08:40:23] I think that someone has to be Derrick, who is travelling at the moment. [08:40:36] a backup for that would be useful [08:41:04] chaskiel can do it. chaskiel can tell you who else can do it. i have no idea who can do it. i did it [08:41:08] rxgk also depends on the pts extended names; I see that that document is final, but is there an actual implementation in the tree? [08:41:32] there isn't. [08:41:57] it's not final, it's "experimental" [08:42:03] implementing pts extended names requires multiple ubik database support [08:43:07] Oh, right. "Approved by the standards group" [08:43:17] Which reminds me, we need to update that documentation based on implementation experience. [08:43:52] Ben: when you get there, if I haven't done so, feel free to ping me. [08:45:39] > there afs3-rxgk-afs ? [08:46:24] Or, if I look at jabber and not mail, multiple ubik database support. [08:46:33] at the pittsburgh hackathon several approaches to implementing multi-db ubik were analyzed and all of them failed the atomicity requirement. Hartmut has an implementation in his OSD patchset but it has not been properly analyzed to determine if it is correct. [08:46:37] No, if you get to implementing pts extended names before Derrick or I have a change to send an update to the spec based on YFS's implementation experience. [08:47:59] > implementing pts extended names requires multiple ubik database support not if you don't implement the optional RPCs to actually configure the mappings, and just rely on implicit mappings [08:49:25] --- Simon Wilkinson has left [08:49:25] --- Simon Wilkinson has become available [08:51:13] I thought Gerrit was supposed to run a remctl post-commit hook to "openafs-backend"? [08:51:54] --- dev-zero@jabber.org has become available [08:52:18] I don't think even using explicit mappings _requires_ multiple database support - it's just that the amount of work goes up hugely if you don't have them (you end up having to redesign the whole ptserver database) [08:52:47] Departmental fileservers has the same problem. If you want to implement them without multiple databases, you end up having to redesign the vldb format. [08:54:57] Ben: I was referring to the ptauthname stuff. You have to make some assumptions that aren't contained in the document. In particular, you have to assume that entities will always have an AFS "name" by which they can be manipulated by the legacy RPCs. [08:57:04] Ah. Presumably the form of that "name" should be generated in a standardized fashion? [08:59:06] At the moment, you have to provide it when you create the entity [09:09:56] ktdreyer: The openafs-web repo does have a post receive hook that calls /usr/local/bin/export_htdocs [09:12:13] cool, so I'm still a little confused about that [09:12:32] export_htdocs runs on both www-01.openafs.org, and on openafs.stanford.edu? [09:13:42] I wonder if there are two slightly different versions of export_htdocs out there [09:20:41] Russ is probably the person to ask about all of this [09:20:58] ok, thanks [09:31:20] Russ is on vacation for a couple of weeks but is responding to e-mail. [10:50:50] Derrick - thank you [13:10:30] --- meffie has left [13:24:28] --- dev-zero@jabber.org has left: offline [13:40:29] --- meffie has become available [14:00:39] --- Brandon Allbery has left [14:25:03] --- meffie has left [14:52:58] --- dev-zero@jabber.org has become available [15:10:37] --- dev-zero@jabber.org has left [15:10:43] --- mvitale has left [15:49:53] --- deason has left [16:31:45] --- jaltman/FrogsLeap has left: Disconnected [16:37:31] --- mdionne has become available [16:38:32] --- jaltman/FrogsLeap has become available [17:44:10] --- shadow@gmail.com/barnowl41322518 has left [17:44:10] --- Derrick Brashear has left [17:47:19] --- Derrick Brashear has become available [17:48:17] --- shadow@gmail.com/barnowl41322518 has become available [22:04:11] --- dev-zero@jabber.org has become available [22:04:58] --- dev-zero@jabber.org has left [22:04:59] --- dev-zero@jabber.org has become available [23:42:54] --- Derrick Brashear has left [23:53:31] --- dev-zero@jabber.org has left