[00:36:06] --- Russ has left: Disconnected [01:00:44] --- dev-zero@jabber.org has become available [02:23:22] --- sxw has left [02:25:06] --- sxw has become available [02:42:48] --- sxw has left [03:42:16] --- SecureEndpoints has left: Replaced by new connection [03:54:56] --- dev-zero@jabber.org has left: Replaced by new connection [03:54:56] --- dev-zero@jabber.org has become available [05:52:00] --- dev-zero@jabber.org has left [07:23:30] --- dev-zero@jabber.org has become available [08:02:55] --- SecureEndpoints has become available [10:01:15] --- tkeiser@sinenomine.net/owl has become available [10:40:21] --- Russ has become available [12:14:28] --- Russ has left: Disconnected [12:33:33] --- Russ has become available [12:52:34] --- dragos.tatulea has become available [13:53:19] --- edgester has become available [15:05:20] --- edgester has left [15:21:35] hi [15:42:56] --- Simon Wilkinson has left [15:45:36] hi dragos [15:46:47] Hi summatusmentis! [15:46:50] What's uo? [15:46:53] up [15:50:09] nothing in particular, working on reading for discrete math :( [15:55:42] hmm [15:56:59] I loved discrete math. I think that was my favorite math course of all the classes I took at Stanford. [15:57:45] --- Simon Wilkinson has become available [15:58:14] it's early, we'll see hw it goes [15:58:16] how* [15:58:25] good luck [15:58:28] thanks :) [15:58:41] Well, I can now disconnect, untar OpenAFS, build it entirely, and successfully reconnect. At this point, I think I'm going to sleep. [15:58:59] dragos: Just noticed you scroll past their. [15:59:08] simon: awesome! [15:59:27] dragos: technically it's "Foundations of Computer Science", but essentially it's discrete math [15:59:31] So soon? [15:59:31] lalala [15:59:50] If you're bored, and want to do some code review, could you take a look at some of the code hanging off bug 124128 in RT? [16:01:44] Simon Wilkinson: Are you asking me? [16:02:15] oh, crap. Simon, I forgot about your AIX testing, I'll get that done tomorrow or tuesday [16:02:20] dragos: Yes, Sorry. Those two sentences were meant to come together. But I can't type quickly enough. [16:02:45] Simon Wilkinson: I'll try reviewing it. [16:03:19] You know the disconnected code better than anyone. It would be good if you could just cast your eyes over them, and make sure that there aren't any glaringly false assumptions. [16:04:01] Are the patches attached to that bug? [16:04:11] Stuff that's resolved has patches which have been committed, stuff that's assigned to shadow, but open has patches that haven't gone in yet, stuff assigned to me is being actively worked on, everything else is pending. [16:04:30] That bug is a tracking bug, it has as its dependencies all of the disconnected bugs I could find. [16:04:31] Has RT gotten faster? [16:04:51] Seems to have, yes. I now just hate it in the same way as I hate our local RT here. [16:05:09] But, once you deal with its RT-ness, it's perfectly usable now. [16:05:18] (thanks jhutz!) [16:05:44] Simon Wilkinson: Okay. I will try to look over the changes this (or next, for others) week. [16:07:04] No rush, just if you're bored ... I _think_ they're all correct - at least, disconnected is now dealing with most things I'm throwing at it. The major outstanding bug is with hardlinks - the way we handle the FID=> name conversion breaks down when there are hardlinks in a directory. [16:07:29] argh [16:08:29] I think that there are ways to deal with it by looking at the type of operation you're trying to do, and the contents of the 'real' and 'shadow' directories. But I need to convince my self that the rules I'm building up don't leave uncaught edge cases. [16:09:56] In the mean time, I might just write some code that makes the directory operations remove and rename refuse to work with files with a link count >1 [16:10:06] (whilst disconnected, that is) [16:15:22] Simon Wilkinson: What is the openafs-discon-for-dragos.patch? [16:17:42] Where's that? [16:20:03] Ah. In my public directory. It's old - I think it's the patch I sent you to disable CStatd checks on vnodes last year (when we were going that way). What we're doing now is that we're requiring CStatd to be set to access files whilst disconnected. [16:20:57] What I think we'll do once pinning's there, is that we'll add a CStatd flag to every pinned vnode when we go disconnected (as not all pinned vnodes will be able to be Statd at the same time). [16:21:35] With extended callbacks we'll be able to tell better whether we've lost a callback because the server won't let us have that many, or if something has really changed. But for now, I think converting pinned -> statd is the best we can do. [16:22:11] let me pin that down [16:22:22] Hey, I should make a wiki page for pinning. [16:22:40] One more step to keep me away from coding :P [16:23:18] The thing is I have some notes collected from the logs on pinning and some conclusions. [16:23:49] Anything contradictory? [16:24:25] erm no [16:25:17] I'm going to sleep now. [16:25:20] Bye. [16:25:26] sleep well [16:25:40] --- dragos.tatulea has left [16:52:19] I did nothing [19:46:33] --- dev-zero@jabber.org has left [20:13:30] --- Russ has left: Disconnected [21:31:01] --- Russ has become available [22:04:43] --- SecureEndpoints has left: Disconnected [23:04:56] --- Simon Wilkinson has left [23:16:33] --- reuteras has become available [23:38:40] --- Simon Wilkinson has become available