[01:44:05] --- dev-zero@jabber.org has become available [01:46:40] --- dev-zero@jabber.org has left: offline [04:38:28] --- RedBear has left [05:04:07] --- mmeffie has become available [05:21:04] --- SecureEndpoints has left: Replaced by new connection [05:23:32] --- SecureEndpoints has become available [05:27:23] > jabber logs are the new bug tracking system? [05:27:35] no, google is the new bug tracking system. besides, cpp is not our bug [05:39:01] * Simon Wilkinson shudders. [05:39:36] Still, I guess it's no different from all my local users who bitch anywhere and everywhere, but at the people who can actually fix the problem they're complaining about. [06:02:37] s/can/can't/? [06:23:58] --- RedBear has become available [06:27:38] Simon - you upgraded what to make the empty /afs go away (re 10.5 )? [06:32:00] istr he took the rest of the system updates [07:25:21] Yup. I took another round of updates and it went away. Unfortunately, I can't remember the set of installed packages I had when it was broken. I suppose there's a log somewhere, if folk are interested to know ... [07:28:11] --- tkeiser@sinenomine.net/owl has become available [07:35:33] unless someone else reports it, i don't care. if they do, one hopes they can tell us about it. [07:36:28] Looking at the backtrace, 124129 is before my vmtruncate patch landed. [07:46:42] --- dwbotsch has left [07:47:01] --- Rrrrred has become available [09:02:40] --- reuteras has left [09:09:57] --- dev-zero@jabber.org has become available [09:10:01] --- dev-zero@jabber.org has left: offline [10:23:49] --- dwbotsch has become available [10:23:50] --- RedBear has left [11:00:05] gsoc webapp has been updated with details of who's been accepted, according to #gsoc. [11:31:19] OpenAFS has been accepted. Mentors please signup at http://socghop.appspot.com/ and sign up to mentor with OpenAFS [12:31:56] --- Russ has become available [13:02:58] --- dragos.tatulea has become available [13:03:12] hi [13:03:18] hi dragos [13:03:20] just wanted to say congrats [13:03:25] for gsoc 2009 [13:05:41] thanks [13:05:53] actually we need to thank you and jake. [13:06:30] why? [13:06:32] right, because I was so useful :-P [13:06:51] Well, you both stuck around. That's a good result. [13:06:53] because you were successful last year. you made us look good. [13:07:44] so, what are the proposals this year? [13:08:33] or it's a secret until it gets officially announced? [13:08:53] It's officially announced now. The current ideas list is on the OpenAFS website. [13:09:06] There may be some more to come ... [13:10:35] and the ones that are there will be fleshed out more. [13:10:45] \\\ [13:10:52] sorry, cat [13:11:02] and we have several students submitting ideas that they have already talked to us about and therefore are not on the ideas page [13:11:04] lol [13:11:13] "We are not trying to scare you off, but OpenAFS is possibly among the worst of all possible development environments for inexperienced developers. " [13:11:24] that's the PR! [13:11:43] hey, it draws them in like moths to a flame :-) [13:12:07] yeah [13:21:19] --- cg2v has become available [13:24:28] --- dragos.tatulea has left [13:26:52] --- dragos.tatulea has become available [13:26:52] > I suspect this is what breaks blogbench... [13:27:47] Right. I never posted. What breaks blogbench is that the _client_ doesn't sillyrename the target away when you rename over a file. it's writing to the old B/file.txt that breaks [13:57:48] Ah - so it's the rename over the existing file in that case? [13:57:55] yes [13:57:57] Not renaming a file that's already open. [13:58:07] Joy. Two rename bugs to squish. [13:58:18] Renaming over an existing open file, that is. [13:58:53] Speaking of the other "bug", where's the evidence that Rename changes fids? I'm looking at the code, and not seeing it at all. [13:59:17] I don't know. Jeff asserted that it does. I postulated on the bad things that that would cause. [14:02:25] I seem to recall that he said its when a file is renamed between directories, and is because the rename is implemented as a delete followed by an insert. But that's off the top of my head, and I haven't looked at the code yet. [14:21:58] --- dev-zero@jabber.org has become available [14:22:00] --- dev-zero@jabber.org has left: offline [17:09:54] --- dragos.tatulea has left [18:50:37] --- mmeffie has left [19:12:55] --- Rrrrred has left [21:49:54] cg2v: I have empircal evidence that the FID changes. Working on the assumption that the FID did not change I added an optimization to the windows client that patched the local copy of the directory with the new name and the old FID. This code resulted in VNOVNODE errors being generated by meredith.dementia.org file server. Checking the actual behavior showed that the FID of the file did in fact change as a side effect of RXAFS_Rename() [22:23:58] --- Russ has left: Disconnected