[05:22:39] --- meffie has become available [06:26:59] --- mvita has become available [08:15:38] --- deason has become available [08:29:55] --- mvita has left [08:29:55] --- mvita has become available [09:14:00] --- Marc Dionne has become available [09:27:05] Oops, I think I just stole some credit from Andrew. Sorry, Andrew. :( [09:27:56] I got confused by the RT log for 131591 showing how-to-rekey.txt having been uploaded by me. [09:48:25] Somewhat off-topic question for jaltman: do you know who should get credit for reporting the gss_pseudo_random counter endianness encoding issue in Heimdal? Nico says that he just did the root cause determination but he did not stumble over it himself. [09:49:23] actually i think i'm the one who stumbled on the bug [09:49:55] whether that deserves any credit is another question... :) [09:50:21] Yeah, I'm not entirely sure what the intended purpose is for the field that I'm being asked about. [09:52:22] it was causing a problem for us because keys are derived on client and server with gss_pseudo_random, and there's an assumption that the result is the same. i was testing with heimdal on one end and MIT on the other [09:52:56] Can I ask what application this was? [09:53:42] Our rxgk implementation [09:53:47] this is with our version of rxgk [09:54:36] Okay, I was mostly wondering if there were non-rxgk consumers of gss_pseudo_random that we knew about. [09:55:05] i was able to narrow it down to gss_pseudo_random giving a different result past the first set of bytes - from there Nico identified the root cause [09:55:26] Yeah, I did the same thing. [09:57:24] (But my heimdal checkout was old, so I didn't notice that Nico had fixed it until after I had spotted the bug.) [09:58:53] ah, that's unfortunate. i know it took me quite a while to narrow it down to that [10:00:04] I was, uh, rather close to the implementation, so it was pretty easy to just print out bits of the prf output and compare them. [10:04:42] my problem was going from the visible symptom to that, and getting misled for a while thinking that the keys matched because i was only printing out the first part - which did match [10:05:26] Right, the first part of the keys matching was really confusing! [10:09:17] kaduk: for what purpose do you need to determine credit? [10:10:23] I've been asked to propose the fix as an errata notice for freebsd, and the EN template has a "Credits" field. I don't have a good sense for what that really means, unfortunately. [10:13:12] looking at the freebsd.org web site not all erratas have credits. but if you are going to submit credits for this issue I would propose "Marc Dionne, Nico Williams, Your File System Inc." in that order [10:14:49] Okay, thanks. [12:11:25] --- meffie has left [12:11:26] --- meffie has become available [12:30:16] --- jaltman/FrogsLeap has left: Disconnected [12:33:54] --- jaltman/FrogsLeap has become available [14:23:33] --- meffie has left [14:33:39] --- jaltman/FrogsLeap has left: Disconnected [14:39:35] --- jaltman/FrogsLeap has become available [15:01:31] --- jaltman/FrogsLeap has left: Disconnected [15:10:28] --- jaltman/FrogsLeap has become available [15:32:44] --- deason has left [19:15:17] --- Simon Wilkinson has left [19:15:30] --- mvita has left [19:16:09] --- Simon Wilkinson has become available [19:22:41] --- mvita has become available [19:23:14] --- ballbery has left [19:27:08] --- ballbery has become available [19:58:42] --- mvita has left [21:58:14] --- Simon Wilkinson has left [22:06:33] --- shadow@gmail.com/barnowl24CEFB49 has left [23:23:45] --- jaltman/FrogsLeap has left: Replaced by new connection [23:23:46] --- jaltman/FrogsLeap has become available