[00:08:34] --- mfelliott has left [00:10:38] --- mfelliott has become available [01:59:03] --- Simon Wilkinson has become available [01:59:28] Google Summer of Code has just been announced for this year ... [02:16:47] while I will admin, I will not be mentoring this Summer [02:17:19] I think we really need to find more mentors, as I think those of us who have been active in the past may find it difficult to find the time. [02:17:29] (isn't this very, very early for you?) [02:19:14] given the number of things that YFS needs to accomplish this Summer, I do not believe that any YFS staff will be taking part as a mentor. Certainly not on company time. From the perspective of the GSoC relationship it may not be a bad idea to take a year off. [02:20:01] However, if there are sufficient mentors I have no qualms about going forward with an application. [02:23:19] --- jaltman/FrogsLeap has left: Replaced by new connection [02:23:21] --- jaltman/FrogsLeap has become available [02:42:44] --- reuteras has left [02:42:44] --- reuteras has become available [02:54:14] --- reuteras has left [02:54:55] --- reuteras has become available [05:23:35] --- jaltman/FrogsLeap has left: Disconnected [05:23:45] --- jaltman/FrogsLeap has become available [05:59:51] --- reuteras has left [05:59:51] --- reuteras has become available [06:18:46] --- reuteras has left: Lost connection [06:18:46] --- mho has left: Lost connection [06:18:46] --- abo has left: Lost connection [06:19:09] --- reuteras has become available [06:19:23] --- abo has become available [06:47:29] --- reuteras has left [06:47:29] --- reuteras has become available [06:52:29] --- reuteras has left [06:55:13] --- reuteras has become available [06:59:54] --- reuteras has left [06:59:54] --- reuteras has become available [07:07:54] --- reuteras has left [07:08:27] --- reuteras has become available [07:11:58] --- deason has become available [07:18:14] --- reuteras has left [08:41:50] --- mho has become available [09:27:57] --- jaltman/FrogsLeap has left: Replaced by new connection [09:27:58] --- jaltman/FrogsLeap has become available [10:50:31] --- jaltman/FrogsLeap has left: Disconnected [10:58:02] --- jaltman/FrogsLeap has become available [13:05:51] --- mdionne has become available [13:10:03] --- jaltman/FrogsLeap has left: Disconnected [13:10:13] --- jaltman/FrogsLeap has become available [13:16:08] Pondering whether I want to try to figure out why PAGs don't currently work (FBSD) or just jump into a credential-based PAG implementation. [13:17:25] since they don't currently work, just build something good that does [13:17:55] does "credential-based" mean some structure in fbsd instead of using e.g. group ids? [13:18:14] I'm not fully convinced that the reason they don't work isn't something dumb like a misplaced #ifdef [13:22:10] deason: I think the thought of a MAC label on process credentials was floated. The current code seems to be playing games with the group list. [13:37:08] yeah, most unix fiddle with the group list; if there's a better way that's feasible, I'd just say go with that, unless maybe you want to backport fixes to older fbsd that doesn't have the capability [13:45:21] There is also the possibility of getting rwatson to put a cr_pag field into struct ucred ... but that would require version-dependence. [13:46:25] Or super-ambitiously implementing keyrings. [13:48:41] Derrick and I looked at using MAC labels to do PAGs on DARWIN, and rwatson came up with a number of reasons why it wasn't such a great idea [14:03:48] Simon: Robert did mention that he had talked with you and Derrick. I'm just going from my copy/paste of irc traffic from long ago, so details are sparse. [14:04:48] From memory the problem with MAC labels boils down to performance, and the fact that some policies don't stack well. Derrick will probably remember more of the details. [14:24:02] i assume the netbsd-style kauth thing won't work. i don't remember. [14:24:36] > deason: I think the thought of a MAC label on process credentials was > floated. The current code seems to be playing games with the group > list. i have code somewhere. i was told by apple not to use it for macos [14:45:30] rwatson says "the trade-offs in mac os x differ a bit", after "I'd be tempted to implement using MAC and see what happens" [14:46:31] i'll see if i can find the code. i think the code is usable [14:47:00] (He does appear to be on IRC now if you were still looking to chat with him.) [14:48:11] well, i'd use aim. irc is scary [15:34:56] --- deason has left [15:36:56] --- jaltman/FrogsLeap has left: Disconnected [15:54:59] --- andersk has left [15:55:24] --- deason has become available [15:59:03] --- deason has left [15:59:03] --- deason has become available [16:17:59] I haven't seen any gerrit emails in a while - just me? [16:20:42] i got some this afternoon [16:23:29] I'm only getting the merged commits (the "cvs" list), but no comments or buildbot messages [16:25:46] i got buildbot submits and comments for sure. [16:28:18] I think it's been 2-3 days since I last saw one [16:29:51] mine were today. so it's you, or the chain leading to you. sadly none of it is in my control [16:36:27] --- mdionne has left [16:48:39] --- mdionne has become available [17:01:47] nevermind... gmail apparently decided that gerrit mail is spam [17:07:49] --- rra has left: Disconnected [17:23:42] --- Russ has become available [18:29:53] --- mdionne has left [19:28:17] --- jaltman/FrogsLeap has become available [19:52:58] --- jaltman/FrogsLeap has left: Disconnected [21:49:21] --- deason has left [22:32:21] --- jaltman/FrogsLeap has become available [22:59:50] --- Russ has left: Disconnected [23:00:32] --- Russ has become available [23:27:02] --- reuteras has become available