[00:03:55] --- reuteras has become available [01:27:20] --- lars.malinowsky has left [02:28:36] --- rra has left: Disconnected [03:07:39] --- lama has become available [03:47:48] --- pod has become available [06:43:44] --- jaltman/FrogsLeap has left: Disconnected [06:51:03] --- jhutz@jis.mit.edu/owl has left: Disconnected [06:51:25] --- jhutz@jis.mit.edu/owl has become available [07:52:10] --- deason has become available [08:10:28] Anyone familiar with the host package fancy taking a look at 129376 ? [09:18:14] --- jaltman/FrogsLeap has become available [09:54:44] --- reuteras has left [11:39:35] --- rra has become available [13:02:41] --- lama has left [13:59:20] ls: /afs/cnf.cornell.edu/shares/public/outside_users/openafs-1.6.0pre3: Permission denied [13:59:28] "public" "outside" [14:03:43] works for me. [14:04:11] and for me [14:04:28] huh. it works for me on my laptop. [14:04:36] me too [14:04:43] oh. shit, i bet johnstown's cellservdb is ancient [14:05:11] somehow i suspect this rhel6 build is missing some kmod rpms [14:13:53] general libafs question: we never seem to attempt to raise mapDV on directory modification vnops; shouldn't we be able to raise it under the same conditions as a normal file? [14:19:00] how is mapDV relevant to directories, which cannot be mapped and never have their contents stored as pages in the OS's page cache? [14:19:41] And which certainly are never written by modifying their pages and storing them to the server... [14:20:44] osi_FlushPages is called for any vnode on solaris during getattr; VNOMAP isn't set on the vnode flags, so we try to flush the pages when mapDV is out of date [14:21:07] if we just shouldn't call that for directories, that's another easy way to not have to deal with it [14:21:57] that's the right way to fix it. [14:40:18] If directories are making it into the AFS page cache, then something is going very wrong. [14:56:13] sxw: sorry about the confusion wrt the distro suffix [14:56:27] No problem. [14:56:48] We do need to try and find some kind of way of suppressing that when we build the srpm. [14:57:24] there was a commit that added %{?dist} to openafs.spec.in, that as far as i looked into it. [14:58:29] The problem is that everything before RHEL6 doesn't define dist outside of koji (the RH build environment) [16:02:39] --- deason has left [16:41:04] --- shadow@gmail.com/owlBDE40F78 has left [16:45:59] --- ktdreyer|w has become available [16:47:07] any word on the final costs of the 2011 AFS & Kerberos Workshop? [18:09:22] --- rra has left: Disconnected [18:28:01] --- mfelliott has left [18:30:49] --- Russ has become available [18:31:56] --- ktdreyer has become available [18:32:45] --- ktdreyer|w has left [18:37:42] --- ktdreyer has left [19:18:49] --- shadow@gmail.com/owl7BA38178 has become available [20:10:08] registration will be the same as last year. housing is the open question at the moment. [21:03:32] --- phalenor has left [21:03:38] --- phalenor has become available [22:17:53] --- phalenor has left [22:17:59] --- phalenor has become available [22:46:17] --- Russ has left: Disconnected