[00:07:38] --- phalenor has become available [00:13:00] --- phalenor has left [00:23:16] --- phalenor has become available [00:26:52] --- kaj has become available [00:34:43] --- haba has become available [00:58:57] --- dev-zero@jabber.org has left [01:07:39] --- Russ has left: Disconnected [01:26:20] --- reuteras has left [01:26:58] --- reuteras has become available [02:04:12] --- kaj has left [02:05:43] --- tharidu has become available [02:07:20] --- kaj has become available [02:21:22] --- kaj has left [02:32:32] --- tharidu has left [02:42:43] --- Simon Wilkinson has become available [03:02:31] --- haba has left [03:13:22] --- tharidufernando has become available [03:14:02] --- tharidufernando has left [03:47:25] --- jaltman has left: Replaced by new connection [03:47:26] --- jaltman has become available [04:01:50] --- Simon Wilkinson has left [04:17:58] --- kaj has become available [04:24:01] --- kaj has left [04:31:08] --- kula has become available [04:32:12] --- haba has become available [04:33:39] --- kaj has become available [05:08:40] --- abo has become available [05:30:47] --- jaltman has left: Disconnected [05:31:36] --- abo has left [06:02:02] --- abo has become available [07:07:15] --- kaj has left [07:11:17] --- deason has become available [07:12:14] --- Simon Wilkinson has become available [07:12:34] --- Simon Wilkinson has left [07:12:44] --- Simon Wilkinson has become available [07:40:01] --- tharidufernando has become available [07:54:37] --- Simon Wilkinson has left [08:28:12] --- Simon Wilkinson has become available [08:35:37] --- reuteras has left [09:09:17] --- Simon Wilkinson has left [09:09:17] --- Simon Wilkinson has become available [09:10:31] --- Simon Wilkinson has left [09:10:45] --- Simon Wilkinson has become available [09:30:02] --- Simon Wilkinson has left [09:53:48] Connection /afs/athena.mit.edu timed out? [10:10:52] Athena's database servers have been renumbered fairly recently ... [10:14:28] that's... annoying [10:15:39] Well, they were physically relocated, too. Though I'm given to understand that they're now virtual machines, so "physically" has less meaning. [10:26:01] anyway I can get afs to ignore those cellservdb entries w/o restarting? [10:27:08] --- tharidufernando has left [10:29:49] I am told that there is an fs newcell invocation that will do the right thing, but I haven't had cause to try it, myself. [10:30:03] yeah... i've used that to add cells not in cellservdb before... [10:30:09] You can use 'fs newcell' to change the in-core entry to whatever you want. Client utilities will reread the CellServDB every time you run them. [10:30:54] what if I just want it to use the afsdb entries instead of what's in cellservdb (by deleltinng what's in cellservdb )? [10:33:54] I don't think there is a UI for that. [10:34:32] What's wrong with correcting the cellservdb entries? [10:35:34] I have an older cellservdb (1.4.10), and they are evidently now wrong [10:36:29] > client utilities will reread the cellservdb every time you run them So you can change the cellservdb, then. [10:36:45] well, I just took them out so that it would use afsdb instead [10:36:51] was still using the old entries [10:36:55] so, had to fs newcell the new ones [10:37:12] and, I wasn't using a client utility, I was using "ls" and "cd" and "cp" [10:37:26] That would be going through the cache manager, then. [11:01:46] --- Russ has become available [11:03:18] --- haba has left [11:37:21] --- jaltman has become available [11:49:37] --- phalenor has left [11:55:12] --- jaltman has left: Disconnected [11:59:28] --- phalenor has become available [12:22:08] --- sxw has become available [12:52:01] --- sxw has left [13:00:55] --- sxw has become available [13:08:01] --- sxw has left [13:08:29] --- meffie has become available [13:12:37] --- sxw has become available [13:50:00] --- phalenor has left [13:55:22] --- summatusmentis has left [14:00:01] --- phalenor has become available [14:28:35] --- sxw has left [14:44:50] --- mdionne has become available [14:46:37] --- sxw has become available [15:33:27] --- deason has left [15:42:14] --- Simon Wilkinson has become available [16:19:31] andersk: Are you around? [16:41:31] Hi. [16:41:57] Any chance you can get a crash dump from your system that's seeing these FlushVCache oopses? [16:42:24] Working out where the corruption lies just from the registers doesn't look like it's going to be possible. [16:43:57] Huh. Yeah, I can try. [16:44:48] That would be cool. I have some hunches, given the codepath that the problem is happening on (for a simple change, ShakeLooseVCaches has caused a massive amount of grief), but need a bit more data to back it up. [16:45:31] If you've got multiple ooopses, could you stick all of them into RT. It would be interesting to see how they differ. [16:47:33] Sure, I can add today’s oops. [16:47:43] Great, thanks. [17:17:21] --- phalenor has left [17:18:26] --- meffie has left [17:25:40] --- phalenor has become available [17:58:23] --- mdionne has left [18:08:39] what does this mean: WARNING: afs_ufswr vcp=e1125700, exOrW=0 [18:12:21] --- Russ has left: Disconnected [18:34:32] --- Russ has become available [18:34:39] --- deason has become available [19:54:40] Why is 'vos ex' attempting to open the file ./.AFSSERVER ? [19:57:48] nfs translator support [19:58:01] rmtsys facility [19:58:02] Ew. [19:58:22] But, a red herring for my actual trouble. [19:58:26] yup [20:59:27] --- abo has left [20:59:43] --- abo has become available [21:12:12] --- Born Fool has become available [22:20:51] --- Born Fool has left [22:23:19] Hm, I apparently still haven't found where volser and friends are grabbing user credentials from. Or do they look at the kerberos credentials cache directly? [22:28:32] you mean "vos" and friends? they ask the kernel [22:28:43] Yeah. [22:29:07] "but where does the kernel get them from?" [22:30:18] from the authenticating user program :) aklog for instance would be looking at the krb ccache, yeah [22:31:08] Well, aklog gets me tokens just fine, but vos fails to authenticate. And I am confused. [22:32:27] (vos also fails to find tokens that I already have from aklog) [22:35:17] it should be asking via the PGetTokens pioctl, I think, if that's what you're asking [22:35:51] Ah. Thank you. [22:35:55] vos and friends would only be using tokens already acquired from aklog &c, unless -localauth [22:56:56] --- deason has left [23:21:13] --- reuteras has become available [23:52:13] --- haba has become available