[00:07:44] --- Simon Wilkinson has become available [00:33:00] --- dev-zero@jabber.org has left [00:47:08] --- dev-zero@jabber.org has become available [01:01:05] --- mdionne has become available [01:10:54] --- Simon Wilkinson has left [01:13:39] --- mdionne has left [01:16:16] --- deason has become available [01:16:36] --- Derrick Brashear has left [01:34:14] --- squinney has become available [01:46:59] from 7c048441: rx/perf.................ok [...] All tests successful. [01:53:05] --- squinney has left [01:56:00] --- meffie has become available [01:57:18] --- Derrick Brashear has become available [01:57:45] --- Simon Wilkinson has become available [02:00:13] deason: The rxperf test passes, but takes between a second and 3 minutes (depending on which way the wind is blowing, I guess) longer for each run. [02:00:19] the issue is timing, namely that the on my test hosts we went from all tests taking 10 seconds total, to rxperf taking 56 seconds on one, and 170 seconds on another [02:21:08] --- meffie has left [03:24:06] --- deason has left [04:25:42] --- jaltman has become available [04:26:03] --- jaltman/FrogsLeap has left: Disconnected [04:42:23] --- deason has become available [04:51:16] --- Derrick Brashear has left [04:56:53] --- deason has left [05:21:36] --- Derrick Brashear has become available [05:25:42] --- Derrick Brashear has left [05:55:30] --- Derrick Brashear has become available [05:59:38] --- Derrick Brashear has left [05:59:41] --- Derrick Brashear has become available [06:01:37] --- deason has become available [06:19:41] --- mvitale has become available [07:01:24] --- Brandon Allbery has left [07:02:19] --- Brandon Allbery has become available [07:12:10] --- Simon Wilkinson has left [07:13:55] --- Simon Wilkinson has become available [07:23:55] --- deason has left [07:24:20] --- Simon Wilkinson has left [07:52:56] --- Simon Wilkinson has become available [07:55:58] --- jaltman/FrogsLeap has become available [07:56:57] --- jaltman has left: Disconnected [07:56:57] --- deason has become available [08:08:09] --- Simon Wilkinson has left [08:15:41] --- abo has left [08:16:55] --- abo has become available [08:18:08] --- abo has left [08:18:43] --- abo has become available [08:41:36] --- Simon Wilkinson has become available [08:53:02] --- Derrick Brashear has left [08:53:06] --- Derrick Brashear has become available [09:01:53] --- Derrick Brashear has left [09:02:28] --- Simon Wilkinson has left [09:03:14] --- Simon Wilkinson has become available [09:03:14] --- Simon Wilkinson has left [09:03:26] --- natefoo has become available [09:04:22] deason: i managed to get another undesired address in my vldb that i can't remove. [09:06:57] --- deason has left [09:37:23] and now i apparently have two readonly copies on the same server, in the same vicep. [09:44:07] what happened here?: https://gist.github.com/3906639 [09:48:20] --- dev-zero@jabber.org has left [09:54:16] --- deason has become available [09:55:01] deason: ping. [10:15:46] --- dev-zero@jabber.org has become available [10:18:42] i used a big hammmer: remove vldb and syncvldb. [10:28:20] --- dev-zero@jabber.org has left [10:34:15] natefoo: sorry, busy for the moment [10:34:58] no worries. [10:42:35] --- deason has left [10:51:16] --- deason has become available [10:57:26] natefoo: okay, well, if you blew it all away obviously there's not much I can do about it now :) [10:57:46] but just... I see a 'vos changeaddr' in there without a -remove [10:57:49] never never do that [10:58:30] that's going to cause problems if you keep doing that; what jeff had mentioned before is adding a check to 'vos changeaddr' to _prevent_ you from doing that (technical details/exceptions elided) [10:59:07] (sorry I'm brief; if you're not aware, the european afs conference is going on now, so me and probably most others are a little distracted :) [11:15:45] --- deason has left [13:05:54] --- dev-zero@jabber.org has become available [13:06:55] --- dev-zero@jabber.org has left [13:06:56] --- dev-zero@jabber.org has become available [13:21:30] --- kula has left [13:21:34] --- kula has become available [14:47:54] --- Brandon Allbery has left [14:49:18] --- Brandon Allbery has become available [14:53:06] --- Simon Wilkinson has become available [14:53:51] --- Derrick Brashear has become available [14:54:59] --- deason has become available [15:00:39] deason: thanks yet again. [15:00:58] i wrote a little guide for anyone who finds themselves in a similar situation: http://www.bx.psu.edu/~nate/doc/vldb.html [15:01:16] enjoy the conference. =) [15:17:42] removing the sysid file and restarting is _not_ what you're supposed to do; you should just be able to restart without removing the sysid file [15:18:37] removing the sysid file and restarting means that the fileserver registers itself as a completely new fileserver, separate from the other fileserver that had all of those volumes on it.... so the entries in the vldb for those volumes now point to a different logical "server" [15:19:31] --- Brandon Allbery has left [15:19:50] that is probably not clear.... I'd need to write something a little more lengthy to really explain that; I'm not sure if that has been done before somewhere else... don't really have time now, but others should know well the issue I'm talking about [15:20:06] --- Brandon Allbery has become available [15:20:19] but the bottom line is: restart the fileserver after you have corrected the address and the netinfo/netrestrict information [15:20:40] that is the way that's supposed to work; if you find some way where that doesn't work, we need to look at it and fix it [15:23:24] and if it doesn't work, see if there are VL_RegisterAddrs errors in FileLog [15:31:04] okay. [15:31:25] i crossed some wires there, i removed sysid down in the disaster recovery. [15:34:17] updated the doc. [15:36:54] --- jaltman/FrogsLeap has left: Disconnected [15:41:57] --- jaltman/FrogsLeap has become available [16:09:28] --- mvitale has left [16:56:00] --- deason has left [17:05:33] --- Simon Wilkinson has left [17:14:36] --- Simon Wilkinson has become available [23:36:23] --- deason has become available [23:37:09] --- deason has left