[00:13:37] --- Simon Wilkinson has left [00:22:34] --- Russ has left: Disconnected [01:00:29] --- Simon Wilkinson has become available [01:25:55] --- kaj has become available [01:35:14] --- rod has left: Replaced by new connection [05:53:41] --- rod has become available [06:04:56] --- reuteras has left [06:58:51] --- jaltman has left: Disconnected [07:12:49] --- kaj has left [07:12:49] --- kaj has become available [07:26:28] --- matt has become available [08:03:28] --- Jeffrey Altman has left: Replaced by new connection [08:03:28] --- Jeffrey Altman has become available [08:04:09] --- meffie has left [08:07:19] --- jaltman has become available [08:09:26] --- jaltman has left: Replaced by new connection [08:09:27] --- jaltman has become available [08:11:56] --- meffie has become available [08:13:18] --- cudave has left [08:13:56] --- dwbotsch has become available [08:46:45] looks like at the moment we're boned on 2.6.18-164.6.1.el5; no linux/exportfs.h. struct export_operations in linux/fs.h. the results of the exportfs.h existance test are not carried into the NEW_EXPORT_OPS test [08:49:46] oh. worse than that then. no fh_to_parent, but no find_exported_dentry [08:50:30] 2.6.18-164.6.1 - isn't that kinda old anyway? [08:50:36] rhel5? [08:50:52] i mean, it's not new, but... [08:51:33] or is this for 1.5.x ? [08:51:51] well, it's what's currently in 1.5.x [08:52:04] ah, i'll shutup then [09:00:11] --- rod has left: Replaced by new connection [09:05:56] --- kaj has left [09:05:58] --- kaj has become available [09:06:21] --- kaj has left [09:06:52] --- Simon Wilkinson has left [09:22:47] --- rra has become available [10:07:31] --- Simon Wilkinson has become available [10:09:56] Derrick: Have you looked into the EXPORT_OPS stuff further? [10:10:09] uh. beyond "we're boned"? [10:10:15] I don't think we are. [10:10:20] If the NFS server works, so should we. [10:10:29] neither old or new exportfs appears to be here? [10:10:46] Just find_exported_dentry is missing, right? [10:10:57] for old? no, also, uh [10:11:20] export_op_default [10:12:25] Yeah. I think that means that we just can't use particular filesystems. [10:13:45] I'd need to check the kernel source for sure, but I reckon that if you make the section of code that uses export_op_default just panic, and remove the bit that uses find_exported_dentry, you'll still work on all of the filesystems we care about. [10:13:45] Obviously, if that does work, we'll need more configure tests ... [10:14:06] i'll try it shortly. [10:14:19] currently writing documentation [10:21:39] Simon, in the Gerrit configuration when you get a chance, could you set sendemail.smtpEncryption to tls? We're requiring encryption for all mail going to our local SMTP server as of next week. [10:27:35] --- Simon Wilkinson has left [10:27:40] --- Simon Wilkinson has become available [10:30:33] --- Simon Wilkinson has left [10:30:41] --- Simon Wilkinson has become available [10:31:41] rra: I'll do it later this evening. On a connection that keeps dropping out at present ... [10:31:48] Thanks! [10:40:58] builds, loads, and i think it wiped out the machine [10:41:27] --- Simon Wilkinson has left [10:41:28] --- Simon Wilkinson has become available [10:42:59] Oh well. I guess I need to go and figure out exactly what the kernel source for that version looks like. [10:43:23] (By wiped out - you didn't get a panic or an ooops?) [10:45:15] i assume it panic'd. it's not reachable to me [10:45:35] --- Simon Wilkinson has left [10:45:39] --- Simon Wilkinson has become available [10:48:00] --- Simon Wilkinson has left [10:48:05] --- Simon Wilkinson has become available [10:48:15] Ah - not a VM. Sorry. [10:48:53] is fine. i will do next try in a VM. "duh" [10:49:14] too many prompts, just not even thining about what i am typing at [10:49:34] fwiw, dafs documentation now in gerrit [10:51:03] ah. i lie. i forgot, i need to rename all to .in [10:58:42] there. fixed to actually make the docs generable [10:58:54] confirming... [10:59:40] hm. fragments/volserver-synopsis.pod isn't being included. uh. [11:00:02] same with fileserver [11:00:21] some of the includes work. some don't. uh [11:00:52] oh. if there's not a blank line before the include it breaks. this is.... unfortunate [11:01:09] i am going to get food. i need a break [11:01:33] The blank line is the thing that Russ meant about the requirement for includes to be POD sections... [11:03:26] --- meffie has left [11:06:21] --- jaltman has left: Disconnected [11:06:28] --- jaltman has become available [11:09:10] I reckon we're being caught out by RedHat's stable-ABI thing. [11:16:54] well, uh. i don't really wanna rewrite it to duplicate more text. shit [11:42:14] --- jaltman has left: Disconnected [11:58:10] --- jaltman has become available [12:14:35] Now, I wish I'd known about this years ago : http://www.driverupdateprogram.com/ [12:14:58] --- meffie has become available [12:21:45] me either [12:23:10] If they're still doing it from RHEL6, it might be worth seeing how our symbol use compares with their whitelist. [12:30:04] In other news, my occasional employer is still seeing problems with tokens disappearing in the cache manager on heavily used machines. We've pulled up all of the 'log when tokens go away' patches from 1.5, so my current thinking is whether we're failing to lock the user structures properly whilst manipulating hash chains. [12:30:05] Yeah, POD syntax requires that POD directives be their own paragraph. I could write an include parser that doesn't require that, but it means that we give up ever getting =include upstream. [12:37:52] pod directives. uh. if someone else wants to tweak it, go nuts. i am well, i want macroes, so i can define my own name and then not care, and then sections are fine. MYNAME can be fileserver or dafileserver or whatecer [12:37:53] Looks like the new config.guess stuff breaks sysname computation. [12:38:10] On 64-bit Mac? If so, Derrick fixed that. [12:38:13] Or do we have another case? [12:38:30] config.*: it broke amd64 macos and i fixed it. what elsr? [12:38:30] Our config.sub was four years out of date, so I was worried that we might have to chase a few cases. [12:39:00] Ah, it might be that I've not got head. I was trying to verify an older patch that Jason has reported as failing, so I'm building a specific revision. [12:39:02] 'shadow' keyboard both rocks and sucks [12:39:13] wah wah [12:39:58] I'm pretty sure Jason's build failure is bogus, anyway. [12:40:09] which [12:40:27] RT #127620 [12:40:55] Basically, patch for dumptool claims to break the ptserver build. [12:41:18] i don't believe 'you' [12:41:44] "you" ? [12:42:00] it would be easy to drink myself into oblivion. trying to get to 1.6 branchpoint makes it tempting [12:42:11] him you [12:42:14] Oblivion isn't much fun. [12:42:26] Not a million miles away from 1.6 branch now, I think. [12:42:43] Everything I'd still like to get in could go in post branch point - they're all little bug fixes. [12:43:18] (I'd really like to make supergroups work on 64bit machines. But before I touch that code, I'd like to make it very clear that by modifying it, I'm not accepting any responsibility for it) [12:43:49] VSALVAGE, rhel5 exportfs, da* docs [12:44:04] ah, the klpd issue [12:44:25] klpd? [12:45:15] Whoohoo, I passed Nathan Neulinger in commit count. Jim Rees is next. :) [12:45:27] You'll never catch me :) [12:45:36] No, but you should watch out for Andrew. :) [12:45:57] (I had a little celebration each time I went up the commit count. Then realised I'm never going to get much higher) [12:45:57] jason needs to learn of git reset and make clean [12:46:05] * rra isn't ever going to beat you people who actually get paid to do development. :) [12:46:11] He should really git clean [12:46:19] git clean -x -f -d is your friend. [12:46:26] klpd: the thing you touched at cmu that became yours forever [12:46:32] And also your fiend, if you run it at the wrong time. [13:33:16] --- jaltman has left: Disconnected [13:53:30] ok, the exportfs issue isn't. i suck [15:34:51] --- Simon Wilkinson has left [15:40:07] --- matt has left [17:28:30] --- matt has become available [17:59:59] --- jaltman has become available [18:01:09] --- rra has left: Disconnected [18:19:15] --- Russ has become available [18:24:26] --- Kevin Sumner has left [20:32:40] --- jaltman has left: Disconnected [20:36:32] --- Jeffrey Altman has left: Replaced by new connection [20:36:35] --- Jeffrey Altman has become available [20:45:12] --- jaltman has become available [21:50:06] --- jaltman has left: Disconnected [21:51:53] --- jaltman has become available [22:11:52] --- Born Fool has become available [23:21:54] --- jaltman has left: Replaced by new connection [23:21:55] --- jaltman has become available [23:50:52] --- Russ has left: Disconnected