[00:01:51] --- Rasmus Kaj has become available [01:21:36] --- reuteras has left [01:21:36] --- abo has left [01:21:44] --- reuteras has become available [01:21:44] --- abo has become available [01:35:54] --- dev-zero@jabber.org has become available [01:46:49] --- Simon Wilkinson has become available [01:49:02] I suspect, if it's an unused global error number, then we should avoid starting to use it. [01:52:01] --- dev-zero@jabber.org has left [01:52:15] --- dev-zero@jabber.org has become available [03:18:11] --- haba has become available [04:29:53] --- Simon Wilkinson has left [04:44:09] --- Simon Wilkinson has become available [04:51:10] --- Simon Wilkinson has left [05:35:11] --- haba has left [05:35:11] --- abo has left [05:35:12] --- reuteras has left [05:35:22] --- haba has become available [05:35:22] --- reuteras has become available [05:35:23] --- abo has become available [06:01:25] --- Jeffrey Altman has become available [06:02:33] --- sjenkins278 has left [06:03:08] --- jaltman has left: Replaced by new connection [06:03:08] --- jaltman has become available [06:21:35] --- reuteras has left [06:21:35] --- haba has left [06:21:35] --- abo has left [06:21:42] --- haba has become available [06:21:44] --- abo has become available [06:21:56] --- reuteras has become available [06:43:40] --- Jeffrey Altman has left [06:55:14] --- deason has become available [06:57:17] --- reuteras has left [07:00:41] --- abo has left [07:00:44] --- abo has become available [07:00:44] --- reuteras has become available [07:00:49] --- reuteras has left [07:08:22] the gerrit, she needs kicking [07:49:29] --- haba has left [07:49:32] --- abo has left [07:49:40] --- haba has become available [07:49:40] --- abo has become available [08:29:26] --- Rrrrred has become available [08:29:52] sysname for 10.6.2 -- is that x86_darwin_100 or x86_darwin_90 ? [08:36:19] ok, looks like it's _100 [08:38:36] nope... compile fails [08:39:30] http://pastebin.com/m7d6346f [09:04:20] --- haba has left [09:14:04] the unsupported architecture trying to compile on the other computer gives me... [09:14:25] http://pastebin.com/m6bdd0ac1 [09:32:39] --- Rasmus Kaj has left [09:47:42] setenv ARCHFLAGS "-arch i386 -arch x86_64" [09:48:22] for the unsupported arch one? [09:48:38] oh [09:48:40] wait. [09:48:47] you' re compiling... what on what? [09:48:57] oafs 1.4.11 [09:49:00] on 10.6.2 [09:49:07] on a brand spanking new 27" iMac [09:49:13] stock 1.4.11? [09:49:30] downloaded from the openafs.org web page [09:49:42] so, no. it will never work [09:49:47] oh? [09:50:49] 1.4.11 was released before 10.6 was final. do the math. [09:51:07] so, where's the correct source? [09:52:14] --- mho has become available [09:52:31] or will just applying this .diff file I see on the 10.6 webpage do the trick? [09:52:39] http://www.openafs.org/release/1.4.11/index-macos-10.6.html [09:52:47] seems like a good idea, doesn't it? [09:53:03] downloading the source and compiling also seemed like a good idea [09:53:30] hell, buying a new mac seemed like a good idea [09:53:57] i hear we announce things. [09:54:07] "In conjunction with the release today of MacOS 10.6 Snowleopard, OpenAFS 1.5.62 has been released with support for 32 and 64 bit Snowleopard, both kernel and userspace. Additionally binary package and source patch are available for OpenAFS 1.4.11 for MacOS 10.6. [09:54:31] a note on the source link might be good... or just applying that patch to the source distro that's downloadable [09:55:02] if i apply the diff, it won't be 1.4.11 anymore [09:55:21] 1.4.11b ? [09:55:26] source tarballs are immutable. they need to stay the same forever [09:56:28] 1.4.11a, maybe. 1.4.11b should be one soon or we should just push 1.4.12. i meant to see if people were willing/able to have a release meeting soon. i should do that [09:57:22] here's where the confusion happens [09:57:29] if you click the "MacOS X" link on the main page [09:57:38] you get binaries. [09:57:42] right [09:57:45] no note about a diff [09:57:47] do they not work? [09:58:00] on one computer, no [09:58:03] on the other, yes [09:58:04] you're living in the space between people who just want it to work, and people who are developers. [09:58:25] did you try 1.5.x on the other computer? [09:58:33] e.g. do we need to backport something>? [09:58:46] to get the page w. the link about the diff... you click 1.4.x maintenance release, and then you DO NOT go to the section labeled "source code" [09:59:08] which is where the source is for every other piece EXCEPT that one diff for 10.6 [09:59:23] to get that, you go to a link that is usually only binaries [09:59:43] to know the binaries didn't work, i assume you got them from somewhere [10:00:09] so, that diff should additionally be linked in the "source code" section of the 1.4.x Maintenance Release [10:00:30] yeah... binaries all came from the web page [10:00:48] probably. guess what? that page is in git. push a change to gerrit and i'll approve it :) [10:01:13] I don't believe I have gerrit access [10:01:24] i don't believe you are correct [10:01:35] there is public access to gerrit? [10:02:15] --- deason has left [10:02:54] --- deason has become available [10:03:00] --- stevenjenkins has left [10:03:11] --- abo has left [10:03:24] --- abo has become available [10:05:42] http://www.dementia.org/twiki/bin/view/AFSLore/GitDevelopers [10:05:49] except, you want module openafs-web [10:05:56] otherwise, it's nothing special [10:06:27] --- stevenjenkins has become available [10:06:39] ok, cool [10:09:07] --- abo has left [10:09:11] --- abo has become available [10:09:20] hmm... patch seems to think that some of this stuff is aloready applied (eg src/afs/DARWIN/osi_inode.c , osi_misc.c ) -- "Reversed (or previously applied) patch detected!" [10:09:56] and now the snowleopard patch is trying to patch freebsd and aix stuff [10:10:07] it will do that. [10:10:20] you know, you could do this the easier way: git checkout openafs-stable-1_4_x [10:10:40] easier ways are awesome [10:10:57] *sigh* does os x not come with a git binary? [10:11:12] alas no [10:12:17] now, as to gerrit needing a restart, that i can probably do now. [10:12:27] let's find out [10:12:38] that would be appreciated [10:13:34] *compiles git* [10:14:08] aw, hell, there's a binary package [10:14:10] let's use that [10:14:20] i compiled it [10:14:30] the binary not work on snow leopard? [10:14:51] well, my compile is going, anyways [10:15:17] ls [10:16:51] fink and macports also have a git package, if you're into that [10:18:57] so, I want... git checkout git://git.openafs.org/openafs.git origin/openafs-stable-1_4_x ? [10:19:09] i used macports, but it builds from source [10:21:05] that command gives me "not a git repository (or any of the parent directories): .git" [10:21:48] ah, changing checkout to clone seems to work [10:23:27] git checkout git://git.openafs.org/openafs origin/openafs-stable-1_4_x [10:23:40] er, clone. yeah. [10:23:49] and the ".git" part of the module is optional [10:23:58] and cloning openafs-web will get you that, fwiw [10:24:06] ok [10:24:12] sorry, network here went to shit. i think someone's torrenting [10:27:16] --- Rasmus Kaj has become available [10:29:06] guess that remctl script is still broken [10:32:03] so, compile still fails on this one computer [10:32:17] you set ARCHFLAGS before trying? [10:32:24] different error [10:32:46] cputype (7) does not match previous archive members cputype (16777223) (all members must match) [10:33:03] you set ARCHFLAGS before trying? [10:33:44] retrying w. that [10:34:16] remember to make clean first [10:34:35] did a distclean and re-configured [10:35:14] seems to be going this time [10:35:19] wonder what's different about this imac [10:37:25] --- deason has left [10:37:48] --- deason has become available [10:40:50] --- Simon Wilkinson has become available [10:44:24] Looks like the IMA change is going to be more convoluted than expected. [10:44:29] --- Simon Wilkinson has left [10:44:29] --- Simon Wilkinson has become available [10:45:21] --- Simon Wilkinson has left [10:47:08] sigh [10:53:53] can we not get them to change the dentry_open semantics wrt ima? that sounded really broken when you last described it [10:54:15] --- Simon Wilkinson has become available [10:54:37] That's the new plan. [10:54:53] But it's a bigger change than just s/EXPORT_SYMBOL_GPL/EXPORT_SYMBOL/ [11:01:19] Gerrit kicked, btw. [11:13:42] thanks [11:30:15] --- abo has left [11:30:15] --- Rasmus Kaj has left [11:30:23] --- abo has become available [11:31:56] --- meffie has become available [11:33:44] --- Simon Wilkinson has left [11:47:27] --- Rasmus Kaj has become available [11:52:46] --- abo has left [11:53:06] --- abo has become available [12:06:16] --- jaltman has left: Replaced by new connection [12:06:20] --- jaltman has become available [12:07:56] --- Rasmus Kaj has left [12:12:22] --- shadow@gmail.com/owl9F142B33 has become available [12:12:23] --- kula has left [12:13:19] --- shadow@gmail.com/owl287E8C29 has left [12:20:12] --- Rasmus Kaj has become available [12:28:44] --- kula has become available [12:34:25] so, recompiling afs pretty much fixed things for this one imac [12:54:05] --- Russ has become available [13:06:46] well, fixed some things [13:06:49] others are still broke [13:07:45] such as huge delays when trying to browse afs with the finder [13:07:53] even when browsing folders in the local cell [13:09:13] just in finder? is accessing via e.g. terminal better? [13:09:20] terminal works fine [13:11:02] --- pod has left [13:11:21] interesting... the program I compiled over on the new imac won't work on this mac mini [13:16:53] Core Duo vs Core 2 Duo [13:23:35] --- dev-zero@jabber.org has left [13:44:25] 32-bit/64-bit weirdness [13:44:37] maybe that's why the openafs pref pane no worky [13:49:23] --- mho has left [13:49:23] --- abo has left [13:49:26] --- abo has become available [13:53:51] try 1.5 [13:54:51] well, it works on the core duo [13:54:56] but not the core 2 duo [13:55:55] none of which says anything about what i said [13:56:17] I take that back... doesn't work on the core duo, either [13:56:22] (it is clearly a 32/64 issue.) [13:56:30] ok, then i also take it back [13:56:49] while it displays and releases tokens just fine [13:56:53] can't obtain tokens with it [13:57:10] cross realm principals, it just returns but never actually gets the token [13:57:15] non-cross realm, it hangs [13:57:51] ah, looks like it finally unhung [14:02:28] but still didn't work [14:05:02] so, in the case of the finder, it seems to sit there spinning for 15, 20, 30 seconds, then finally just pops up [14:05:46] no bulkstat [14:05:49] also, try 1.5 [14:06:01] what is bulkstat? [14:06:36] finder wants to know what every object is, and access rights. you stat() every one. bulkstat is.... bulk stat() [14:07:05] I should also add that the tcpdump and fstrace dump seem to show nothing happening.. then all of a sudden, all those stats and results seem to pretty much happen at once [14:08:14] i'd suggest fstrace, but only with 1.5 [14:08:42] is 1.5 considered production ready? [14:09:07] what answer do you want? [14:09:35] the nswer i will give you is if 1.5 works for you and 1.4 doesn't, it depends whether production is "want to use afs" or "want to stand on principle" [14:09:35] since I need to give the computer to a user [14:09:44] i use only 1.5 [14:09:55] i seem to give you binaries that exist and are not trashed [14:10:01] take it for what you will [14:10:06] is 1.5 considered (it fixes a lot of stuff but has a lot of things that will break normal everyday use) [14:10:18] --- deason has left [14:10:24] i know of nothing weird in 1.5.x now [14:10:28] --- stevenjenkins has left [14:10:39] --- deason has become available [14:10:44] ok. so, we'll give 1.5 a shot [14:10:49] d/l from git? [14:11:27] yeah, sure, i guess we aqre due for 1.5.67 and you'd get 1.5.66 which i know of fixes since [14:11:32] not that any should matter [14:12:11] --- deason has left [14:12:16] ok, I see there's a dmg [14:12:20] will start w. that [14:12:44] --- deason has become available [14:14:18] --- stevenjenkins has become available [14:50:05] --- dev-zero@jabber.org has become available [15:01:20] --- RedBear has become available [15:01:33] I sorta un-take it back [15:01:49] unless nothing I am typing is showing uphere [15:01:53] ok [15:01:57] I sorta un-take it back [15:02:06] on the core 2 duo [15:02:32] the afs prefs piece did not work at all... when I recompiled from source, that gave me an afs prefs piece broken in the same way as on the core duo [15:02:53] so, anyway, we'll try 1.5.x (tomorrow, hopefully) and see what happens w. that and w. the finder [15:05:02] finder won't change [15:05:02] --- summatusmentis has left: Lost connection [15:05:08] no bulkstat. no bulkstat [15:05:16] unless there's something else going on. well. try it [15:05:16] --- summatusmentis has become available [15:05:24] --- summatusmentis has left: Lost connection [15:05:53] --- dev-zero@jabber.org has left [15:06:15] --- summatusmentis has become available [15:06:15] --- summatusmentis has left: Lost connection [15:06:16] --- summatusmentis has become available [15:06:18] --- dev-zero@jabber.org has become available [15:06:23] --- summatusmentis has left: Lost connection [15:10:16] --- summatusmentis has become available [15:10:17] --- summatusmentis has left: Lost connection [15:15:17] --- summatusmentis has become available [15:15:21] --- meffie has left [15:15:28] something else, as it didn't use to be that bad [15:17:06] --- Russ has left: Disconnected [15:32:31] --- mdionne has become available [15:32:38] --- summatusmentis has left: Lost connection [15:35:18] --- summatusmentis has become available [15:35:23] --- summatusmentis has left: Lost connection [15:40:18] --- summatusmentis has become available [15:40:18] --- summatusmentis has left: Lost connection [15:45:18] --- summatusmentis has become available [15:45:18] --- summatusmentis has left: Lost connection [15:46:25] --- shadow@gmail.com/owl9F142B33 has left [15:48:10] --- shadow@gmail.com/owlDF620AF4 has become available [15:50:19] --- summatusmentis has become available [15:50:19] --- summatusmentis has left: Lost connection [15:55:19] --- summatusmentis has become available [15:55:21] --- summatusmentis has left: Lost connection [16:00:19] --- summatusmentis has become available [16:00:24] --- summatusmentis has left: Lost connection [16:05:20] --- summatusmentis has become available [16:28:57] --- dev-zero@jabber.org has left: Replaced by new connection [16:28:58] --- dev-zero@jabber.org has become available [16:43:30] --- abo has left [16:53:11] --- abo has become available [16:53:17] --- abo has left [17:00:56] --- abo has become available [17:07:40] --- deason has left [17:07:43] --- deason has become available [17:07:52] --- deason has left [17:12:53] --- deason has become available [17:16:48] --- mdionne has left [17:35:15] --- deason has left [17:37:43] --- mdionne has become available [17:42:08] > Looks like the IMA change is going to be more convoluted than expected. [17:44:16] No kidding. The tone of the discussion is disappointing, but I guess there should be no surprise there. But from the last message in the thread on LSM I get a feeling that there may be enough cooler heads to prevail. [18:22:13] --- summatusmentis has left [19:06:22] --- mdionne has left [20:35:56] --- abo has left [20:36:05] --- Rasmus Kaj has left [20:47:09] --- abo has become available [20:47:09] --- abo has left [20:47:09] --- abo has become available [20:47:28] --- Rasmus Kaj has become available [21:05:37] --- kula has left [22:20:04] --- kula has become available [22:31:11] --- reuteras has become available [23:06:01] --- reuteras has left [23:06:04] --- abo has left [23:11:16] --- abo has become available [23:11:17] --- abo has left [23:15:32] --- reuteras has become available [23:15:37] --- reuteras has left [23:15:39] --- abo has become available [23:15:40] --- reuteras has become available [23:33:45] --- dev-zero@jabber.org has left