2008-03-05T00:13:44 *** greg-g has quit IRC 2008-03-05T00:13:58 *** greg-g has joined #openils-evergreen 2008-03-05T00:19:11 *** greg-g has quit IRC 2008-03-05T00:39:15 *** EGUEST804 has joined #openils-evergreen 2008-03-05T00:39:37 *** EGUEST804 has left #openils-evergreen 2008-03-05T01:17:29 *** gmcharlt_ has joined #OpenILS-Evergreen 2008-03-05T01:26:37 *** gmcharlt__ has joined #OpenILS-Evergreen 2008-03-05T01:29:43 *** gmcharlt has quit IRC 2008-03-05T01:34:04 *** gmcharlt_ has quit IRC 2008-03-05T02:18:29 *** pmurray_away has quit IRC 2008-03-05T02:22:09 *** pmurray_away has joined #openils-evergreen 2008-03-05T02:25:49 *** Mark__T has joined #openils-evergreen 2008-03-05T02:37:47 *** tristanbob_ has quit IRC 2008-03-05T02:37:47 *** kados has quit IRC 2008-03-05T02:40:30 *** tristanbob_ has joined #openils-evergreen 2008-03-05T02:40:30 *** kados has joined #openils-evergreen 2008-03-05T06:37:24 *** gmcharlt__ has quit IRC 2008-03-05T06:37:35 *** gmcharlt has joined #OpenILS-Evergreen 2008-03-05T07:55:43 *** dmcmorris_esi has joined #openils-evergreen 2008-03-05T08:22:39 *** greg-g has joined #openils-evergreen 2008-03-05T08:32:57 *** dbs has joined #OpenILS-Evergreen 2008-03-05T09:24:04 *** pmurray_away is now known as pmurray 2008-03-05T09:42:02 miker_++ # acq models 2008-03-05T09:42:11 does it seem non-broken? 2008-03-05T09:42:12 * berick attacks the ML 2008-03-05T09:42:19 looks good so far 2008-03-05T09:44:13 good 2008-03-05T09:52:05 haha .. %s/acqple/jub/g 2008-03-05T09:52:46 yeah 2008-03-05T09:52:59 other class names changed too... 2008-03-05T09:53:31 anything that started with acqpoli before (other than that exactly) should now start with acqli 2008-03-05T09:53:34 s/po// 2008-03-05T09:53:47 k 2008-03-05T10:04:17 *** gmcharlt has quit IRC 2008-03-05T10:10:52 i wonder if we should just rebrand MALTA as a regional code4lib 2008-03-05T10:11:00 *** Mark__T has left #openils-evergreen 2008-03-05T10:11:19 people seem to want that 2008-03-05T10:11:33 yeah 2008-03-05T10:11:42 multiple c4lc I mean 2008-03-05T10:11:52 although i'm afraid it would alienate the people i was trying to enfranchise with malta in the first place 2008-03-05T10:12:05 of course, that means we'll never see the RealThing near hear again... 2008-03-05T10:12:09 with dirty canadians and yankees 2008-03-05T10:12:44 we probably should try to have another malta soon 2008-03-05T10:12:51 * rsinger cleans out his garage to host it. 2008-03-05T11:11:19 *** gmcharlt has joined #OpenILS-Evergreen 2008-03-05T11:17:59 durrrty 2008-03-05T11:27:22 berick: man, I can't believe you actually used that -c flag 2008-03-05T11:27:23 dbs-- 2008-03-05T11:27:23 dbs: Error: You're not allowed to adjust your own karma. 2008-03-05T11:27:46 pinesol-- 2008-03-05T11:28:06 dbs: haha.. 2008-03-05T11:29:14 i often get "takes a paramter" and "option required" mixed up with getopt 2008-03-05T11:29:53 time to rm dbs from svnserve's password-db 2008-03-05T11:30:38 berick: yeah, well bad on me for not testing the existing use case and focusing only on the behaviour I was aiming for 2008-03-05T11:31:16 dbs: re svn access, wait till you see the steaming pile of opensrf i'm gonna commit (hopefully in short order) ;) 2008-03-05T11:31:25 it's gonna rock yr world 2008-03-05T11:31:29 ooh 2008-03-05T11:31:41 * dbs goes to get a coffee in preparation 2008-03-05T11:32:15 hey, where did this steaming pile come from? 2008-03-05T11:32:20 well, i'm banging on ACQ right now. if i can get the ML and controllers updated in time, i'll be doing osrf this afternoon 2008-03-05T11:33:44 rsinger: it's new (optional) functionality for running multi-domain opensrf. private services registered on one private domain, and public services registered on both private and public domains. 2008-03-05T11:34:05 berick: oh, cool 2008-03-05T11:34:23 my world is rocking 2008-03-05T11:34:27 it's a security measure.. makes it easier to develop gateways/interfaces without having to build new security into those module 2008-03-05T11:34:32 er, modules 2008-03-05T11:34:42 berick: so in a jangle context: you could have a public item service but private actor? 2008-03-05T11:35:59 rsinger: in general, yes, if you didn't need to publicly expose any actor methods. 2008-03-05T11:36:17 so i can really become a pines borrower even though i'm in fulton county? 2008-03-05T11:36:32 berick: right, that's a crude example 2008-03-05T11:36:36 we tend to segregate the apps based on whether the methods handle their own authorization/authentication. for example, the DB apps are wide open, thus private 2008-03-05T11:36:47 * rsinger nods. 2008-03-05T11:37:10 rsinger: yeah, anyone in GA can 2008-03-05T11:37:39 clayton county is close enough 2008-03-05T11:58:49 *** gmcharlt has quit IRC 2008-03-05T12:07:20 *** pmurray has quit IRC 2008-03-05T12:08:35 *** pmurray has joined #openils-evergreen 2008-03-05T12:12:21 *** kados has quit IRC 2008-03-05T12:12:21 *** tristanbob_ has quit IRC 2008-03-05T12:14:18 *** kados has joined #openils-evergreen 2008-03-05T12:17:56 *** tristanbob_ has joined #openils-evergreen 2008-03-05T12:18:35 *** greg-g has quit IRC 2008-03-05T12:20:29 *** greg-g has joined #openils-evergreen 2008-03-05T12:24:27 *** greg-g has quit IRC 2008-03-05T12:24:37 *** greg-g has joined #openils-evergreen 2008-03-05T12:29:00 *** Karen_ has joined #OpenILS-Evergreen 2008-03-05T12:48:42 *** sylvar has joined #openils-evergreen 2008-03-05T12:50:43 ...so is there any reason why the OPAC must display all of these item statuses, even when there's only one copy and therefore only one column is relevant? Seems to me that it'd be nice to have a setting like "on the bib page in OPAC, display only the item statuses that are applicable to this bib" 2008-03-05T12:52:56 sylvar: consistency, I would assume; it might be disorienting for a user to see different displays for every bib 2008-03-05T12:52:57 I can image some folks expecting consistency in the column display 2008-03-05T12:53:02 hah 2008-03-05T12:53:53 sylvar: we just need to make a smaller set of columns be the "norm" for your installation 2008-03-05T12:53:59 for a single-institution Evergreen instance, you would certainly want to customize the opac to provide a simpler display; but it makes sense for state-wide instances 2008-03-05T12:54:01 Fair enough. 2008-03-05T12:56:52 miker_, would you be willing to give me some suggestions on postgres tuning? I think the postgresql.conf is pretty much standard, and I'd like to speed it up so that it can actually return OPAC results within 60 seconds when the search term is very common. 2008-03-05T12:57:53 sylvar: I'd aim lower ;) maybe within 4 seconds.. I could tolerate that :) 2008-03-05T12:58:12 well, 60 seconds is when it gives up. :) 2008-03-05T12:58:31 true enough 2008-03-05T12:59:02 http://evergreen-test.library.utah.edu/ is the server. A search for "androgenic" returns results. A search for "utah" doesn't. 2008-03-05T12:59:25 And I've broken down the SQL query into small enough parts to convince myself that that's where the bottleneck is. 2008-03-05T12:59:38 It can sort a few dozen results, but not 3000. 2008-03-05T13:00:18 sylvar: shared_buffers is the most important piece, I believe 2008-03-05T13:00:47 800MB at the moment 2008-03-05T13:01:03 ah, well good that you've increased that :) 2008-03-05T13:01:53 it could also be some inherent bottleneck in the VM's host's disks or something, I don't know 2008-03-05T13:02:16 if I were Google I'd have the whole db in a ramdisk, but... :) 2008-03-05T13:04:23 max_fsm_pages could be useful too 2008-03-05T13:04:31 oohhh - vm? Hmm. 2008-03-05T13:05:08 kinda scary, that 2008-03-05T13:06:00 max_fsm_pages = 153600 2008-03-05T13:06:48 if you have enough ram, could try pre-caching all the bib records :) 2008-03-05T13:07:50 riiiiiight. 2008-03-05T13:08:00 you could probably bump max_fsm_pages up to 1000000 safely 2008-03-05T13:09:28 but that probably won't make much difference to sort speed 2008-03-05T13:11:22 crank up work_mem 2008-03-05T13:11:31 that's what controls the sorting 2008-03-05T13:12:34 what are you at with work_mem right now? 2008-03-05T13:13:42 work_mem is default (1MB?) 2008-03-05T13:14:35 oof 2008-03-05T13:14:40 what's better? 2008-03-05T13:15:07 how much ram do you have 2008-03-05T13:15:56 (if you've bumped shared_buffers from 24MB to 800MB, I would be tempted to bump work_mem up by a similar scale) 2008-03-05T13:18:42 Overall about 4GB. 2008-03-05T13:19:26 So you think about 32MB then? 2008-03-05T13:20:36 sounds like a good start! 2008-03-05T13:20:52 about to eat lunch, but yeah, you need a lot more work_mem 2008-03-05T13:20:59 (anything that doesn't fit into the work_mem space gets dumped to disk, which will _kill_ performance) 2008-03-05T13:21:00 also, VM-- for anything "real" 2008-03-05T13:21:06 hah :) 2008-03-05T13:21:14 I take the blame for vm :) 2008-03-05T13:21:18 nono 2008-03-05T13:21:32 VM is great for showing off a few thousand records 2008-03-05T13:21:40 the scale of a small library, say 2008-03-05T13:21:55 hey - i'm part of equinox apparently: http://blogs.talis.com/panlibus/archives/2008/03/code4lib_2008_l.php 2008-03-05T13:22:03 * dbs looks for his pay cheque 2008-03-05T13:22:03 but for utah's 1M+ bibs? no way 2008-03-05T13:22:12 miker_: I know, I know :) 2008-03-05T13:22:22 OK, if you recommend a physical server, I"ll see what I can do. 2008-03-05T13:22:42 sylvar: ideally, with a nice RAID array of fast disks 2008-03-05T13:22:48 sure 2008-03-05T13:23:35 sylvar: also, 1.4 will have an improved search mechanism ... same results, much faster 2008-03-05T13:23:45 that's what I'm working on this week 2008-03-05T13:24:07 Cool. Any guesses about a release date for 1.4.0.0? 2008-03-05T13:24:44 as soon as we're ready ;) ... I'm still holding out hope for this month ... RC for sure this month 2008-03-05T13:26:14 * dbs makes a mental note to stop filing expense reports and to get back on the i18n juggernaut 2008-03-05T13:26:25 Excellent! 2008-03-05T13:30:07 well, when I increased work_mem and restarted postgres, my next search gave me "* Fix Me * Error parsing JSON [SyntaxError: missing } after property list]" 2008-03-05T13:31:55 do I need to shut down everything? 2008-03-05T13:32:41 sylvar: if you restart PG, you will need to restart everything else as well 2008-03-05T13:36:13 including memcached 2008-03-05T13:39:06 right: apache2 stop, memcached stop, stop_all, start_all, memcached start, apache2 start. 2008-03-05T13:46:38 Hmm. work_mem is now 32MB, everything's restarted, I've got about half a gig less free RAM, but performance hasn't changed much. I'm thinking disk bottleneck, maybe. 2008-03-05T13:48:16 *** dmcmorris_esi has left #Openils-Evergreen 2008-03-05T13:48:29 "free" ram doesn't mean much. it's probably cached 2008-03-05T13:48:43 cached data, I mean. by the OS 2008-03-05T13:49:24 Linux likes to use all available RAM - and that's a good thing. 2008-03-05T13:49:31 indeed 2008-03-05T13:49:51 * sylvar nods and runs bonnie++ to benchmark the disk 2008-03-05T13:52:31 hey guys, what's the status/deal with the django interface? are we still doing anything with that? 2008-03-05T13:53:09 phasefx: undecided, I think 2008-03-05T13:53:39 I'm planning the dojo+opensrf interface, but haven't had a chance to start writing code 2008-03-05T13:53:48 if you want to jump in, be my guest 2008-03-05T13:54:36 I have a full plate, someone was just asking about it. Thanks! 2008-03-05T13:55:16 miker_: it would conceivably take the place of the bootstrap cgi and maybe the other admin interfaces as well? 2008-03-05T13:56:01 bootstrap cgis and django ... maybe some of the admin interfaces, but dunno yet 2008-03-05T13:56:27 what is django doing now that's different from the bootstrap cgis? 2008-03-05T13:56:56 deleting things 2008-03-05T13:57:01 haha 2008-03-05T13:57:37 and not using a tree :) 2008-03-05T13:57:40 but the role of those two currently are the same? 2008-03-05T13:57:47 more or less 2008-03-05T13:57:58 I think I have a picture now. Thanks again! 2008-03-05T13:58:00 django doesn't understand the structure 2008-03-05T13:58:17 or parts of it 2008-03-05T14:00:45 hmm... /var/lib/postgresql/8.2/main/base/ would be my database, most likely, yes? Didn't realize it was 55GB already. Furrfu! 2008-03-05T14:02:32 sylvar: have you analyze'd and vacuum'ed recently? 2008-03-05T14:03:15 I haven't, but a colleague might have... would it be a good thing or a bad thing? 2008-03-05T14:03:40 good thing 2008-03-05T14:06:58 analyze helps query plans, vacuum recovers disk space 2008-03-05T14:12:10 OK, running "VACUUM ANALYZE;" now. 2008-03-05T14:14:47 miker_: i'd like to propose one additional change to opensrf_core.xml with this new multi-domain stuff: putting a container element around the 1 or more chunks. this will allow the router to start as many processes as it needs based on the single config 2008-03-05T14:15:44 berick: I like it ... we've already broken compat with so one more won't hurt 2008-03-05T14:15:54 my thought exactly 2008-03-05T14:39:10 Drat. I lost my network connection before vacuum analyze finished. Do I need to start that again? (perhaps as nohup vacuumdb) Or is it probably continuing without me? 2008-03-05T14:49:31 sylvar: i'd check 'top' to see if you see a vacuum process running 2008-03-05T15:18:39 dbs: I put a user generator on the wiki; it's very brain dead, but you might find it useful for your vmware images 2008-03-05T15:35:55 berick: cool! 2008-03-05T15:36:03 err, phasefx: cool! 2008-03-05T15:37:07 made for some very interesting demographics when I demo'ed reporting a while back :) had folks from all over the US in each of these Georgia libraries 2008-03-05T15:37:29 *** phasefx__ is now known as phasefx_ 2008-03-05T15:39:03 * dbs still thinks an evergreen-contrib ("use at own risk! even riskier than evergreen!") subversion repository would be useful 2008-03-05T15:39:36 does sound useful 2008-03-05T15:40:45 dbs: egfoundry ;) 2008-03-05T15:42:22 miker_: sounds good to me; I could run it at coffeecode.ca 2008-03-05T15:43:03 dbs: well, you have access to open-ils.org too 2008-03-05T15:43:12 but wherever, really 2008-03-05T15:43:43 if you're comfy with me adding a subversion repository to open-ils.org, okay 2008-03-05T15:45:01 any reason not to just add a repo to svn.open-ils.org? 2008-03-05T15:48:42 none that I can think of off hand ... checking into that now 2008-03-05T15:49:05 i would vote for svn.open-ils.org as well 2008-03-05T15:49:44 and we'd just give a directory and commit to anyone who asks? 2008-03-05T15:50:49 phasefx_: well... anyone that proposes a project on the list which gets (at least) a second 2008-03-05T15:50:52 IMO 2008-03-05T15:50:59 sounds like a good ground rule 2008-03-05T15:51:19 works for me 2008-03-05T15:51:19 I mean, we're not trying to be sourceforge, but if there's interest and it's eg-centric, and more than one person cares, yeah 2008-03-05T16:04:14 Any ideas about this error? 2008-03-05T16:04:18 2008-03-05 14:03:07 MST HINT: This error usually means that PostgreSQL's request for a shared memory segment exceeded your kernel's SHMMAX parameter. You can either reduce the request size or reconfigure the kernel with larger SHMMAX. To reduce the request size (currently 860839936 bytes), reduce PostgreSQL's shared_buffers parameter (currently 102400) and/or its max_connections... 2008-03-05T16:04:19 * Restarting PostgreSQL 8.2 database server 2008-03-05T16:04:20 * The PostgreSQL server failed to start. Please check the log output: 2008-03-05T16:04:22 2008-03-05 14:03:07 MST LOG: could not load root certificate file "root.crt": no SSL error reported 2008-03-05T16:04:23 2008-03-05 14:03:07 MST DETAIL: Will not verify client certificates. 2008-03-05T16:04:25 2008-03-05 14:03:07 MST FATAL: could not create shared memory segment: Invalid argument 2008-03-05T16:04:26 2008-03-05 14:03:07 MST DETAIL: Failed system call was shmget(key=5432001, size=860839936, 03600). 2008-03-05T16:04:28 ...parameter (currently 100). 2008-03-05T16:04:29 If the request size is already small, it's possible that it is less than your kernel's SHMMIN parameter, in which case raising the request size or reconfiguring SHMMIN is called for. 2008-03-05T16:04:32 The PostgreSQL documentation contains more information about shared memory configuration. 2008-03-05T16:04:34 ...fail! 2008-03-05T16:05:24 shared_buffers = 800MB, work_mem = 32MB, max_fsm_pages = 1000000 2008-03-05T16:05:44 max_connections=100? 2008-03-05T16:06:09 yes 2008-03-05T16:06:52 should it be lower, especially for a test server? 2008-03-05T16:07:38 just reading the error message 2008-03-05T16:09:20 no, you need to up the OS shared mem 2008-03-05T16:09:21 sec 2008-03-05T16:10:23 sysctl blah blah 2008-03-05T16:10:42 echo 400000000 > /proc/sys/kernel/shmmax 2008-03-05T16:10:57 I put that in the script I use to start PG 2008-03-05T16:12:08 actually, you'll need /at least/ 860839936 2008-03-05T16:14:22 or sysctl -w kernel.shmmax= 2008-03-05T16:14:29 (same thing) 2008-03-05T16:14:52 indeed 2008-03-05T16:15:04 or set it in /etc/sysctl.conf 2008-03-05T16:15:17 (this is like db2 config tuning all over again) 2008-03-05T16:16:02 heh 2008-03-05T16:23:35 dbs: feel free to go to town on svn.open-ils.org ... are you thinking of a trac-based thing (since it does svn+wiki+bugtracking) or more? 2008-03-05T16:24:07 just track 2008-03-05T16:24:12 s/k// 2008-03-05T16:25:58 but yeah, turning on tickets + opening up the wiki for project pages makes sense 2008-03-05T16:26:41 it's per-project, right? 2008-03-05T16:26:49 it is 2008-03-05T16:28:06 tickets can have arbitrary categories though 2008-03-05T16:28:44 but fine... if you want something that's a bit more granular (which would make even more sense) I can dig something up 2008-03-05T16:29:25 for the foundry? 2008-03-05T16:29:31 i think trac will work just fine 2008-03-05T16:29:33 aye 2008-03-05T16:30:07 nono, I was just wondering 2008-03-05T17:00:15 *** pmurray is now known as pmurray_away 2008-03-05T17:23:35 *** dbs has quit IRC 2008-03-05T18:38:52 miker_: do I need to do anything after echo 860839936 > /proc/sys/kernel/shmmax ? restart services? restart the server? 2008-03-05T18:40:51 apparently not: http://www.performancewiki.com/linux-tuning.html says it takes effect immediately 2008-03-05T18:41:05 I've added it to /etc/sysctl.conf for the next boot 2008-03-05T18:53:44 *** gmcharlt has joined #OpenILS-Evergreen 2008-03-05T19:21:52 And it seems to have helped somewhat. I can now get results for a search with 400-500 results. Very common search terms still time out, though... 2008-03-05T19:32:32 *** Karen_ has left #OpenILS-Evergreen 2008-03-05T19:34:39 *** gmcharlt has quit IRC 2008-03-05T21:02:52 *** dbs has joined #OpenILS-Evergreen 2008-03-05T21:51:10 berick: I have tested opensrf-trunk + evergreen-trunk (just a single domain) and it is good 2008-03-05T21:55:51 dbs++ awesome 2008-03-05T21:56:36 berick: just a super-simple test (things run! yay!) 2008-03-05T21:57:05 well, that's how most people will be using it for now, i imagine 2008-03-05T21:57:07 btw - we got tabs and whitespace gunked up in the dang *.xml.example files 2008-03-05T21:57:38 arg 2008-03-05T21:57:41 berick-- 2008-03-05T21:57:41 berick: Error: You're not allowed to adjust your own karma. 2008-03-05T21:57:58 i can add some modelines 2008-03-05T21:58:22 I've got a typo to fix in there anyways 2008-03-05T22:04:29 should we bump the opensrf version attribute in opensrf.xml? 2008-03-05T22:04:36 it's been at 0.0.2 for a while... 2008-03-05T22:09:11 oh, that's a good point. it should change with these changes. guess we could bump it up to .3 for now, then push it up 1.0.0 when osrf goes 1.0? 2008-03-05T22:09:20 sure 2008-03-05T22:11:53 whitespace "consistency" - I went with 2-space indents in opensrf_core.xml, and 4-space in opensrf.xml, as that was the prevailing style for each file 2008-03-05T22:12:15 right on 2008-03-05T22:27:20 mmm, forgot that python-dev would be a dependency for python in opensrf 1.0 (and to enable building python in opensrf's install.conf) 2008-03-05T22:30:26 god I hate python-memcached and debian's version of setuptools 2008-03-05T22:31:25 @tell dbs_doc_writer_of_the_future warn people to install python-memcached manually until debian 4.0 goes away 2008-03-05T22:31:25 dbs: Error: I haven't seen dbs_doc_writer_of_the_future, I'll let you do the telling. 2008-03-05T22:34:18 hah - I have an awesome hackity-hack for that 2008-03-05T22:35:20 i can't stand the python DNS handling 2008-03-05T22:39:49 it is.. irksome 2008-03-05T22:40:44 debian 3.1 is going away soon. debian 4.0 should be gone (unsupported) about one year after lenny comes out. :) 2008-03-05T22:42:11 so, about Sep 2009 2008-03-05T22:43:31 (not that any of that is helpful ;) 2008-03-05T22:49:22 jeff - no, that's helpful - but I've managed to route around debian 4.0's brain-damaged version of setuptools 2008-03-05T22:49:22 is it less brain-damaged in testing/unstable? 2008-03-05T22:53:26 *** tristanbob_ has quit IRC 2008-03-05T22:53:26 *** denials has quit IRC 2008-03-05T22:53:26 *** scottmcd has quit IRC 2008-03-05T22:53:26 *** phasefx has quit IRC 2008-03-05T22:53:26 *** sarabee has quit IRC 2008-03-05T22:53:26 *** greg-g has quit IRC 2008-03-05T22:53:26 *** sylvar has quit IRC 2008-03-05T22:53:26 *** pmurray_away has quit IRC 2008-03-05T22:53:26 *** lisppaste6 has quit IRC 2008-03-05T22:53:26 *** kados has quit IRC 2008-03-05T22:53:26 *** berick has quit IRC 2008-03-05T22:53:26 *** jeff has quit IRC 2008-03-05T22:53:26 *** peanutb has quit IRC 2008-03-05T22:53:26 *** dbs has quit IRC 2008-03-05T22:53:26 *** gsf has quit IRC 2008-03-05T22:53:26 *** bradl has quit IRC 2008-03-05T22:53:26 *** phasefx_ has quit IRC 2008-03-05T22:53:26 *** miker_ has quit IRC 2008-03-05T22:55:05 *** eeevil has joined #openils-evergreen 2008-03-05T22:55:05 *** dbs has joined #openils-evergreen 2008-03-05T22:55:05 *** sylvar has joined #openils-evergreen 2008-03-05T22:55:05 *** greg-g has joined #openils-evergreen 2008-03-05T22:55:06 *** tristanbob_ has joined #openils-evergreen 2008-03-05T22:55:06 *** kados has joined #openils-evergreen 2008-03-05T22:55:06 *** pmurray_away has joined #openils-evergreen 2008-03-05T22:55:06 *** jeff has joined #openils-evergreen 2008-03-05T22:55:06 *** phasefx has joined #openils-evergreen 2008-03-05T22:55:06 *** scottmcd has joined #openils-evergreen 2008-03-05T22:55:06 *** denials has joined #openils-evergreen 2008-03-05T22:55:06 *** phasefx_ has joined #openils-evergreen 2008-03-05T22:55:06 *** gsf has joined #openils-evergreen 2008-03-05T22:55:06 *** bradl has joined #openils-evergreen 2008-03-05T22:55:06 *** sarabee has joined #openils-evergreen 2008-03-05T22:55:06 *** berick has joined #openils-evergreen 2008-03-05T22:55:06 *** lisppaste6 has joined #openils-evergreen 2008-03-05T22:55:06 *** peanutb has joined #openils-evergreen 2008-03-05T22:55:49 dbs: what's brain-damaged about it? i'd hate to see humanity suffer if lenny/5.0 is still brain-dead. 2008-03-05T22:55:49 s/suffer/suffer further/ 2008-03-05T22:55:50 old version of setuptools doesn't understand FTP protocol, apparently 2008-03-05T22:58:03 Setting up python-setuptools (0.6c3-3) ... 2008-03-05T22:58:03 INFO: using unsupported version '/usr/bin/python2.5' 2008-03-05T22:58:06 * jeff frowns 2008-03-05T22:58:39 heh 2008-03-05T23:07:44 *** eeevil is now known as miker_