Team Chat Logs

2008 7
Mo Tu We Th Fr Sa Su
        1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 31

August 23, 2008

[00:04:15] * rblank has joined #trac
[00:12:42] * hpnadig has quit IRC
[00:35:58] * crump has joined #trac
[00:37:53] <evil_twin> New news from t.e.o: Ticket #7556 (1 hour per day database is locked) closed <http://trac.edgewall.org/ticket/7556#comment:1>
[00:41:02] <ldlework> http://ldlework.pastebin.com/m39f28c4c
[00:41:28] <coderanger> and that is a wsgi script made by trac-admin deploy?
[00:41:41] <ldlework> ah no I don't think so
[00:42:29] <ldlework> http://ldlework.pastebin.com/m283ab486
[00:42:35] * crump has quit IRC
[00:42:49] * cmlenz has joined #trac
[00:43:09] <coderanger> right, do not use that, use the one generated by trac-admin deploy
[00:43:25] <ldlework> it'll generate in cwd?
[00:46:34] <coderanger> no, in the folder you give as the argument to deploy
[00:46:51] <coderanger> well, inside a cgi-bin folder in the argument you give
[00:47:02] <ldlework> If my env is at /svn/gnosis/trac and my apache readable directory is /svn/gnosis/trac/apache how do I use deploy?
[00:47:11] <coderanger> @tracdoc
[00:47:11] <evil_twin> coderanger: "tracdoc" is https://coderanger.net/~coderanger/tracdoc/install/index.html
[00:47:18] <ldlework> heh
[00:47:22] <ldlework> ughhh
[00:47:31] <coderanger> See the fcgi page
[00:48:20] * liquidat0r has joined #trac
[00:48:27] * asmodai waves at coderanger
[00:48:31] <coderanger> yo
[00:48:43] <coderanger> asmodai: You better be on for the last playtest!
[00:50:54] * rblank has quit IRC
[00:52:11] * RobinD has quit IRC
[00:54:14] <asmodai> coderanger: last playtest? on Wednesday?
[00:54:32] <coderanger> yah
[00:55:10] <coderanger> well last of this round
[00:55:51] <asmodai> yeah, and I am supposed to be on how? the time is in the middle of my night and I have to work the day after :S
[00:56:07] <coderanger> lots of coffee! :)
[00:56:31] * coderanger passes out
[00:56:34] <asmodai> heh
[00:58:41] <ldlework> coderanger : I'm now using the deploy generated wsgi script
[00:58:45] <ldlework> and the authentication is still a no go
[01:00:20] <ldlework> nevermind
[01:00:20] <ldlework> I got it
[01:00:21] <ldlework> thanks
[01:17:22] * liquidat0r has quit IRC
[01:24:44] * freaky[t] is now known as fReAkY[t]
[01:24:50] * fReAkY[t] is now known as freaky[t]
[01:30:24] * EnTeQuAk has joined #trac
[01:31:42] * tinh has joined #trac
[01:33:39] * timte has joined #trac
[01:37:46] * EnTeQuAk has joined #trac
[01:38:07] * cy_` has quit IRC
[01:38:42] * davidfraser has joined #trac
[01:43:46] * cy_` has joined #trac
[01:49:46] * hwinkel has joined #trac
[02:10:11] * davidfraser has quit IRC
[02:14:26] * hpnadig has joined #trac
[02:18:16] * pto has joined #trac
[02:28:40] * gakman has quit IRC
[02:50:30] * izibi has joined #trac
[02:56:21] * m_g has joined #trac
[03:06:19] <HenrikV> So running off trunk is generally advisable?
[03:07:24] * Crissi has joined #trac
[03:07:31] <Crissi> hello
[03:07:53] <Crissi> how i can create a ticket by command line (python program is ok)
[03:07:56] <Crissi> ?
[03:10:03] * hwinkel has quit IRC
[03:10:09] * tct13 has quit IRC
[03:10:13] <HenrikV> trac-admin ?
[03:10:39] <Crissi> it can delete tickets
[03:10:43] <Crissi> but not create?
[03:11:58] <Crissi> there is only ticket remove there
[03:12:19] * rblank has joined #trac
[03:26:33] * |igor_ has joined #trac
[03:26:55] <Crissi> hrrr
[03:27:10] <Crissi> after upgrading from 10.3 to 11.0 it doest work
[03:27:27] <Crissi> cant display any ticket
[03:28:17] <Crissi> i had installed the new trac and the update the environment
[03:36:16] * izibi_ has joined #trac
[03:37:27] * to_ has quit IRC
[03:37:47] * zoli2k has joined #trac
[03:38:03] <zoli2k> Hi, I have a problem of configuration of trac with lighttpd
[03:38:31] <zoli2k> http://pastebin.com/m602a73c5
[03:38:46] <zoli2k> this is the message of lighttpd
[03:39:25] <zoli2k> I configured it according to document: TracFastCgi
[03:39:45] <zoli2k> can anyone help me how to debug this problem?
[03:40:05] <zoli2k> the environment works well with tracd but not with lighttpd
[03:42:17] * Tetraboy has joined #trac
[03:43:03] <Crissi> what the hell is that: http://paste.debian.net/15432/
[03:43:44] * izibi has quit IRC
[03:46:39] <Crissi> ok.. got it... account manager plugin must be updated too
[03:47:32] * Tetraboy2 has quit IRC
[03:50:21] * |igor_ has quit IRC
[03:50:26] <Crissi> fuckll
[03:50:28] <Crissi> fuck
[03:50:28] <Crissi> fuck
[03:50:29] <Crissi> fuck
[03:50:51] <Crissi> sometimes it works sometimes i got the unknown encoding traceback
[03:52:56] <Crissi> why the hell it does not work after upgrade...
[03:53:11] * davidfraser has joined #trac
[04:05:11] <Crissi> there is an bug in trac
[04:05:54] <Crissi> if a backtrace cames up, the content type is always text (also if the content is html)
[04:06:02] <Crissi> trac 0.11.1
[04:09:21] <evil_twin> New news from t.e.o: Ticket #7557 (RSS feed doesn't include port in URL) created <http://trac.edgewall.org/ticket/7557>
[04:13:27] <Crissi> found the problem. the webserver must be restarted too
[04:40:19] * Crissi is now known as CrisAWAY
[04:57:07] * cmlenz has quit IRC
[04:59:52] * bkuhn has joined #trac
[05:03:56] * bkuhn is now known as bkuhnIdle
[05:25:01] * sdboyer has joined #trac
[05:25:51] * bkuhnIdle is now known as bkuhn
[05:49:10] * tct13 has joined #trac
[05:58:41] * EnTeQuAk has quit IRC
[06:01:58] * yoswink has joined #trac
[06:06:27] * zoli2k has quit IRC
[06:12:45] * hpnadig has quit IRC
[06:12:45] * jcran has quit IRC
[06:12:45] * Kapsel_ has quit IRC
[06:12:45] * bergheim has quit IRC
[06:12:45] * bencer has quit IRC
[06:12:45] * coling has quit IRC
[06:12:46] * aat has quit IRC
[06:12:46] * CrisAWAY has quit IRC
[06:12:46] * ryan_ has quit IRC
[06:12:46] * lisppaste5 has quit IRC
[06:12:46] * __t has quit IRC
[06:13:34] * CrisAWAY has joined #trac
[06:13:34] * hpnadig has joined #trac
[06:13:34] * jcran has joined #trac
[06:13:34] * __t has joined #trac
[06:13:34] * lisppaste5 has joined #trac
[06:13:34] * bergheim has joined #trac
[06:13:34] * Kapsel_ has joined #trac
[06:13:34] * ryan_ has joined #trac
[06:13:34] * bencer has joined #trac
[06:13:34] * coling has joined #trac
[06:13:34] * aat has joined #trac
[06:25:02] * __t has quit IRC
[06:27:53] <rocky> coderanger: don't suppose you could throw up a screenshot for http://trac-hacks.org/wiki/NewsFlashMacro ? :)
[06:35:28] * macmaN6789 has joined #trac
[06:42:57] * liquidat0r has joined #trac
[06:48:36] * miau^_ has quit IRC
[06:51:30] * zoli2k has joined #trac
[06:53:55] * tct13 has quit IRC
[07:02:37] * |igor_ has joined #trac
[07:03:03] * to_ has joined #trac
[07:03:07] <to_> hi
[07:03:21] <to_> Is there exist a timeline macro?
[07:08:31] <to_> For example, I'd like to be able to display a timeline of the day in my wikistartpage...
[07:22:15] * |igor_ has quit IRC
[07:25:55] * Tetraboy has quit IRC
[07:26:32] * yoswink has quit IRC
[07:28:19] * yoswink has joined #trac
[07:36:46] * SlashV has joined #trac
[07:40:00] * tinh has quit IRC
[07:46:39] * EnTeQuAk has joined #trac
[07:47:08] * SlashV has left #trac
[07:50:05] * prologic__ has joined #trac
[07:50:18] * prologic_ has quit IRC
[08:19:05] * CrisAWAY is now known as Crissi
[08:20:56] * zoli2k has quit IRC
[08:26:25] * tinh has joined #trac
[08:30:23] * __t has joined #trac
[08:46:49] * yoshi has joined #trac
[08:48:29] * yoswink has quit IRC
[08:56:54] <rblank> @
[09:06:05] <coderanger> !
[09:07:22] * ricky has quit IRC
[09:08:33] * Tetraboy has joined #trac
[09:09:53] * tinh has quit IRC
[09:10:46] * ricky has joined #trac
[09:14:43] * Shadowhand has joined #trac
[09:14:49] <Shadowhand> hello
[09:15:08] <Shadowhand> i'm about to attempt a trac relocation/upgrade from 0.10 to 0.11
[09:15:18] <Shadowhand> any specific docs on the best way to approach this?
[09:16:16] <coderanger> install it, run trac-admin upgrade, go back to sleep
[09:16:21] <coderanger> (that last part is optional)
[09:17:29] <Shadowhand> "install it" meaning the database?
[09:17:45] <coderanger> No, install trac
[09:18:10] <Shadowhand> that's already done
[09:18:25] <Shadowhand> more specifically, i want to know which files to move from the old install to the new install
[09:18:29] <Shadowhand> they are on different servers
[09:18:48] <coderanger> the trac environment folder
[09:19:18] <coderanger> and the database if you aren't using sqlite and want to move it to a new machine, or just adjust the server name to point at where the DB is
[09:19:45] <Shadowhand> ok
[09:22:21] <Shadowhand> if i copy db/ and attachments/, then run trac-admin upgrade, that will be enough?
[09:23:30] * cmlenz has joined #trac
[09:23:38] <coderanger> No
[09:23:45] <coderanger> Copy the trac env folder
[09:24:08] * parti has joined #trac
[09:24:34] <Shadowhand> the entire thing?
[09:24:37] <Shadowhand> i don't want to do that
[09:24:46] <coderanger> Why not, it should be tiny
[09:24:51] <Shadowhand> templates with 0.10 and 0.11 aren't compatible
[09:24:58] <coderanger> So
[09:25:04] <Shadowhand> why can't i just copy db and attachments?
[09:26:26] <coderanger> Because you 1) also need the conf/ folder, 2) some plugins store things in other folders in the env, 3) probably want to include your plugins, 4) you need the VERSION file for Trac to work, 5) you probably want the htdocs folder if you have anything in it (like the logo)
[09:26:47] <coderanger> Now that sentence can be expressed much faster if I just say "Copy the env folder"
[09:27:08] <Shadowhand> 1) no, i don't, 2/3) no, i already reinstalled the plugins, 4) ok, i'll copy that too, 5) i have nothing in htdocs
[09:27:41] <coderanger> Why in the name of all that is good are you arguing this?
[09:28:05] * retracile mumbles something unintelligable about mornings.
[09:28:13] <coderanger> Are you charged by the byte or something?
[09:28:17] <Shadowhand> because i've already spent a lot of time configuring the new trac and installing the plugins i want
[09:28:21] <retracile> It's still morning? Why is it still morning?
[09:28:33] <Shadowhand> the old trac is fairly out of date
[09:28:45] <coderanger> retracile: Because its a saturday, which makes morning okay
[09:28:47] <Shadowhand> i just want the minimum files that i need to copy to run trac-admin upgrade
[09:29:08] <retracile> coderanger: heh. I'm not sure I completely agree, but I won't argue. ;)
[09:29:18] <coderanger> Shadowhand: So try it your way, see if anything breaks, and stop asking for help when you don't actually care what the answer is
[09:29:33] <coderanger> Assuming you are right that you don't need anything else, it will work fine
[09:29:58] <Shadowhand> thanks, good enough for me
[09:31:31] * coderanger wonders why he bothers
[09:31:59] <Shadowhand> hey man, i'm not trying to bust balls... like i said, i already spent a lot of time configuring things
[09:32:06] <Shadowhand> it would really suck to lose all that time
[09:32:36] <Shadowhand> slightly OT, but what's the sqlite3 command to check the character set of the database?
[09:33:29] <coderanger> there isn't one
[09:34:31] <Shadowhand> k
[09:34:54] <coderanger> trac normalizes everything to utf8 in the DB (and anywhere we store things)
[09:35:49] <Shadowhand> pysqlite2.dbapi2.DatabaseError: file is encrypted or is not a database
[09:35:59] <coderanger> sqlite version mismatch
[09:36:04] <coderanger> @wiki TracUpgrade
[09:36:04] <evil_twin> coderanger: http://trac.edgewall.org/wiki/TracUpgrade
[09:36:07] <Shadowhand> 2 v 3?
[09:36:16] <coderanger> or vice versa
[09:36:26] <rblank> @
[09:36:35] <rblank> Oops, wrong window again. Sorry.
[09:36:53] <Shadowhand> well, i'm 100% sure the new trac is sqlite3
[09:37:16] <s0undt3ch> Shadowhand: then user the right binary to open it
[09:37:21] <s0undt3ch> *then use
[09:37:28] <s0undt3ch> sqlite3 usually
[09:37:42] <Shadowhand> so, open with 2, dump and reload into 3?
[09:37:51] <coderanger> that would be what the page says
[09:38:07] <s0undt3ch> he he he
[09:38:47] <Shadowhand> that page says nothing about upgrading sqlite
[09:38:56] <Shadowhand> unless i've still got sleep sand in my eyes
[09:40:54] * tct13 has joined #trac
[09:44:57] <s0undt3ch> Shadowhand: http://trac.edgewall.org/wiki/0.10/TracUpgrade
[09:45:11] <s0undt3ch> Shadowhand: http://trac.edgewall.org/wiki/0.10/TracUpgrade#From0.8.xto0.9
[09:47:53] <Shadowhand> already got it, thanks
[09:48:04] <Shadowhand> google to the rescue ;)
[09:53:13] * RobinD has joined #trac
[10:36:29] * |igor_ has joined #trac
[10:40:34] <ldlework> does anyone know why I might be getting this error in apache: http://ldlework.pastebin.com/m5661e53e
[10:41:45] <Shadowhand> ldlework: compat doesn't exist in your trac.util
[10:43:04] <ldlework> okay
[10:43:10] <ldlework> what is compat and where do I get it?
[10:43:18] <Shadowhand> that i don't know
[10:43:37] <s0undt3ch> ldlework: compat is a 2.3 compatabiliy module trac implements
[10:43:48] <ldlework> is it not in the newest verison?
[10:44:01] <ldlework> I have python 2.5 installed
[10:44:05] <ldlework> iwth 0.11.1
[10:44:07] <ldlework> trac
[10:44:17] <s0undt3ch> ldlework: for < 0.12 should be present
[10:44:43] <ldlework> how can I easilly confirm the version of trac I have installed?
[10:45:15] <s0undt3ch> ldlework: tracd -V ?
[10:45:31] <s0undt3ch> ldlework: tracd --version
[10:45:45] <ldlework> hmm wierd
[10:45:47] <ldlework> I do
[10:45:53] <ldlework> from python
[10:45:58] <ldlework> from trac.util import python
[10:45:59] <ldlework> err
[10:46:00] <ldlework> compat
[10:46:01] <ldlework> and it works
[10:46:16] <s0undt3ch> ldlework: python -c "from trac import __version__; print __version__"
[10:46:30] <ldlework> 0.11.1
[10:46:41] <s0undt3ch> ldlework: ls #
[10:46:42] <s0undt3ch> /usr/lib/python2.5/site-packages/Trac-0.11.1-py2.
[10:46:42] <s0undt3ch> #
[10:46:45] <s0undt3ch> errr
[10:47:11] <s0undt3ch> ldlework: ls /usr/lib/python2.5/site-packages/Trac-0.11.1-py2.5.egg/trac/util
[10:47:35] <ldlework> compat.py
[10:47:55] <s0undt3ch> ldlework: dunno why it won't find it
[10:47:55] <ldlework> wierd I wonder why it can't find it
[10:48:18] <Shadowhand> thanks for the help everyone, trac has been successfully upgraded
[10:48:21] * Shadowhand has left #trac
[10:48:42] <s0undt3ch> One more happy customer
[10:49:11] <ldlework> s0undt3ch, what I have attempted to do
[10:49:21] <ldlework> Is migrate a trac env from one machine to another
[10:49:55] <ldlework> http://gnosis.unnamedforums.com/
[10:50:00] <ldlework> Its like partially working
[10:50:10] <ldlework> But something is wrong
[10:50:18] <ldlework> The nav is mostly missing
[10:51:30] <ldlework> where are all the modules?
[10:51:31] * __t has quit IRC
[10:51:42] <s0undt3ch> ldlework: I don't have perms to see it
[10:51:58] <ldlework> gnosis gnosis
[10:52:33] <s0undt3ch> ldlework: Iternal Server Error ;)
[10:52:39] * izibi_ is now known as izibi
[10:53:07] <s0undt3ch> ldlework: did you looose any perms on migration?
[10:53:16] <ldlework> ah
[10:53:17] <ldlework> thanks!
[10:53:19] <ldlework> that worked
[10:53:21] <ldlework> <3
[10:53:24] <ldlework> 2 happy customers!
[10:54:08] <s0undt3ch> ldlework: remove ADMIN perms to that user you just passed me :)
[10:54:30] <s0undt3ch> ldlework: annonymous don't have perms on purpose?
[10:55:01] * handyande has quit IRC
[10:55:08] <s0undt3ch> ldlework: what's that theme thing on your admin panel?
[10:55:32] <s0undt3ch> ldlework: ThemeEngine?
[10:55:36] <ldlework> Yeah
[10:55:44] <ldlework> This is messed up
[10:55:51] <ldlework> I'm getting internal server erros randomly
[10:56:32] <ldlework> its randomly failing to import compat
[10:56:33] <ldlework> ...
[10:58:09] <ldlework> could it be
[10:58:17] <ldlework> because all the plugins that are installed into trac
[10:58:21] <ldlework> arent installed into python?
[11:01:20] <ldlework> so wierd
[11:02:37] <s0undt3ch> ldlework: that's the only trac install you have?
[11:02:56] <s0undt3ch> ldlework: you could make sure python patch is pointing to the right spot
[11:03:09] <ldlework> python patch?
[11:03:33] <s0undt3ch> ldlework: no, you're using mod_wsgi right?
[11:05:54] <s0undt3ch> ldlework: sys.path.insert(0, path_to_the_trac_you_are_using) although since it's on site-packages that shouldn't be needed
[11:14:00] * igor has joined #trac
[11:14:05] * yoshi is now known as yoswink
[11:23:36] <ldlework> How can I setup authentication to be linked to either the local linux users or the svn users?
[11:29:51] * Crissi has left #trac
[11:31:23] * |igor_ has quit IRC
[11:32:36] <s0undt3ch> ldlework: trac-hacks.org is most likely your friend ;)
[11:40:48] <retracile> ldlework: might check accountmanager
[11:49:38] * sdboyer_ has joined #trac
[11:49:51] * pete__c has quit IRC
[11:55:51] * __t has joined #trac
[11:55:52] * EnTeQuAk has quit IRC
[12:00:46] * igor has quit IRC
[12:03:36] * chori has joined #trac
[12:06:54] * sdboyer has quit IRC
[12:16:53] * sdboyer_ is now known as sdboyer
[12:18:56] * sdboyer has quit IRC
[12:22:45] * sdboyer has joined #trac
[12:33:30] * bkuhn is now known as bkuhnIdle
[12:34:55] * yoswink has quit IRC
[12:36:39] * pete__c has joined #trac
[12:42:13] <elb> if any traccers are around ... Pidgin could really use some help, still
[12:42:30] <elb> 0.11 is still unbearably slow for us
[12:46:14] <elb> this time at least it's honest CPU usage, not strange artifacts :-)
[12:47:36] * pete__c has quit IRC
[12:47:48] <coderanger> elb: The profile file you sent me doesn't seem to be valid
[12:47:59] <elb> coderanger: I'm glad I'm not the only one who can't load it!
[12:48:22] <coderanger> can you pastebin the profiler fcgi script as you have it now
[12:48:25] * coderanger has idea
[12:49:19] <elb> http://pidgin.im/nopaste/114
[12:53:43] * freaky[t] is now known as fReAkY[t]
[12:54:04] * pete__c has joined #trac
[12:54:29] <coderanger> http://paste.pocoo.org/show/83123/
[12:54:31] <coderanger> Try that
[12:55:05] <elb> is that going to fill my /tmp up with profiles? ;-)
[12:55:54] <coderanger> probably :-)
[12:56:01] <elb> not really ap roblem
[12:56:07] <coderanger> you only need it running for a few seconds
[12:56:09] <elb> especially if it helps fix this
[12:58:03] <elb> OK, I have a million .prof files
[12:58:15] <elb> and an equally large number of .goodprof files ;-)
[12:58:15] <coderanger> Any .goodprof?
[12:58:29] <coderanger> okay, pick one or 10 and send them on over :)
[12:58:32] <elb> actually, a lot more goodprof files
[12:59:20] <coderanger> Quick check, how many different pids.
[12:59:26] <coderanger> It should be a low number
[12:59:30] <elb> 4
[12:59:38] <coderanger> okay, that sounds about right
[12:59:39] <elb> which is correct
[12:59:43] <elb> our config is set up to run 4
[13:03:24] <elb> sent
[13:04:05] <elb> woo and these load
[13:04:38] <elb> although it's not clear ot me some operations werne't failing, which may skew results
[13:05:03] <coderanger> it was overwriting the profiles constantly, so at any given time before you would get a half-finished one
[13:05:19] <coderanger> I should just stick that script on my site so I don't keep rewriting each time
[13:07:36] <coderanger> hmm, I wonder if you can merge stat files somehow
[13:07:53] <coderanger> since the timing on a single request is so useless
[13:08:14] <elb> yeah
[13:08:17] <elb> I just looked at some
[13:08:31] * coderanger plays
[13:08:43] <elb> I can tar up *all* the goodprofs for you if you want
[13:08:45] <elb> I just picked some
[13:09:02] <coderanger> lemme figure out how/if to merge them
[13:09:17] <elb> I left out the earliest ones, because they likely include the permissions caching
[13:09:20] <coderanger> IIRC its marshall data
[13:09:24] <elb> which is really expensive but only happens once
[13:11:56] <coderanger> oh nice, the hotshot log parser is written partially in the C code
[13:11:58] <coderanger> brilliant
[13:12:16] <coderanger> okay, this seems possible at least, I'll poke at it for a bit
[13:12:21] <elb> excellent
[13:12:46] <elb> because our server load in pages/sec has fallen to like 1/2 what it was at the 2.5.0 release
[13:12:54] <elb> and it's still pretty unbearable
[13:13:22] <coderanger> How is memory usage (read: is it hitting swap)
[13:15:51] <elb> no
[13:15:58] <elb> ~600MB free
[13:16:06] <elb> 800+MB of disk cache
[13:16:09] <coderanger> okay
[13:16:45] <coderanger> when they ever get around to release the next 2.5 it should drop our steady state memory use by a good bit
[13:17:03] <elb> we're still running 2.4
[13:17:26] <elb> I should say ... most of the trac processes probably have swapped out junk
[13:17:33] <elb> their VSS is 4x RSS
[13:17:38] <elb> but they don't seem to be *hitting* it
[13:17:48] <elb> because they manage to consume 100% CPU ;-)
[13:29:31] * eikonos has joined #trac
[13:29:59] <coderanger> okay
[13:30:18] <coderanger> http://paste.pocoo.org/show/83125/
[13:30:29] <coderanger> That should make a single, merged profile file
[13:36:08] <coderanger> elb: Just from the 11 requests I have, it looks like you may be hitting threading issues
[13:36:20] <coderanger> try bumping the number of FCGI procs perhaps?
[13:37:07] <elb> hmm, it's been higher
[13:37:15] <elb> but I'll give that a shot
[13:37:46] <elb> would 8 be unreasonable?
[13:37:49] <coderanger> threading:acquire is second only to the main request function in terms of time
[13:38:03] <coderanger> its all based on how much ram you have
[13:38:07] <elb> right -- but doesn't that mean something is *doing* something in those other threads?
[13:39:08] <coderanger> It means two threads are trying to access the same thing, in one of the few places we have explicit mutexing
[13:39:23] <elb> and how will more threads make that better?
[13:39:33] <coderanger> not more threads, more procs
[