**** BEGIN LOGGING AT Sun Jul 16 02:59:57 2006 Jul 16 03:01:24 what's the trick to do (unsupported, I know) SMP builds? Jul 16 03:30:54 jacques: i'd imagine OEMAKE_EXTRA += "-j2" would do it for a single .bb file, but i'm guessing you mean the entire tree Jul 16 03:31:11 * zwelch wonders if you could just add that to base.bbclass and be off Jul 16 03:33:36 zwelch, yeah, I meant the entire tree. hi BTW Jul 16 03:33:49 * zwelch grins and waves Jul 16 03:34:08 well, let me know if the base.bbclass hack works Jul 16 03:34:15 i can't see why it wouldn't Jul 16 03:34:20 I thought there was a nicer way Jul 16 03:34:28 but i am still fairly shortsighted when it comes to OE/bb Jul 16 03:34:40 there probably is; i'm just trying to help :) Jul 16 03:35:01 that, and say hi in a topic-relevant manner ;) Jul 16 04:03:00 jacques, I remember seeing something about parallel builds in the local.conf.sample file Jul 16 04:04:46 cinix, looking now Jul 16 04:06:46 # PARALLEL_MAKE = "-j 4" Jul 16 04:06:52 cool :-) Jul 16 04:07:49 I'll let this build complete to get a time, then try with that set (to -j 3) Jul 16 04:11:32 i belive there are still issues with Parallel Jul 16 04:11:56 and i suspect its not supported at all for 64 bit hosts Jul 16 04:13:26 even gentoo has issues with it - I found out tonight that netpbm won't build parallel (in gentoo) Jul 16 04:15:43 but even just building glibc and gcc parallel saves a *lot* of time Jul 16 04:15:49 and the kernel Jul 16 04:24:21 well unless your building for x86, you can expect a lot more Jul 16 04:25:08 why does the arch matter? Jul 16 04:25:23 support and maturity of development Jul 16 04:28:40 where can i download kernel for arm? arm.kernel.org? Jul 16 04:28:43 * thejapa hides Jul 16 04:29:12 thejapa, i hope your joking Jul 16 04:29:17 sorry couldn't resist remembering yesterday Jul 16 04:32:55 hrm ... i just need a simple connector ... Jul 16 04:54:08 PARALLEL_MAKE doesn't seem to have any effect Jul 16 04:55:21 well, there's always the sledgehammer approach that i suggested ;) Jul 16 04:55:39 you're probably better off figuring out what PARALLEL_MAKE doesn't work though Jul 16 04:56:33 ah, well it did nothing for gettext, but it's working for binutils Jul 16 04:56:58 heh, right, looking at its usage, i'd be flumoxed if it didn't: Jul 16 04:56:58 maybe they have a whitelist Jul 16 04:57:16 conf/bitbake.conf:EXTRA_OEMAKE_prepend_task_do_compile = "${PARALLEL_MAKE} " Jul 16 04:57:18 conf/local.conf.sample:# PARALLEL_MAKE = "-j 4" Jul 16 04:57:37 it's that simple, from what i can tell Jul 16 04:57:47 so, it depends on whether or not the .bb file uses oemake Jul 16 05:00:23 hmm Jul 16 08:24:40 morning Jul 16 08:43:52 03koen 07org.oe.dev * rbb3f0620... 10/conf/machine/include/tune-arm920t.conf: tune-arm920t: add -mcpu=arm920t since gcc will happily generate armv5 isns (clz, bx) when using -march=armv4t, spotted by Lennert Buytenhek Jul 16 09:30:50 03koen 07org.oe.dev * rd0938254... 10/packages/librsvg/librsvg_2.11.1.bb: librsvg: disable mozilla-plugin since mozilla isn't in DEPENDS, should close #1178 Jul 16 10:15:57 morning folks Jul 16 10:16:04 hey Jul 16 10:24:31 yo zecke Jul 16 10:24:45 yo pb Jul 16 10:24:54 yo woglinde Jul 16 10:25:24 ah, how tedious, apt has upgraded my monotone to 0.26 and 0.25 is no longer in the debian repository Jul 16 10:25:43 pb hm in unstable thats since 3 months Jul 16 10:25:48 pb_: that is not THAT bad Jul 16 10:26:21 zecke: true, there is a silver lining. if I can't pull at all, then at least I don't have to wait around for it to apply the revs. Jul 16 10:26:41 pb_: right and within 24 hours the services will shutdown Jul 16 10:26:43 hail pb_ Jul 16 10:26:47 pb_: and OE will be discontinued Jul 16 10:26:48 long time no see, where did you hide? Jul 16 10:27:06 hail mickey|thesis Jul 16 10:28:10 yeah, long time indeed. Jul 16 10:28:16 zecke: ah. Jul 16 10:28:33 pb_: good timing for apt. as you know we'll upgrade to .27 over the next days, you're well prepared to start contributing again ;) Jul 16 10:28:40 pb_: on Monday we will shutdown monotone.vanille.de Jul 16 10:28:52 mickey|thesis: heh. very good. Jul 16 10:28:57 pb_: then we crunch the numbers Jul 16 10:29:07 pb_: yeah right, try to get a momotone 0.27 Jul 16 10:29:20 debian unstable should have 0.27 Jul 16 10:29:34 zecke: okay Jul 16 10:30:07 I will be away on vacation from Wednesday for about ten days, so I guess the new 0.27 repository should be working when I get back. Jul 16 10:31:02 pb_: right :) Jul 16 10:31:08 pb *g* Jul 16 10:31:20 I will clone zeke 50 times Jul 16 10:31:25 and we will see Jul 16 10:31:31 pb_: sounds good. same here starting on Thursday. We're you going to? Jul 16 10:31:39 50 handicapped bastards, that sounds nice Jul 16 10:32:01 woglinde: heh. all fear the mad scientist Henning Jul 16 10:35:31 mickey|thesis: somewhere in France.; Dordogne kind of area I think. Jul 16 10:36:06 pb_: have fun :) Jul 16 10:36:27 zecke: I'll try Jul 16 10:37:32 mickey|thesis: how's your thesis, by the way? Jul 16 10:37:48 * zecke considers going to canossa, to switch back to emacs Jul 16 10:39:15 pb_: on tuesday I'll present a first release candidate to my advisor. until now he just got it chapter wise once and then. now i'm giving him a first state in which i'd say "if you don't object, please just let me submit that damn thing" :) Jul 16 10:41:04 i don't expect him letting me go with that, but my hope is to be able to submit in august then. Jul 16 10:41:35 the sooner you submit, the sooner you'll have your shiny new macbook Jul 16 10:41:44 that's right. that is #1 motivation ;) Jul 16 10:41:49 mickey|thesis: ah, very good. hope your advisor is suitably impressed. Jul 16 10:43:02 * mickey|thesis hopes as well Jul 16 10:43:09 this advisor has a slight problem Jul 16 10:43:13 he is very good at criticizing Jul 16 10:43:19 he has a problem in doing the opposite Jul 16 10:43:27 and no, it's not just me :D Jul 16 10:43:35 he can't say "that's cool so far" to anyone Jul 16 10:44:11 he knows how to motivate a team Jul 16 10:44:31 ya Jul 16 10:45:36 guess that's my problem for choosing an advisor that's about to retire soon Jul 16 10:45:41 *shrug* Jul 16 10:48:46 heh Jul 16 10:48:48 oh well Jul 16 10:49:27 i'm going to give a party that this town hasn't seen in the last couple of years when I'm through this mill Jul 16 10:49:38 :-) Jul 16 10:50:05 mickey|thesis: better than the WM? Jul 16 10:50:22 well... not quite in this dimension :) Jul 16 10:57:14 * mickey|thesis just sent read-only reminder Jul 16 10:59:17 good idea. it'd be silly to let people write to your reminders. Jul 16 10:59:51 hehehe Jul 16 11:00:00 point taken. i need to improve my grammar :D Jul 16 11:00:42 btw., i'm still looking around for a english native speaking co-author for the forthcoming OE book... Jul 16 11:02:02 heh. actually, your grammar was fine in that case. English is just ambiguous like that. Jul 16 11:02:47 *g* Jul 16 11:02:57 heh. i like that invisible stressing "in that case" :D Jul 16 11:03:03 :-) Jul 16 11:04:59 * mickey|thesis goes back to doing the nickname suffix a bit more justice Jul 16 11:05:23 enjoy Jul 16 11:05:29 thanks ;) Jul 16 11:05:33 * pb_ goes to water the garden Jul 16 11:05:45 hehe Jul 16 11:05:53 pb_: I'm not asking about the rain ;) Jul 16 11:06:04 zecke: heh. it hasn't rained here for some days. Jul 16 11:06:20 almost a week, actually, if I remember right Jul 16 11:06:39 wow Summer in Cambridge must be incredible Jul 16 11:07:23 http://www.wunderground.com/global/stations/03571.html Jul 16 11:07:35 looks like it might rain on Thursday Jul 16 11:07:41 hey Jul 16 11:18:15 03koen 07org.oe.dev * refa06284... 10/conf/machine/include/tune-arm920t.conf: tune-arm920t: set PACKAGE_ARCH to armv4t Jul 16 11:50:38 03coredump 07org.oe.oz354x * r01514cf5... 10/packages/alsa/ (3 files in 3 dirs): alsa: Add alsa-settings: Default ALSA preferences and pref-loading for non-clamshell devices Jul 16 11:57:05 zecke: for stuff like this, taking the libtool command and adding --dry-run --debug helps a lot Jul 16 11:57:21 ah :) Jul 16 11:57:38 I'm still compiling away, after my build failed unnoticed due binutils 2.17 Jul 16 11:58:19 I just got glibmm to finally package with manual hacking Jul 16 11:58:50 the problem was, once the glibmm populate step failed, it corrupted the work dir Jul 16 11:59:13 so I had to fix libsigc++'s .la *before* the entire glibmm build Jul 16 12:00:15 any idea how to have it unbroken automatically? Jul 16 12:00:26 I'm struggling with lack of sleep and failing builds Jul 16 12:00:31 I think it is time to get a new disk Jul 16 12:00:43 well, glibmm's libtool looks to work ok-- strange that the build system's ld.so.conf gets in the config for it, tho Jul 16 12:00:51 but that's the case with all the other libtools in my work dir Jul 16 12:01:17 so it's just a matter of making libsigc++'s .la lack x86_64-linux/lib Jul 16 12:01:31 not sure how it gets there, though Jul 16 12:05:10 lrwxrwxrwx 1 ich ich 16 2006-07-16 13:50 libesd.so -> libesd.so.0.2.28 Jul 16 12:05:11 lrwxrwxrwx 1 ich ich 16 2006-07-16 13:50 libesd.so.0.2.28 -> libesd.so.0.2.28 Jul 16 12:05:19 can anyone share this symptom? Jul 16 12:05:28 ROFL Jul 16 12:05:34 zecke: do13 had Jul 16 12:05:46 ah fine Jul 16 12:05:48 I suspect that means you had one in staging when libesd trying to populate Jul 16 12:06:00 so it wrote *through* the symlink for the real one Jul 16 12:06:05 and overwrote that with a link Jul 16 12:06:12 koen: I wonder what happens if I use oe_libinstall properly Jul 16 12:06:39 * Luke-Jr 'd nuke em both and rm the populate_staging stamp Jul 16 12:07:03 Luke-Jr: will do, after fighting the MAD fcl Jul 16 12:17:42 later Jul 16 13:38:08 Morning Jul 16 13:38:48 Looking for help setting up OE Jul 16 13:40:10 Followed Wiki steps, get following message trying to build nano Jul 16 13:40:22 NOTE: Using cache in '/OE/build/tmp/cache' Jul 16 13:40:33 ERROR: no files to build. Jul 16 13:41:06 Trying to figure out why sources aren't being downloaded.. Jul 16 13:41:56 Set sources directory in local.conf.. yet not pulling down sources Jul 16 13:42:15 bitbake isn't supposed to download anything Jul 16 13:42:22 jerrybme, is your BBPATH setup right ? Jul 16 13:42:32 Erm, actually it is. Jul 16 13:42:41 Sorry, got confused for a moment. Jul 16 13:42:48 Yes, doubled check but will triple check now.. Jul 16 13:42:53 I suppose you have downloaded the bb files with monotone? Jul 16 13:43:03 yep Jul 16 13:43:45 And BBFILES is like this: "${HOME}/src/OpenZaurus/org.openembedded.dev/packages/*/*.bb" Jul 16 13:43:48 Dr. Who: export BBPATH=/OE/build:/OE/org.openembedded.dev Jul 16 13:44:38 I have those in the reverse order but I _think_ it shouldn't affect this... Jul 16 13:45:01 yup looks right Jul 16 13:45:03 How about the BBFILES in local.conf? Jul 16 13:45:30 yes order is irrelevant Jul 16 13:45:40 DataBeaver: BBFILES:= "?OE/org.openembedded.dev/packages/*/*.bb Jul 16 13:45:48 thats wrong Jul 16 13:45:53 why do you have a ? Jul 16 13:46:11 poor typing ;) Jul 16 13:46:23 it's not in local.conf Jul 16 13:46:57 it nees to be Jul 16 13:47:00 needs* Jul 16 13:47:26 BBFILES:= "/OE/org.openembedded.dev/packages/*/*.bb is in local.conf Jul 16 13:47:53 I hope you have the latter quote as well there... Jul 16 13:48:01 you have OE at the root of your FS? Jul 16 13:48:01 my poor typing is not in local.conf.. Jul 16 13:48:25 yes ending " in there Jul 16 13:48:47 emte: yes /OE is at root Jul 16 13:48:57 i'll just mention that is a very bad place Jul 16 13:49:09 Why? Jul 16 13:49:20 doubly so if your using bitbake as root Jul 16 13:49:32 emte: nonsense Jul 16 13:49:44 (about the dir hierarchy) Jul 16 13:49:46 if one of the packages is hardcoded and things escape badness will happen Jul 16 13:51:01 jerrybme: /OE is fine Jul 16 13:51:01 What happens if you do bitbake -i and then type parse to the prompt? Jul 16 13:51:21 koen, then you have never notices any hardcodeded ../../../../../../../usr/lib type manglings by makefiles? Jul 16 13:51:46 Ugh, does someone really use stuff like that? Jul 16 13:51:51 emte: by your reasoning it wouldn't be safe to install OE at all Jul 16 13:51:58 i've run across a handful of things Jul 16 13:52:09 I never had troubles with /OE Jul 16 13:52:16 and I built 'world' a few times Jul 16 13:52:38 koen, i belive it isnt, but the new changes in the security system with chroot make it dificult to implement Jul 16 13:53:14 DataBeaver: same message as above: Note: Using cache in 'OE/build/tmp/cache' & Error: no files to build Jul 16 13:53:28 hmh... Jul 16 13:53:35 i know qemu was still putting garbage where it shouldent the last time i enabled it Jul 16 13:54:09 namely in /tmp Jul 16 13:54:19 jerrybme: your BBFILES is wrong Jul 16 13:54:45 emte: programs are encourages to put temporary files in /tmp Jul 16 13:54:49 hence the name Jul 16 13:54:59 /tmp Jul 16 13:55:11 as in /TeMPorary Jul 16 13:55:52 i set TMPDIR to use another location because i dont have the room on the main drive for garbage like that Jul 16 13:56:58 mine lives in build/tmp/temp Jul 16 13:58:13 koen: where's the error in my BBFILES? Jul 16 13:58:25 jerrybme, its not at the specified location Jul 16 14:00:43 emte: not clear to me the bb files are @ /OE/org.openembedded.dev/packages/ Jul 16 14:02:16 jerrybme: paste the env-var shell script in pastebin, as well as your local.conf Jul 16 14:02:35 ~pastebin Jul 16 14:02:37 methinks pastebin is a place to paste your stuff without flooding the channel - try http://pastebin.com/ (BROKEN AND SUCKING NUTS), or http://pastebin.ca, or http://channels.debian.net/paste, or for #oe use http://oe.pastebin.com, or http://bzflag.pastebin.ca/ Jul 16 14:03:31 and if pastebin is still useless use http://www.rafb.net/paste Jul 16 14:04:07 What's the correct virtual for kernel? Jul 16 14:04:44 virtual/kernel Jul 16 14:05:52 Thanks Jul 16 14:08:08 jerrybme: and triple check if BBPATH points the dir containing conf/local.conf Jul 16 14:08:42 if you use bash, tab is your friend Jul 16 14:08:53 and gpm too Jul 16 14:08:58 er Jul 16 14:09:00 gmp Jul 16 14:09:33 koen: http://www.rafb.net/paste/results/dcLYS991.html for OE env script Jul 16 14:10:32 and you local.conf is at /OE/build/conf/local.conf? Jul 16 14:11:34 Koen: that maybe the problem, it's @ /OE/build/local.conf Jul 16 14:11:52 my local.conf: http://www.rafb.net/paste/results/lGj8Lx65.html Jul 16 14:15:45 Well it needs to be under the conf directory Jul 16 14:16:09 Thanks koen et al: moving the local.conf to the right place & removing the spaces from 2 lines of the conf file it's compling nano now Jul 16 14:16:55 too bad it wasnt fam, then you coulda used the cool bitbake setup tools Jul 16 14:18:10 re Jul 16 14:19:45 New problem... I added virtual/kernel:linux-openzaurus-2.6.17 to PREFERRED_PROVIDERS, but it still wants to use 2.6.16 Jul 16 14:20:06 hm geode is dead Jul 16 14:20:31 DataBeaver: preferred_provider doesn't handle versions, only providers Jul 16 14:20:41 bah. Jul 16 14:20:45 lol Jul 16 14:20:58 Guess I'll have to mess with the .bb file then... Jul 16 14:21:12 ? Jul 16 14:21:21 PREFERRED_VERSION_linux-openzaurus = "2.6.17" Jul 16 14:21:39 Ah, of course. Jul 16 14:25:59 wheres the yellow wire ? Jul 16 14:27:27 no take the red one Jul 16 14:28:11 lol Jul 16 14:28:25 signal wires for an spi bus, reds not an option Jul 16 14:28:37 *g* Jul 16 14:28:51 tought you have a bomb or soemthing Jul 16 14:30:41 well if i miswire it ... something might explode] Jul 16 15:32:09 Having trouble getting nano to compile see: http://www.rafb.net/paste/results/Q6It7K46.html Jul 16 15:42:05 jerrybme: your OE tree is too old Jul 16 15:48:08 koen: mm I just did a: monotone --db=/OE/OE.db pull monotone.openembedded.org org.openembedded.dev & checkout yesterday Jul 16 15:48:19 yes Jul 16 15:48:24 like I said, it's too old Jul 16 15:48:29 I'll do again Jul 16 15:48:40 allways pull and updates before reporting an error Jul 16 15:50:05 OK, thanks for your patience with newbies like me.. Jul 16 15:55:45 uh Jul 16 15:59:52 hi minipanda Jul 16 16:00:21 woglinde: ooops Jul 16 16:00:48 ooops? Jul 16 16:01:02 i dont oops only my kernels sometimes Jul 16 16:02:17 woglinde: hehe, i have a basic patch and .bb file for LTP, Jul 16 16:02:37 woglinde: so is that possible for checkin to OE.db in near future? Jul 16 16:04:44 minipanda: attach it to the bugtracker and someone will have a look at it Jul 16 16:05:09 minipanda but beware, my msp patches didnt go in Jul 16 16:05:53 aware Jul 16 16:05:56 hms Jul 16 16:07:02 woglinde: kick zecke for the msp stuff Jul 16 16:07:37 no Jul 16 16:07:46 I will beg for dev account Jul 16 16:07:52 after the term Jul 16 16:07:59 zecke has enoigh todo Jul 16 16:23:02 morning all Jul 16 16:23:18 Morning Jul 16 16:37:22 hey RP Jul 16 16:38:06 jerrybme: and pull/update as often as possible, or else when you do it will take hours Jul 16 16:38:41 jerrybme: Daily is probably best.. you could do it every week if you want it to take ~1-2hr with monotone 0.25 Jul 16 16:39:25 not really Jul 16 16:39:25 or you could take a vacation from OE and take a day when you get back :) Jul 16 16:39:37 there's only 5 hours left to check in changes using monotone 0.25 Jul 16 16:40:23 after that no more changes can go in Jul 16 16:40:49 oh, so the migration to 0.27 is happening before 0.27 is actually available? =p Jul 16 16:40:50 after that monotone 0.26? Jul 16 16:41:07 * Luke-Jr blames Gentoo Jul 16 16:41:09 * koen points to http://openembedded.org Jul 16 16:41:27 Luke-Jr: debian installed 0.27 last week on my machine Jul 16 16:41:54 koen: it is impossible to get 0.27 with Gentoo Jul 16 16:42:12 * koen doesn't use gentoo Jul 16 16:42:22 I have better things to do than being 1337 Jul 16 16:43:23 gentoo isn't leet anymore Jul 16 16:43:29 gentoo has gone down the drain in the past two years Jul 16 16:43:38 it's now a fucking bunch of crap Jul 16 16:44:20 giel: indeed, but still better than most of the stuff out there Jul 16 16:44:32 I'm still using it because my replacement isn't complete Jul 16 16:48:52 Hmm.. is anyone else having problems building glibc-2.3.5? Tried settting the preferred version to 2.4 in local.conf, no luck.. seems unavailable. Jul 16 16:48:59 Er, that's off .dev. Jul 16 16:51:48 * koen points to http://www.openembedded.org/wiki/OeFaq Jul 16 16:52:41 koen: Thankye Jul 16 17:22:53 NOTE: package libxxf86dga-1.0.1-r0: task do_configure: failed Jul 16 17:22:53 -- | configure: error: Package requirements (xproto x11 xextproto xext xf86dgaproto) were not met: Jul 16 17:23:13 missing dependencies (or maybe just misbehaving bitbake) Jul 16 17:24:26 in this case, xf86dgaproto wasn't built Jul 16 17:25:10 looks like a missing DEPENDS Jul 16 17:25:51 file a bug so people won't forget Jul 16 17:26:29 aww, why not just do it? =p Jul 16 17:27:55 * koen just forgot Jul 16 18:25:11 hi, is that possible for me to get return value from cmd operate like "bitbake -c fetch bootstrap"? Jul 16 18:32:07 hi Jul 16 18:32:43 the ${PATH} variable links to the native staging directory bin Jul 16 18:33:23 but I need to execute a binary from the target system Jul 16 18:33:48 (directfb-config), in order to get information about linkage etc Jul 16 18:34:05 how can I add something to the start of $PATH Jul 16 18:34:31 I tried with PATH =+, but that doesn't seem to add it to the very beginning Jul 16 18:34:48 and the native staging bin dir always seems to come first Jul 16 19:12:25 tkp: 'inherit binconfig' (iirc) Jul 16 20:05:22 mickey_away: hey Jul 16 20:54:24 NOTE: 1h to read-only. Jul 16 20:55:58 * koen buys mickeyl a new watch Jul 16 20:56:29 actually i don't wear watches Jul 16 20:56:53 can't have anything around my wrist Jul 16 20:57:00 feels uncomfortable Jul 16 20:58:03 so that wristmounted pda thingy isn't for you Jul 16 20:59:04 definitly not. If anything, I'd use a startrek ng communicator thing Jul 16 20:59:17 mickeyl: did you happen to see the discussion about openzaurus-unstable becoming an 'insane' configuration for bitbake, after your recent config changes? was i right to conclude that you did so intentionally, with the point being to force people to realize they're dealing with an "unsupported" configuration? Jul 16 20:59:35 zwelch: no, it was an accidant Jul 16 20:59:46 okay :) Jul 16 20:59:53 which i fixed by giving sanity.bbclass a second chance to succeed in verifying DISTRO Jul 16 20:59:55 well, i gave you the benefit of the doubt :) Jul 16 20:59:59 :) Jul 16 21:00:25 so i can probably kill my local sanity.conf now, right? Jul 16 21:00:52 .conf or bbclass? Jul 16 21:00:53 * zwelch will likely test it when his current build finishes Jul 16 21:01:12 well, with an 'insane' config, bb tells you to touch sanity.conf Jul 16 21:01:17 oh Jul 16 21:01:21 never did that :) Jul 16 21:01:26 i touched build/conf/sanity.conf, and *poof* it started working again Jul 16 21:01:36 yeah, it took me an hour just to read the error it was giving me Jul 16 21:01:44 hah Jul 16 21:02:00 in point of fact, it raises an exception caused by not importing sys.exit Jul 16 21:02:23 i fixed that in bitbake as well :D Jul 16 21:02:27 the exception has the effect on my brain to mask the informative error message it tried to print Jul 16 21:02:32 12:40 : (not that our devs excel in reading error messages....) Jul 16 21:02:35 *nod* tracebacks are not for users... Jul 16 21:02:35 sweet :) Jul 16 21:02:58 users being developers in OE, but anyway. I prefer nice errors Jul 16 21:03:03 koen: i read them... but as mickeyl just said, it helps to give users the proper message Jul 16 21:03:54 zwelch: I made that comment this afternoon about people complaining about not being able to push after we go read-only Jul 16 21:04:06 * zwelch rolls his eyes Jul 16 21:04:11 it's for what, a week? Jul 16 21:04:23 something like that Jul 16 21:05:27 * zwelch tries to come up with a witty observation about how the world has lost the notion of patience in the age of the internet, but fails Jul 16 21:07:42 incidentally, i don't remember seeing one when i grabbed my copy, but what are the odds that someone will seed a torrent to distribute converted DB? Jul 16 21:08:39 or was i mistaken in the impression that it will be necessary to 'start fresh' in that way? Jul 16 21:08:53 * zwelch goes and reads the migration document again Jul 16 21:09:36 a torrent would be much of an improvement, since we'd have daily snapshots Jul 16 21:09:50 so you'd need to switch to a new torrent everyday Jul 16 21:11:22 i'm thinking more about the initial re-launch Jul 16 21:11:28 03coredump 07org.oe.oz354x * redd003af... 10/conf/machine/poodle-2.6.conf: poodle-2.6.conf: Add alsa-settings Jul 16 21:11:30 i.e. immediately post-migration Jul 16 21:11:50 ... when everyone will want a copy, and when the p2p effect will be most effective Jul 16 21:12:18 also, if pulls are that much faster, it may be more economical on bandwidth to publish less frequently than a daily snapshot Jul 16 21:12:25 we have bandwidth to spare Jul 16 21:12:35 heh, then it doesn't matter much Jul 16 21:13:05 2x 100Mbit boxen with unlimited traffic Jul 16 21:13:21 anyway, i was just thinking outloud. y'all seem to have put a lot of thought into this migration, and i'm sure it will go smoothly Jul 16 21:13:50 does OE have boxes at the OSL? Jul 16 21:14:00 OSUOSL, that is Jul 16 21:14:09 hey, are you planning on generating stats for migration downloads? it could tell how much ppl uses oe actively nowadays. Jul 16 21:14:19 no, at utwente.nl and strato.de Jul 16 21:14:29 ah Jul 16 21:14:44 well, i can get the project hosting at the OSL, if necessary Jul 16 21:14:53 i can probably toss a server in along with it Jul 16 21:14:59 * zwelch shrugs Jul 16 21:15:08 that would be nice as a long term plan Jul 16 21:15:28 well, the offer is here while i am ;) Jul 16 21:15:38 but they are easy to approach directly too Jul 16 21:16:01 * zwelch just happens to be a few blocks away from them, that's all Jul 16 21:16:49 anyway, thanks for the feedback; i'm off to play my mandolin Jul 16 21:19:31 why a week? I did a complex CVS->Svn in under a day Jul 16 21:20:07 Luke-Jr: and how did you verify everything worked as expected? Jul 16 21:20:08 see migrationplan Jul 16 21:20:26 We create a schedule, we announce it X times Jul 16 21:20:31 and then someone always knows better Jul 16 21:20:34 * zecke shakes head Jul 16 21:20:57 zecke: I did all that before the actual migration Jul 16 21:21:03 and someone always complains Jul 16 21:21:06 YOU ALL SUCK Jul 16 21:21:15 Luke-Jr: ah you tested the result Jul 16 21:21:23 Luke-Jr: before creating it? Jul 16 21:21:25 * zecke bows down Jul 16 21:21:47 :) Jul 16 21:21:48 I simply updated the result for the actual migration Jul 16 21:21:51 just kidding :) Jul 16 21:22:19 which reduced testing to a very small period of time Jul 16 21:22:20 Luke-Jr: Why does it take so long? Jul 16 21:22:30 ? Jul 16 21:22:31 Well: a) the rosterify itself takes 8 hours Jul 16 21:22:37 O.O Jul 16 21:22:54 b) testing viewmtn, mailing lists, checking stuff out, irc bots Jul 16 21:23:11 ah yes Jul 16 21:23:12 takes human resources, I do not know if they are allocatable Jul 16 21:23:27 'b' could have occurred on a copy prior to the actual migration =p Jul 16 21:23:28 I should diff viewmtn against mainline to see how much I changed Jul 16 21:23:35 then we might stumble a bug in mtn that delays another X time items Jul 16 21:24:22 Luke-Jr: well, testing a real system on real world load will always be different Jul 16 21:24:45 only the load amounts Jul 16 21:25:39 koen: you're here during the complete migration phase? Jul 16 21:25:42 why not have the mtn server cache a value for 'has had signature verified' and let clients skip verifying them when it's already been done? :p Jul 16 21:25:49 Luke-Jr: feel free to do the migration for us within a day Jul 16 21:26:00 Luke-Jr: I would be glad if things work out more soon Jul 16 21:26:06 mickeyl: as a said, I'm staying out of the migration biz Jul 16 21:26:18 zecke: my mtn usage is currently merely pull && update Jul 16 21:28:13 koen: hmm.... my vacation starts on 20th, so i'm afraid i will have to miss the last two days. Jul 16 21:29:15 koen: will we have a read-only svn/cvs mirror soonish after the migration? Jul 16 21:29:50 if Jul 16 21:29:54 a) people want one Jul 16 21:30:03 b) someone steps up to maintain it Jul 16 21:30:08 I want a read-write Svn mirror! Jul 16 21:30:09 a) absolutely Jul 16 21:30:28 :) Jul 16 21:30:40 the script is in ~/monotone/ Jul 16 21:30:42 b) shouldn't be more than a hourly cronjob if we're satisfied with 1h lag? Jul 16 21:30:48 heh Jul 16 21:30:55