**** BEGIN LOGGING AT Wed Jun 29 23:59:56 2005 Jun 30 00:08:29 jf-work: ping? :) Jun 30 00:09:39 pong Jun 30 00:10:04 <-- running one of the two first 266Mhz slugs :D Jun 30 00:10:44 Speeding Slug :) Jun 30 00:11:14 let me take caffeine. bbiab Jun 30 00:14:27 Any non-io stuff goes twice as fast Jun 30 00:16:09 SpeedSlug...Oxymoron Jun 30 00:16:13 hehe Jun 30 00:17:23 Timing buffered disk reads: 20 MB in 3.00 seconds = 6.67 MB/sec Jun 30 00:17:36 Not exactly impressive speed still Jun 30 00:18:21 Now, if I can figure out a way to test the network interface... Jun 30 00:23:42 urgh Jun 30 00:38:27 NAiL: I miss probably something. What is this speedslug ? Jun 30 00:38:42 An overclocked slug Jun 30 00:38:47 266mhz instead of 133 Jun 30 00:39:57 running openslug ? Jun 30 00:40:12 Mine is, yes. Jun 30 00:41:16 what about heat ? weather is hot actually... Jun 30 00:41:28 38 here! Jun 30 00:41:35 celcius ? Jun 30 00:41:36 I'd say ~40C Jun 30 00:42:10 (do not know aynthing about °F conversion) Jun 30 00:42:22 jf-work, that is C here... Jun 30 00:42:38 jf-work: 37C is ~100F Jun 30 00:42:59 you've got 38/40°C.. wowww. Jun 30 00:43:12 38C is about 98deg F Jun 30 00:43:20 and that is the Air Temp here today. Jun 30 00:43:33 I'm talking slugtemp though :P Jun 30 00:45:20 how long will the slug survive this... :/ Jun 30 02:26:31 OK, monotone DB is now down, to be replaced by a new fresh DB. Jun 30 02:26:49 yup Jun 30 02:26:53 You must export your private and public keys from your existing DB, and burn them to CD and print them out. Jun 30 02:27:01 done Jun 30 02:27:18 then delete your old DB. You must *not* transfer data from old DB to new DB except via a patch file. Jun 30 02:27:20 I've got the keys in like 6 locations. Jun 30 02:27:33 I've got a dir with only the patches. Jun 30 02:29:15 The makefile will be updated? Jun 30 02:29:23 yes, tonight Jun 30 02:29:37 (i.e. within the next 6 hours it should be back to normal) Jun 30 02:30:18 Ah, by the time I get back from work Jun 30 02:34:25 so did it work? Jun 30 02:35:18 like a charm Jun 30 02:35:30 Cool! Jun 30 02:36:25 dyoung: check jtag Jun 30 02:36:41 Theres a lot of stuff there. Jun 30 02:36:45 Still reading Jun 30 04:06:31 <[g2]> hey dyoung or should that be speedy-dyoung ! Jun 30 04:53:57 hi Jun 30 05:45:17 morning Jun 30 06:28:09 Anybody got the monotone build to work for openslug? Jun 30 06:32:06 <[g2]> VoodooZ_Work, I built it many days ago Jun 30 06:32:24 <[g2]> but I haven't tired in the last couple days Jun 30 06:32:39 <[g2]> I'd wait until after the 1st and the OE transition Jun 30 06:32:50 VoodooZ_Work: "mkdir /home/slug ; cd /home/slug ; wget http://www.nslu2-linux.org/Makefile ; make" Jun 30 06:33:01 OE transition is now complete. Jun 30 06:33:09 <[g2]> has OE switched ? Jun 30 06:33:19 (it's already the 1st in Kiribati) Jun 30 06:33:22 yep Jun 30 06:33:43 we've all switched, the bk repos are dead, the monotone repos are live. Jun 30 06:33:51 <[g2]> I think VoodooZ_Work want's make openslug right ? Jun 30 06:33:51 I did all this but I keep getting the error: onotone: network exception: name resolution failure for ewi546.ewi.utwente.nl Jun 30 06:33:57 yep Jun 30 06:34:15 VoodooZ_Work: you need to start again from scratch with the new Makefile Jun 30 06:34:20 I just want to get up to speed on the changes so I can start adapting my old code if need be. Jun 30 06:34:28 really? Jun 30 06:34:34 from scratch? Jun 30 06:34:40 new monotone db Jun 30 06:34:54 you *cannot* use the one which was used for testing. Jun 30 06:35:05 hostnames have also changed Jun 30 06:35:07 ok. So I can just delete the whole folder and start over then? Jun 30 06:35:25 up to you. but you can' use the old db Jun 30 06:35:30 can't use Jun 30 06:35:46 ok, I'll try to delete everything then. Jun 30 06:36:26 I get a different error now: monotone: network exception: name resolution failure for monotone.vanille.de Jun 30 06:37:00 Unless it's my box but that woould be odd as everything works fine on it. Jun 30 06:37:02 it resolves for me Jun 30 06:37:05 you've got local network problems then Jun 30 06:37:14 ok. I'll look into it. thanks Jun 30 06:37:20 both those hosts resolve fine Jun 30 06:37:23 Non-authoritative answer: Jun 30 06:37:23 Name: monotone.vanille.de Jun 30 06:37:23 Address: 141.2.14.64 Jun 30 06:37:41 I'll try resolving it from my home box to isolate the issue. Jun 30 06:38:23 strange. nslookup resolves it with no problem on the same machine! Jun 30 06:38:38 Maybe the error description is bogus. Jun 30 07:32:42 anybody know if monotone requires special ports or does it use plain http? It seems something is blocking me. Jun 30 07:39:00 ov2k: Were you involved with the Ov511 webcam driver at some point? Jun 30 07:39:32 nope Jun 30 07:39:41 just a fanciful coincidence of names Jun 30 07:40:20 I did sign on with a question about updating the openvpn ipkg for unslung Jun 30 07:40:37 as in, I was wondering how I might do it and whether it was possible right now Jun 30 07:40:58 considering the whole bk thing Jun 30 07:41:03 sorry, I had you confused with somebody else. Jun 30 07:41:20 I unfortunately don't know much about unslung as I'm an openslug dude. Jun 30 07:41:39 the whole bk thing is changing real soon so I would wait. Jun 30 07:41:44 that's what I figured Jun 30 07:42:12 I know the fix for the problem that's been going over the nslu2-linux list, but I can't help put it in Jun 30 07:42:32 oh well Jun 30 07:42:37 I have other work to do anyway Jun 30 07:42:38 ov2k: bk is now gone Jun 30 07:42:45 right Jun 30 07:42:54 but we have the monotone replacement up and running Jun 30 07:43:15 does monotone have anything to do with the package repo, though? Jun 30 07:43:24 rwhitby: monotone seems to work fine on my home box. I'll try stuff. Jun 30 07:43:33 Optware (formerly known as unslung packages) repo? Jun 30 07:43:40 yes Jun 30 07:43:46 ov2k: they use CVS. Jun 30 07:43:46 there's a wiki page refering to cvs Jun 30 07:44:11 ok, then I'll follow the wiki page Jun 30 07:44:13 if you've got a sourceforge id, I can give you write access to that right now Jun 30 07:44:46 it's ov2k Jun 30 07:44:50 surprise surprise Jun 30 07:46:44 I have to go to work now, but hopefully I can get this sorted out later Jun 30 07:47:14 I'll probably be back, assuming I'll run into some sort of stupid problem Jun 30 07:47:50 ov2k: you've now got write access Jun 30 07:48:56 great, thanks! Jun 30 07:49:08 np, go fix something :-) Jun 30 07:50:04 I'll be trying Jun 30 07:50:17 that's all that any of us do :-) Jun 30 08:31:01 VoodooZ: I get: Jun 30 08:31:02 ERROR: Nothing provides virtual/armeb-linux-depmod-2.3 Jun 30 08:31:02 ERROR: dependency virtual/armeb-linux-depmod-2.3 (for unslung-kernel) not satisfied Jun 30 08:31:02 NOTE: no buildable providers for unslung-kernel Jun 30 08:32:23 what? unslung? Jun 30 08:33:02 yep. building openslug now - will let you know tomorrow Jun 30 08:33:09 * rwhitby-asleep goes to bed Jun 30 08:33:47 ok. thanks. Jun 30 08:33:51 goodngiht Jun 30 08:34:41 VoodooZ_Work, and anyone else who sees connection problems, monotone uses a netsync protocol on port 5253, initiated from the local side, so outgoing connections on that TCP port must be possible. Jun 30 08:35:18 jbowler-zzz: thanks. I'll check this out. Jun 30 08:36:12 But i think it's a problem with monotone. and recompiling it is a pain as I'm still running redhat 9 so I might upgrade to fedora core4 first. Jun 30 08:53:29 Yo peeps Jun 30 08:53:43 Any other frankensluggers out there? Jun 30 08:54:47 beewoolie-afk, openjtag and openslug Jun 30 09:47:23 NAiL: ping Jun 30 09:57:42 re Jun 30 09:59:12 any mailing list moderators here? need some help... Jun 30 10:00:24 <[g2]> hey Jun 30 10:00:51 <[g2]> kitno455, hey Jun 30 10:01:36 tommy, having this thread with guy about sane Jun 30 10:01:52 cant figure out where on yahoo page to allow him to post Jun 30 10:02:04 keep having to approve his every msg Jun 30 10:02:53 hmm is he coming from same address every time ? Jun 30 10:03:01 <[g2]> hmmm... I thought there was a url for enabling that Jun 30 10:03:03 yes Jun 30 10:03:18 he is daka@thg.se Jun 30 10:03:27 i just cant find link. i think i am blind Jun 30 10:05:13 also, is there a way to look at the libusb build scripts online? i want to see if openslug libusb has any patches applied Jun 30 10:21:59 found it. find the user, then click 'edit membershit' then go thru three more pages.... Jun 30 15:49:06 javques: ping Jun 30 15:49:16 jacques, ping Jun 30 15:50:25 anyone know how hot a un-underclocked slug runs? Jun 30 15:50:43 do you need a heat sink in other words Jun 30 15:50:44 hi christopher Jun 30 15:50:52 hi Jun 30 15:51:00 does the un-underclocked slug run hot? Jun 30 15:51:02 christopher, we can't detect any temperature change Jun 30 15:51:10 wow Jun 30 15:51:11 cool Jun 30 15:51:17 exactly! Jun 30 15:51:18 I need to do that then Jun 30 15:51:24 if it's there it's just a couple of degrees Jun 30 15:51:53 I compiled on mine for hours and it didn't feel any warmer than before Jun 30 15:52:03 which days logs have the initial conversation? Jun 30 15:52:45 well much of the discussion happened in #openjtag Jun 30 15:52:49 i c Jun 30 15:52:56 which I think is also logged Jun 30 15:52:58 so, how did you figure out that it was that resistor then? Jun 30 15:53:11 intel doco Jun 30 15:53:15 i c Jun 30 15:53:38 well I need to go find someone who has a soldering iron seeing as I don't have one here Jun 30 15:53:44 I was just chosen to make the post, I had very little hand in discovering the mod Jun 30 15:54:05 but several ppl did lots of testing and intel doc reading to come up with it Jun 30 15:54:06 beewoolie and dyoung beat the doco to death for an hour plus Jun 30 15:54:21 My reclock slug maxed out at 52degC over a 10 hour loadvg2 torture test Jun 30 15:54:24 ok, well a big thanks to all of those involved Jun 30 15:54:34 at some point in time, beewoolie said to himself "screw it" and removed a resistor Jun 30 15:54:42 that he planned to move to another location Jun 30 15:54:43 thanks for the info dyoung-zzzz Jun 30 15:55:02 at this point in history, we are unsure as to why beewoolie decided to check the results Jun 30 15:55:18 good thing he did :-) Jun 30 15:55:21 was it madness or a hint of genuis that caused beewoolie to boot his slug Jun 30 15:55:23 ppl really want to find a reason why it came that way from the factory, but we may never know, and it may have nothing to do with stability/performance Jun 30 15:55:45 mostly cus I mentioned the resistor positions and couldnt actually do it right thing (because I was about to get beat up) Jun 30 15:55:59 so he had one handy, so said what the h ck. Jun 30 15:56:16 Right *then*. Jun 30 15:56:19 yeah, I wonder why linksys did that Jun 30 15:56:39 one thing to remember, I don't think linksys designed the board Jun 30 15:56:40 but beewoolie booted his slug and unfortunately for those in the western hemispere (late in the evening) he announced his results Jun 30 15:56:49 we know they didn't do the OS integration Jun 30 15:56:49 they don't need the power for their situation so maybe they were trying to save thermally Jun 30 15:57:15 ByronT: beewoolie-afk's announcement was perfectly timed for me :-) Jun 30 15:57:33 the board may have had another application and the design was just copied Jun 30 15:57:46 rwhitby, except that your wife left you with the kids.... Jun 30 15:58:00 php will run much faster now :-) Jun 30 15:58:00 and you were soldering with a 6 month old in arms Jun 30 15:58:07 ;) Jun 30 15:58:09 yeah, otherwise I would have been #2. Jun 30 15:58:10 haha Jun 30 15:58:18 or very early in the OS porting they had a problem, decided to try underclocking, it fixed the issue and they left it that way, but later kernels fixed it in software Jun 30 15:58:25 no, my wife came home before I started soldering. let me make that perfectly clear. Jun 30 15:58:43 actually NAiL tried within a minute of beewoolie announcing Jun 30 15:58:44 well, I'm going to go find a soldering iron, this is exciting Jun 30 15:58:46 why is she reading over your shoulder? ;-) Jun 30 15:58:53 thanks for the information guys Jun 30 15:58:54 it was a good 2-3 hours before anyone else showed up Jun 30 15:59:08 yeah I woke up at midnight my time I think Jun 30 15:59:15 I know... I was trying to wake up jacques and pull dyoung from getting his butt kicked Jun 30 16:03:19 03marceln * 10unslung/Makefile: Addes net-snmp. So that people can test it and tune the config. Jun 30 16:05:41 glc: get our yer soldering iron! Jun 30 16:05:57 Used a fingernail clipper ! Jun 30 16:06:20 heh - classic! Jun 30 16:06:22 Boots kinda fast. Jun 30 16:08:38 nail clipper - neat Jun 30 16:09:09 glc: make sure you put that in the comments in the database Jun 30 16:09:40 mmh, i'm having some minor issuse with openslug (otherwise it's fantastic), question is should i post slugbugs for them? Jun 30 16:10:04 yep, there is a openslug category I think. Jun 30 16:11:03 or i'd like to have "empty" (maybe commented) files in /etc/modutils ... and stuff like this Jun 30 16:11:27 or i can't find a way to set the slug's timezone Jun 30 16:12:07 i was just wondering because the slugbug category looks pretty empty. Jun 30 16:12:46 Already did. Jun 30 16:12:47 dr_nick: openslug 2 is not very old. and most people using it at the moment are experienced enough to just fix the problems themselves. Jun 30 16:13:54 lol we're still arguing about the best way to set the TZ Jun 30 16:14:09 I cp a /etc/localtime file from my desktop Jun 30 16:14:22 other ppl like to set TZ env var Jun 30 16:15:33 rwhitby: yes. most things are more like annoyances, which would just be nice to be there from the beginning. but don't get me wrong, thanks for everything. openslug is great, i love it. Jun 30 16:15:55 if he only knew..... Jun 30 16:17:56 ka6sox-office, starting it up... Jun 30 16:18:24 ka6sox-office is not here... Jun 30 16:25:18 just ssh'ing in I can tell it's faster Jun 30 16:27:11 yep, dropbear is notieably faster Jun 30 16:27:17 since it's cpu-limited normally Jun 30 16:27:23 (much more so than openssh) Jun 30 16:28:01 well, this has been fun, but I gotta go to work now. Jun 30 16:28:12 christopher, that was fast - you already modded? Jun 30 16:28:13 * dyoung-zzzz me toos Jun 30 16:28:25 make sure you make an entry in the turboslug database Jun 30 16:28:51 glc is #14 Jun 30 16:29:02 so christopher will be #15 Jun 30 16:29:27 not bad for 18hrs Jun 30 16:29:42 not bad at all, tho rwhitby-away doing three at once helped the count :-) Jun 30 16:30:20 Calibrating delay loop... 263.78 BogoMIPS Jun 30 16:30:34 glc, thats the same bogomip number I had. Jun 30 16:30:40 unslung seems to report that. Jun 30 16:30:40 that's an unslung turboslug Jun 30 16:30:44 bah, 2.4 kernel looosers Jun 30 16:30:54 2.6 gets 266.24 Jun 30 16:30:58 I was feeling ripped off. I want my 3 bogomips dammit Jun 30 16:31:18 i have more bogomips than you nyah nyah Jun 30 16:31:18 glc gets to put the "Nail Clipper R83 method" in the db Jun 30 16:31:32 hey Jun 30 16:31:39 :/ Jun 30 16:32:33 tnats nail clipper, not NAiL Clipper. Jun 30 16:32:55 Great logo would be a slug with a turbocharger as a snail shell. Jun 30 16:33:25 that's TurboSlug! Jun 30 16:33:34 ;) Jun 30 16:33:46 ooooh that would rock! Jun 30 16:33:54 they already look like snail shells Jun 30 16:39:57 jacques: yes up and running Jun 30 16:43:29 now I need to perform the operation to my other slug, but it doesn't even have a serial mode yet :-) Jun 30 16:44:37 neither of my slugs are serial'd (though FrankenSlug is getting closer to his operation) Jun 30 17:32:20 NAiL: ping Jun 30 17:34:32 NAiL: I'm trying to generate a list of all the diffs between the old monotone db and the new (as yet unmodded one). Jun 30 17:34:51 By 'all the diffs' I mean all the ones we made for NSLU2 - minus all the ones OE made.. Jun 30 17:35:39 jbowler: sorry about leaving out your key, but I needed confirmation that you had saved it somewhere else first. Jun 30 17:35:49 Is mine in there? Jun 30 17:35:58 Oh yes, it's backed up on my other slug. Jun 30 17:36:03 :) Jun 30 17:36:15 pong Jun 30 17:36:19 jbowler: CD or hard-copy preferred :-) Jun 30 17:36:32 Tiersten: have you confirmed non-volatile backup of your key? Jun 30 17:36:52 Yes, I went for the CD option, I don't trust the OCR on my scanner. Jun 30 17:36:54 NAiL: is the Master Makefile working fine? Jun 30 17:37:11 still building. But seems to work fine Jun 30 17:37:24 .mt-attrs is going to be an enormous source of annoyance Jun 30 17:37:28 Yes. It's on a DVD backup I made the other day Jun 30 17:37:41 ah, yes... Jun 30 17:37:53 * NAiL realizes cron is gonna fail to build Jun 30 17:38:04 And ypserv Jun 30 17:38:19 (Unless you linked downloads) Jun 30 17:38:36 Tiersten: it's in now Jun 30 17:38:37 * NAiL fixes downloads Jun 30 17:38:58 Thanks Jun 30 17:39:18 Anybody who has a magnifying glass is free to fill in the 4 missing gaps in my config register table :) Jun 30 17:39:53 Should the overclocking page be linked into the main howto page? Jun 30 17:40:14 As it has been announced, I can't see why not Jun 30 17:40:22 k Jun 30 17:40:42 jbowler: Can you push the .mt-attrs change? Jun 30 17:41:07 I'm still trying to work out what the changes are: there's a whole load of unslung stuff missing. Jun 30 17:41:20 The sveasoft crowd are going to think we're even more geeks now :) Jun 30 17:41:27 NAiL: push anything you need to - I've got the original db and won't sync until I know the changes. Jun 30 17:41:48 I've got the original repo for backup as well Jun 30 17:41:50 So I'll have a list of what I think needs to be put back, and if some stuff is already back that's fine. Jun 30 17:42:36 samba is back up to what it used to be Jun 30 17:42:52 as is packages/cron, except that attrib stuff Jun 30 17:43:30 Pusing the .mt-attrs change line-by-line is probably best. Jun 30 17:43:54 Yeah Jun 30 17:44:00 It's actually a LUA hook, it could probably be made to be per-directory. Jun 30 17:44:14 I've only pushed small changes. Easier to back up to a good state if something is screwed Jun 30 17:44:47 Oh, better idea: you or I could fix cron to install that .sh file rather than just using it in place. Jun 30 17:45:19 Looks stupid, but install -c -m 755 ${WORKDIR}/install-sh . Jun 30 17:45:20 install it? I don't get it. Jun 30 17:45:26 aaah Jun 30 17:45:29 yeah Jun 30 17:45:31 got it Jun 30 17:46:07 I've been trying to eliminate executable attributes on files in the packages/* directories anyway. Jun 30 17:46:18 yeah Jun 30 17:46:20 I figure it's better to put the a+x on them in the work tree. Jun 30 17:46:31 I'll stuff it in there and see if it works in my build Jun 30 17:46:56 hmm.. are the .bb's read only on the start of the build? Jun 30 17:47:07 Oh well, I can wait until cron fails, then restart. Jun 30 17:47:22 Yes: bb's cached at the start. Jun 30 17:47:45 So the downloads fix works, but changes to .bbs don't. Jun 30 17:49:35 NAiL: could you check what this does on your db: monotone --db=your.db list certs 1b7c629d1f084951148d8e7d5ff691dc5d061f7c Jun 30 17:50:06 (It should just say that rwhitby "Bumped OpenSlug version") Jun 30 17:50:23 lists four rwhitby-stuffs.. Bumped is one Jun 30 17:50:31 aga Jun 30 17:50:32 aha Jun 30 17:50:39 The changelog part says Bumped Jun 30 17:51:08 Thx, needed to double check what I was doing (I've got too many monotone db's lying around...) Jun 30 17:51:17 heheh ;) Jun 30 17:51:30 What's the current version of Openslug? 2.1pre? Jun 30 17:52:08 Look in conf/distro/openslug.conf Jun 30 17:52:27 2.1-beta Jun 30 17:52:38 that was just my bump to be able to create the org.openembedded.nslu2-linux branch. it is of no consequence :-) Jun 30 17:52:38 nice Jun 30 17:52:54 ok Jun 30 17:53:15 I'm not pushing the udev changes yet. Not until I figure out what's happening at boot Jun 30 17:53:15 So the current version is 2.0-beta, the next version will probably be 2.2-beta. Jun 30 17:53:30 2.1-beta means someone built from monotone. Jun 30 17:53:36 aaah Jun 30 17:53:44 good idea :) Jun 30 17:54:06 But, to quote someone else, it's just a monotonically increase number. Jun 30 17:56:27 This is so annoying... I'm shopping for a new mobo... But all the cards with S-ATA have RAID-controllers. I don't want no steenking hardware raid. Jun 30 17:56:46 I have to pay a lot extra to *not* have S-ATA RAID :\ Jun 30 18:05:48 [g2]:) Jun 30 18:06:55 <[g2]> hey NAiL ! Jun 30 18:07:32 <[g2]> hard to believe it's not even 24 hours :) Jun 30 18:07:56 hehe, another three hours or so, and my slug has been up for 24 hours :D Jun 30 18:11:32 NAiL: good idea not to push the udev changes Jun 30 18:11:44 yeah Jun 30 18:12:06 Atleast not till I'm sure what actually happens, and can fix it. Jun 30 18:12:16 if you want, you could check them in on an org.openembedded.nslu2-linux.udev branch if you need to collaborate. Jun 30 18:12:52 I need to learn how to work with multiple branches first Jun 30 18:13:07 Right now, I'm sticking to this one branch, so I don't stuff up anything. Jun 30 18:24:00 Good Plan Jun 30 18:33:57 NAiL: did you add EDITOR/TERM to base-files/profile? Jun 30 18:34:17 yes Jun 30 18:34:42 Ok, so that's part of the 'cron' set of changes. Jun 30 18:34:48 yup Jun 30 18:35:33 Weird, but it looks like OE removed the celf-ash-builtins patch from busybox. Jun 30 18:36:17 hmm Jun 30 18:36:49 It's in my new repo Jun 30 18:37:09 busybox/busybox-1.00/celf-ash-builtins.patch Jun 30 18:37:40 Yes, but it's not in the SRC_URI in busybox_1.00.bb Jun 30 18:37:44 oh Jun 30 18:37:54 Unless, I have a local change... Jun 30 18:38:28 It's removed from SRC_URI, yes Jun 30 18:38:38 jbowler: I think I remember seeing that mentioned on #oe, but don't remember the details. Jun 30 18:39:24 I discussed some problems, or at least told them about some, and had nslu2-linux fixes - they maybe just removed the patch entirely. Jun 30 18:39:51 It looks like someone added CONFIG_TUN to the kernel too (as a module) - dyoung? Jun 30 18:39:54 Can anyone explain the purpose of /.recovery ? I've been hit by my slug not booting twice. Both times, rm .recovery fixes the problem Jun 30 18:39:57 jbowler: nope, me Jun 30 18:40:06 It's for openvpn Jun 30 18:40:36 Ok, is the nslu2-io.c change the one you got from mr_claus? Jun 30 18:40:41 yeah Jun 30 18:41:20 config_tun was someone, but I forget who. it was a good change. Jun 30 18:41:33 rwhitby-web: twas me, I tell ya :) Jun 30 18:42:06 oh, yeah, you did. I read that as "not me" :-) Jun 30 18:42:15 hah, ok :) Jun 30 18:42:34 NAiL: there are quite a few Python changes in there, but I think they are all from OE. Jun 30 18:42:48 I haven't touched anything Python Jun 30 18:43:51 Then there's strace, I think that is OE because strace_4.5.11.bb has an extra quota.patch Jun 30 18:45:08 jbowler: Yeah, it's from koens branch Jun 30 18:45:20 jbowler: You don't check "mt log"? :) Jun 30 18:47:05 Hum, I thought they had just dropped the latest revisions from bk (I hadn't checked their logs) Jun 30 18:48:05 I'm actually diffing two working trees - I'll put up a list of the diffs I see on pastebin (mostly what I want to find are my own changes...) Jun 30 18:48:12 <[g2]> there's an strace .12 which was checked in a couple days ago Jun 30 18:50:07 if they'd checking 4.5.9 we'd have something Jun 30 18:50:28 another thing that was moved for no good reason Jun 30 18:50:33 removed Jun 30 18:51:02 AFAICT no strace after 4.5.9 works for armeb Jun 30 18:51:19 my post to the strace devel list (which is 95% spam) was totally ignored Jun 30 18:51:42 <[g2]> I think .12 built fine Jun 30 18:51:53 <[g2]> I added it back in Jun 30 18:52:13 yeah they all build Jun 30 18:52:24 but they didn't work for me Jun 30 18:52:29 even native-built Jun 30 18:52:30 <[g2]> actually.... I saw the checkin from pb_ in BK but I didn't see it in our repo Jun 30 18:52:47 <[g2]> I just built an old one Jun 30 18:53:50 strace, bison, and file are on my build but don't work list Jun 30 18:54:25 <[g2]> http://nslu2-linux.bkbits.net:8080/openembedded/cset@1.3644.1.17?nav=index.html|ChangeSet@-3d Jun 30 18:54:34 jacques: What's wrong with file? Jun 30 18:54:42 NAiL: my list is on http://paste.uni.cc/7435, I grouped the changes Jun 30 18:54:50 NAiL, it kept saying everything was 'data' Jun 30 18:54:59 its way of saying "I dunno" Jun 30 18:55:28 haha Jun 30 18:56:34 jbowler: The udev.inc change isn't mine, but other than that it looks right Jun 30 18:56:41 <[g2]> DOH where did that come from that's the wrong patch Jun 30 18:57:30 <[g2]> jacques, this one http://openembedded.bkbits.net:8080/openembedded/cset@1.3656?nav=index.html|ChangeSet@-3d Jun 30 18:58:57 i bet it works fine on armel Jun 30 18:59:03 <[g2]> hey Jun 30 18:59:18 Someone added installation of /etc/udev/rules.d/50-udev.rules Jun 30 18:59:21 <[g2]> where's Jacmet when you need'e Jun 30 18:59:23 <[g2]> m Jun 30 18:59:39 The change is an add in OE though (I didn't look closely enough). Jun 30 18:59:44 jbowler: Which is a good idea. But not mine ;-) Jun 30 19:00:40 Ok - so are you happy checking in all those changes (there's some overlap in our change set in samba and cron) Jun 30 19:01:16 I'll get the openslug-packages up-to-date and try to fix up .mt-attrs. Jun 30 19:01:44 Yeah Jun 30 19:02:11 hmm Jun 30 19:02:15 atftp doesn't compile Jun 30 19:02:31 In file included from tftp.c:43: Jun 30 19:02:31 argz.h:184: error: parse error before '{' token Jun 30 19:04:51 Missing macro definition? Jun 30 19:05:11 Quite possible. Haven't investigated. My C-skills are rather weak. Jun 30 19:05:53 I'm still battling with trying to patch busybox Jun 30 19:07:12 I can't figure out how it handles the parameters. It confuses me ;-) Jun 30 19:07:23 That's a typically useless C compiler error message - it could mean almost anything, but if the code compiles on other systems it's probably a pre-processor/header file issue. Jun 30 19:07:38 How bb handles the parameters to the commands? Jun 30 19:07:44 yeah Jun 30 19:08:25 I pasted some of the comments in the source here a while back. That was fun :-P Jun 30 19:09:01 Each command has it's own 'main' routine (syntax something like ls_main) - that just gets argc/argv from the real busybox main. Jun 30 19:09:49 Yeah, well... It seems that start-stop-daemon doesn't get "--quiet" right. So it makes noise when it shouldn't. Jun 30 19:10:41 There's two checks missing as well Jun 30 19:12:58 It's just an aesthetic issue, but it screws up how the initscripts look :( Jun 30 19:16:18 That's why I was deleting the --quiet - I had problems with it before (and didn't try to fix them) Jun 30 19:16:50 Ok, someone has already done openslug-packages - the only thing left to do is to move 'cron' from the broken list. Jun 30 19:19:08 Read the 'meld' windows wrong, cron is not in the BROKEN list. Jun 30 19:20:20 oh Jun 30 19:20:22 yes Jun 30 19:20:47 * NAiL feels guilty Jun 30 19:21:35 Should I check in the previous changes? Jun 30 19:21:40 (I.e. to cron) Jun 30 19:21:50 Cron is checked in Jun 30 19:22:42 I don't see any changes. Jun 30 19:23:20 no Jun 30 19:23:23 what the? Jun 30 19:23:30 Did I forget cron? Jun 30 19:23:44 Strange... Jun 30 19:24:34 haha, yes, I did Jun 30 19:24:40 my build just failed Jun 30 19:24:40 repvik@dev:~/nslu2/openslug$ cat /home/repvik/nslu2/openslug/tmp/work/cron-3.0pl1-r0/temp/log.do_install.23136 Jun 30 19:24:44 install: cannot stat `/home/repvik/nslu2/openslug/tmp/work/cron-3.0pl1-r0/install-sh': No such file or directory Jun 30 19:24:50 Ok, you check in cron ;) Jun 30 19:25:43 Ok, I'll do the .mt-attrs too then, because cron seems to have the only relevant change (there's some unslung stuff, but I don't know the status of that.) Jun 30 19:33:13 NAiL: did you check in base-files/base-files/profile? Jun 30 19:34:43 Yes, changeset de56403f43f1ceeaffcfca380aff23d0c1426036 Jun 30 19:35:45 Need viewmtn on nslu2-linux.org... Jun 30 19:36:02 * NAiL is still learning Jun 30 19:36:59 I did a checkin of the ypserv rename, but I wasn't synced to your changes, so I created two heads in org.openembedded.nslu2-linux. Jun 30 19:37:21 I then merged the two heads, which makes it very difficult to see where the changes you checked in came from. Jun 30 19:37:33 viewmtn would show the graph, and it would be more obvious... Jun 30 19:39:16 aha Jun 30 19:40:08 The command line 'mt log' makes it look like I checked in a whole load of files at once - had me worried for a while. Jun 30 19:40:52 Cron is in, revision cfa6faedd18c389ac5ce1553f4fe870765069a53 Jun 30 19:42:42 jbowler: you can push to ewi, and look at viewmtn there. Just don't propagate. Jun 30 19:43:24 That's vanille.de now presumably? Jun 30 19:43:39 no viewmtn is only on ewi I think Jun 30 19:43:48 ewi syncs with vanille every 3 hours I heard Jun 30 19:43:54 vanille is the master Jun 30 19:44:14 I expect vanille will eventually get viewmtn Jun 30 19:44:44 We should probably propagate pretty soon - once we have a stable build - quite a few changes have accumulated. Jun 30 19:45:47 yep, as soon as we have a stable known-good build Jun 30 19:46:07 jbowler: do you have write access to #oe's server yet? Jun 30 19:47:57 I do not know - what I saw in #oe suggested that koen/mickeyl had added just your key, but there is no way of telling (without trying a push). Jun 30 19:50:35 I get a "failed, disconnecting" from ewi, so I think the answer is no. Jun 30 19:51:52 However, that actually proves nothing - until a few minutes ago my public key wasn't going to be transmitted when a sync happens to OE. Jun 30 19:52:12 Now I've checked in it will be, so I might have access after those checkins get synced into OE. Jun 30 19:54:32 bbl, supper time Jun 30 20:03:43 jbowler-away: just send your key to mickeyl Jun 30 20:53:21 btw, was someone working on getting a bot up that notified of changesets? Jun 30 20:57:44 its broken Jun 30 20:58:20 its a program that will be called by cron that talks to CIA in channel. Jun 30 21:00:22 03bzhou * 10unslung/ (make/py-bluez.mk Makefile): py-bluez ready for cross testing Jun 30 23:34:07 repvik: ping Jun 30 23:40:22 repvik: the packages/samba/files/{init,configure.lfs.patch} new files were missing, I just commited them Jun 30 23:42:51 In the unslung build: ERROR: Nothing provides virtual/armeb-linux-depmod-2.3 Jun 30 23:43:52 I don't understand the '2.3' - it looks like the changes to classes/kernel.bbclass, which should append the kernel major version number (2.4?) fail somehow. Jun 30 23:50:05 The get_kernelmajorversion use on the DEPENDS, where it uses PV, is flawed - it should apparently use KERNEL_MAJOR_VERSION, hum. Jun 30 23:55:41 strace also fails. **** ENDING LOGGING AT Thu Jun 30 23:59:57 2005