**** BEGIN LOGGING AT Sun Jul 03 23:59:56 2005 **** BEGIN LOGGING AT Mon Jul 04 00:13:30 2005 Jul 04 00:49:43 when i do make, it fails at bluez-utils-nodbus Jul 04 00:57:16 CHAOSiTEC: Did I talk to you about that the other day? Jul 04 00:57:23 or was that someone else? Jul 04 00:57:37 * NAiL does 70 things at once nowadays :) Jul 04 00:57:40 yeah.. i tried to delete the whole slug dir, and started from scratch Jul 04 00:57:53 but alas it still fails same spot Jul 04 00:58:04 right now, im trying bb bluez-utils-nodbus Jul 04 00:58:40 lol NAiL Jul 04 01:00:48 Ok, I've just set up a new repo, I'm gonna try building -nodbus from that. Jul 04 01:01:08 ok (atm bb is building binutils here Jul 04 01:02:05 Exactly how did you set up stuff? Jul 04 01:02:14 howtos stepbysetep Jul 04 01:03:08 then got the makefile, did makefile setup;make Jul 04 01:03:19 lol, not makefile setup, but make setup Jul 04 01:03:50 uh.. first stepbystep, then the makefile? Jul 04 01:03:59 ya Jul 04 01:04:17 ie. the http://www.nslu2-linux.org/Makefile ? Jul 04 01:04:43 Because you don't need both. Jul 04 01:04:50 lol Jul 04 01:05:02 if i dont run the makefile, then bb will fail Jul 04 01:05:29 You only need to make a directory (/home/nslu2), cd into it, wget http://www.nslu2-linux.org/Makefile, make Jul 04 01:05:44 ok Jul 04 01:06:08 then i really dont need to do the stepbystep thn? Jul 04 01:06:12 nope Jul 04 01:06:13 then* Jul 04 01:06:20 lol Jul 04 01:06:58 but will it download the whole oe stuff etc? Jul 04 01:07:22 That'll download everything you need to build Jul 04 01:07:34 and for making new packages? Jul 04 01:07:34 (except monotone, cvs & svn) Jul 04 01:07:39 yes Jul 04 01:07:48 Have you installed monotone? Jul 04 01:07:57 What distro are you running? Jul 04 01:07:58 ok lol then i suddenly found out what the problem might be lol Jul 04 01:08:01 gentoo Jul 04 01:08:12 yeah monotone is up and running Jul 04 01:08:29 emerge sync && emerge monotone (and subversion and cvs if you don't have those installed already) Jul 04 01:08:50 i do have subversion and cvs installed as well Jul 04 01:08:59 ok, then you'll probably just need monotone Jul 04 01:09:12 It should install "boost", and monotone 0.19 Jul 04 01:09:14 monotone is installed Jul 04 01:09:18 good Jul 04 01:09:42 then you just make the dir, wget the makefile, make openslug-firmware (IIRC) and then you're set. Jul 04 01:09:52 had a few problems getting it installed, until i unmasked so that i could get the latest (i.e. 0.19) Jul 04 01:09:59 going to run it now, lol Jul 04 01:10:48 Shouldn't be masked... Bit strange that. I've emerged it on three boxen, and not had problems with any of them Jul 04 01:10:58 but it doesn't matter as long as it's installed ;) Jul 04 01:11:25 (im running on amd64 Jul 04 01:11:31 oh Jul 04 01:11:45 I haven't tested on amd64 yet. Jul 04 01:11:52 I'm getting my amd64 box *today* Jul 04 01:12:09 there was an install script problem with boost, so affected the install of monotone, until i did the unmasking, i.e. newer version Jul 04 01:12:17 aha Jul 04 01:12:46 mine runs rock stable ;-) Jul 04 01:12:48 do you have the url to the stepbystep page? Jul 04 01:14:30 http://www.nslu2-linux.org/wiki/OpenEmbedded/StepByStepSetup Jul 04 01:15:46 well, deleted the dir, and its now runnning make setup Jul 04 01:16:18 good Jul 04 01:16:44 after that you can enter the openslug dir, make conf/local.conf and make setup-env, then you can try bb bluez-utils-nodbus Jul 04 01:17:08 yeah, but that will proberly not be until i get back from work, lol Jul 04 01:17:20 make setup won't take that long Jul 04 01:17:29 It just fetches OE and a few other things Jul 04 01:17:34 i have 30 minutes left, lol Jul 04 01:17:45 give it 5-10 mins, and it should be done :) Jul 04 01:17:57 ok Jul 04 01:21:49 0, Jul 04 01:21:50 , Jul 04 01:22:23 current progress:monotone: verifying new revisions (this may take a while) Jul 04 01:24:16 yeah, that takes a bit of time Jul 04 01:24:23 yeah lol Jul 04 01:24:45 so, looking forward for your amd 64 bit system? Jul 04 01:24:58 are you kidding? ;) Jul 04 01:25:22 lol Jul 04 01:25:43 Replacing this dusty old XP2700+/768mb with a AMD64 3700+/1GB is... good :) Jul 04 01:25:46 i bought mine the same day the shops had the 0.09 cores out Jul 04 01:26:22 its the slowest one lol, 3000+ that i have Jul 04 01:26:41 i upgraded from the 1800+ athlon Jul 04 01:26:46 ah Jul 04 01:27:01 My 2700+ will be downgraded to a dedicated build-box now ;) Jul 04 01:27:12 and, im thinking going x2 this year Jul 04 01:27:18 Dunno what I'll do with my 2000+ :P Jul 04 01:27:36 nice!, mine is just laying in the closet, but alas, i have 4 comps..lol Jul 04 01:27:55 I've got three right now. Jul 04 01:28:06 xp 2700, 2000 and 1800 Jul 04 01:28:24 3 running linux, one runs MCE 2005 beta 2 (sigh, linux freak betatesting microsft product, lol) Jul 04 01:28:49 03jeanfabrice * 10unslung/Makefile: Demote wget-ssl for wiley as it seems to link against librt.so which doesn't exist on wiley Jul 04 01:29:59 03jeanfabrice * 10unslung/Makefile: Change UNSLUNG_TARGET to nslu2 Jul 04 01:30:38 Two running linux, and one running Windows XP. Jul 04 01:30:48 My desktop is a Wintendo-box Jul 04 01:30:56 i dual boot on my main, i NEED my games, lol Jul 04 01:31:02 hmm Jul 04 01:31:07 I don't have any games installed at all Jul 04 01:31:37 well.. I've got this spider thingy, hears... and minesweeper. But nothing more than that :P Jul 04 01:31:40 hearts Jul 04 01:31:51 I don't exactly play a lot ;) Jul 04 01:32:16 lol, i play Battlefield 2, and eq2 Jul 04 01:32:17 If I want to play something, I'll sit down and play Day of the Tentacle on my modded XBox ;) Jul 04 01:32:53 lol Jul 04 01:33:44 Or burnout, if I need to get my head off things Jul 04 01:34:05 ;-) good game Jul 04 01:34:24 Maybe I'll sit down and try God of War one of these days. Looks kinda neat. Jul 04 01:34:36 How's make setup doing? Jul 04 01:35:06 just started bb bluez-utils-nodbus right now Jul 04 01:35:12 cool Jul 04 01:35:52 03jeanfabrice * 10unslung/Makefile: Promote net-snmp for wiley Jul 04 01:36:15 im hoping i will be able to get lvm2 up for the box Jul 04 01:37:00 yeah Jul 04 01:37:13 I don't have enough disks for lvm2 to make sense :P Jul 04 01:37:21 i have 600GB just sitting there, waiting, lol Jul 04 01:37:58 then, next month i will buy 2 other slugs, 1 with 600GB hds, for backing up the first lol Jul 04 01:38:11 hahah Jul 04 01:38:13 the second slug, will be dev Jul 04 01:38:30 I'm waiting for my budget to allow a few more slugs myself Jul 04 01:38:40 the disks are a bit more expensive though :( Jul 04 01:38:45 true Jul 04 01:38:55 Right now, one of my slugs use one of my iPods as root disk :P Jul 04 01:39:18 but in total i now have 960Gigs of storage on all my comps, but i want most of it to be on the network Jul 04 01:39:42 lol, i was actually thinking of using my creative zen micro lol Jul 04 01:41:55 but i settled for 512 mb swap, and 10gigs or root, way to much lol Jul 04 01:42:13 hehe Jul 04 01:42:32 i actually bought the slug, because of nslu2-linu.org Jul 04 01:42:42 linux* Jul 04 01:43:04 Yeah, me too I think :) Jul 04 01:43:33 i wasnt satisfied with the stuff it could do, out of the box, thats why i never bought it before, lol Jul 04 01:43:47 hehe, exactly Jul 04 01:43:54 Now it can do so much more :D Jul 04 01:43:59 ohh yeah! Jul 04 01:44:00 lol Jul 04 01:44:36 just need to get perl working lol Jul 04 01:44:59 Works if compiled natively. If you can fix the cross build, I'd really be happy :P Jul 04 01:45:05 (by cross compiling, so that its just a matter of ipkg install perl Jul 04 01:45:29 but alas, i cant figure out why it fails, must be a problem, when compiling a specific lib Jul 04 01:45:44 (symbol problem as i recall Jul 04 01:45:49 It's not quite my department either, unfortunately Jul 04 01:46:20 last time i developed in either C or C++ was back in the good ol' dos days lol Jul 04 01:46:47 i gave up on developing on windows. too much of a pain, and never really tried on linux Jul 04 01:48:59 hmm, do you remember what package 'file' is from? Jul 04 01:49:17 no, and afaik, file is broken? Jul 04 01:49:28 it's in "file" :P Jul 04 01:49:29 ahh ok Jul 04 01:49:32 lol Jul 04 01:50:08 Haven't tried file, someone (g2?) talked about it being broken, and just answering "binary data" or something like that Jul 04 01:50:19 i c Jul 04 01:50:39 i think g2 talked me into using my dev slug for embedded gentoo lol Jul 04 01:51:05 haha Jul 04 01:52:39 lol, i havnt even deunderclocked my slug yet, nor fattened it Jul 04 01:52:55 I don't think there's that many fatslugs yet Jul 04 01:53:08 But the deunderclocked stats are growing quite fast :D Jul 04 01:53:13 then there will be, soon, lol Jul 04 01:53:24 I've got two slugs deunderclocked. #2 and #20 Jul 04 01:53:32 yeah i noticed on the mailing list Jul 04 01:54:59 but 40 slugs, use distcc, now that would be fun, rofl Jul 04 01:55:13 but=buy Jul 04 01:55:24 "Imagine a BeoWulf cluster of..." Jul 04 01:55:34 rofl, yeah Jul 04 01:55:45 of course, all deunderclocked, and fattened..rofl Jul 04 01:55:50 yes :D Jul 04 01:56:03 man, glibc takes forever to package Jul 04 01:56:09 i bet Jul 04 01:56:52 hmm, cant we build perl natively, then package them, instead of having the broken cross compiled one?? Jul 04 01:57:17 yes. That is kinda my job ;) Jul 04 01:57:25 lol Jul 04 01:58:07 * CHAOSiTEC is a madocist, and wants webmin up on th slug Jul 04 01:58:53 hahah Jul 04 01:59:22 hmm a slugfarm, managed all by webmin (wonders what cpu cycles are left for actual jobs, rofl Jul 04 02:00:13 nah, webmin will only take cpu when you're actually making changes, won it? Jul 04 02:00:16 won't Jul 04 02:00:41 uses a little for the httpd perl script, but shouldnt be too much i think Jul 04 02:03:12 ohh well, one smoke, then im off for work Jul 04 02:07:10 see you around NAiL, im off for work... Jul 04 02:07:17 later Jul 04 02:07:25 laterz Jul 04 02:07:51 NAiL: do you know a way to stop udev without killing it by killall? Jul 04 02:11:54 mr_claus: hmm... using ps|grep|sed and kill? :P Jul 04 02:14:22 0, Jul 04 02:14:22 0. Jul 04 02:28:08 CHAOSiTEC_away: bluez-utils-nodbus built now. So it should build for you as well. Jul 04 02:48:38 NAiL: ps is not working without proc filesystem Jul 04 02:49:45 ah, heh Jul 04 02:50:12 very... nice... Jul 04 02:50:25 I'll take a look at it later today, got my hands full right now Jul 04 05:41:00 <[g2]> any apache users out there ? Jul 04 06:01:01 OT: Anybody knows how to configure a fedora box so it tries the local passwd before ldap? (so when ldap server is down you can still login) Jul 04 06:44:05 VoodooZ_Work: it's probably something in /etc/pam.d Jul 04 06:44:12 not much of a help, i know Jul 04 06:48:26 <[g2]> kergoth, HEY! Jul 04 07:05:18 Could someone do me a favour please? Jul 04 07:05:46 I don't post to the Yahoo mail list and am not about to start - spam being a big reason. Jul 04 07:06:24 Set up a Yahoo account just for posting then Jul 04 07:06:48 Could someone who does post tell the guys in the Sane scanner thread to read the comments in optware/make/libusb.mk ? Jul 04 07:07:07 they'll save themselves a lot of running around in circles. Jul 04 07:08:08 libusb is fundamentally broken in the way it accesses structures and there are no easy patches to fix the problem for good. Jul 04 07:08:28 Didn't you abandon it and write your own access functions? Jul 04 07:08:34 exactly Jul 04 07:11:21 Okay. Sent. Just this time though. Sign up a throwaway email account next time okay? :) Jul 04 07:11:45 Tiersten: thanks. Jul 04 07:12:10 peteru-home: you have your own domain, right? Jul 04 07:12:25 just make up a new local address, and bounce it Jul 04 07:12:54 I've never gotten spam from yahoo mailing list posts Jul 04 07:14:06 I suppose I should go to sleep. Jul 04 07:14:45 heh Jul 04 07:19:08 rwhitby-away: I tend to "rotate shields" by using a lot of email addresses and then black listing the ones that go bad, but the Yahoo sign up process is just that little bit of too much effort :-) Jul 04 07:19:20 * peteru-home is a lazy man. ;-) Jul 04 07:19:50 you could set up a yahoo account + email and never check your yahoo email Jul 04 07:19:52 :) Jul 04 07:21:21 yep Jul 04 07:21:59 peteru-home: hm... Jul 04 07:22:48 is it really that bad? Jul 04 07:25:22 peteru-home: did you make your own access functions for sane, or something else? Jul 04 07:25:56 peteru-home wrote puppy. Jul 04 07:25:58 ~puppy Jul 04 07:26:15 hmm Jul 04 07:26:26 Puppy is a network interface tool for the toppy. Jul 04 07:26:28 ~toppy Jul 04 07:26:28 hmm... toppy is a dual tuner digital receiver and personal video recorder - see http://www.topfield.co.kr/product_e/pr_feature.asp?cb=DTR&cm=PVR&pn=TF5000PVRt Jul 04 07:26:35 ahh Jul 04 07:27:21 Unfortunately a lot of packages assume that the entire world is a 32 bit x86 Jul 04 07:27:34 what, it isn't? Jul 04 07:27:48 Nope. I've got a 16 bit x86 somewhere... :) Jul 04 07:28:04 I've just got a 64bit x86 :P Jul 04 07:28:33 actually, libusb works, just seems to gives sane 32 bits of image data instead of 24. If sane could just skip the extra 8 bits all would be fine... everything else sane uses libusb for works great Jul 04 07:29:07 DaKa2: Sure but it doesn't fix the underlying problem. You can put in a hack to discard the extra data if you want but it's a bit ick Jul 04 07:29:30 ~ube Jul 04 07:29:31 ube is, like, Ugly But Effective. Or sometimes Unorthrodox But Effective Jul 04 07:29:48 hardware is bearable because it *should* be a one off :) Jul 04 07:30:11 well, yeah, just dont know how to fix libusb, seems like they know they bade a big mistake and is remakeing the entire api for version 1.0 Jul 04 07:30:45 DaKa2: I ended up using usbfs directly, instead of fixing the problems with libusb Jul 04 07:31:48 the proper way to fix libusb is not to declare the structs as packed, but for the code to actually unpack the individual bytes from the URBs as needed. Jul 04 07:32:20 effectivelly, that means rewriting enormous chunks of libusb Jul 04 07:32:32 hm.. that sounds like a lot of work... Jul 04 07:33:32 it was a lot easier to talk to usbfs directly, although I find that usbfs has some issues in 2.6.11 (at least on MIPS) Jul 04 07:34:06 oh fun.. another place that things can go wrong.. Jul 04 07:34:39 yeah, libusb 1.0 refactor has been on the drawing board for a long time. Jul 04 07:34:44 starting to think running 2.4 with the scanner module isnt such a bad idea Jul 04 07:35:35 I invested at least a month trying to get 2.6.xx going for my project. USB ended up being unworkable, so eventually I ended up going back to 2.4.xx Jul 04 07:35:44 03rwhitby@nslu2-linux.org 07org.openembedded.nslu2-linux * rd6fdb02a91d1c19b5e0e6c9c0e484f3e14b2ca77 10/ (21 files in 16 dirs): Jul 04 07:35:44 propagate from branch 'org.openembedded.dev' (head b378da256e147a5c55663f95c0cc3df118a4a63d) Jul 04 07:35:44 to branch 'org.openembedded.nslu2-linux' (head e04a7f684ef5855d717ea0e66d63fd31465c451a) Jul 04 07:35:57 It works ;) Jul 04 07:36:08 whee! monotone messages :) Jul 04 07:36:41 yep, should be working now Jul 04 07:36:55 it runs every 5 minutes Jul 04 07:37:02 cool Jul 04 07:37:27 did you guys see my monotone issue earlier today? Jul 04 07:38:11 monotone: misuse: no unique private key for cert construction Jul 04 07:38:28 oh well.. maybe should try making some ugly fix in SANE... i really dont want to go to 2.4 Jul 04 07:38:33 03nail@nslu2-linux.org 07org.openembedded.nslu2-linux * rfc7a16d864477c5d989f50430896d4cf6df31265 10/packages/meta/openslug-packages.bb: Added working openntpd. Jul 04 07:38:49 *ding-ding* Jul 04 07:38:54 was there anything I could have done to resolve that, or is it just a matter of waiting? Jul 04 07:39:33 peteru-home: you needed to create a private key, so that you could resolve the merge. Jul 04 07:40:49 but it's whoever left the repo unmerged who should have fixed it Jul 04 07:42:13 rwhitby-away: but, if I resolve the merge myself, will that mean that I have a local fork of the file in question? Jul 04 07:42:36 and does not that create the possibility that this will bite me later on? Jul 04 07:43:09 Yoda grammar I suffer from. ;) Jul 04 07:43:37 create the possibility is Jul 04 07:43:45 03nail@nslu2-linux.org 07org.openembedded.nslu2-linux * ra96120ae4432f090395e5fa91129bb55062af8ee 10/ (11 files in 8 dirs): Jul 04 07:43:45 merge of d6fdb02a91d1c19b5e0e6c9c0e484f3e14b2ca77 Jul 04 07:43:45 and fc7a16d864477c5d989f50430896d4cf6df31265 Jul 04 07:46:47 ok, I'm off to push some Zzzs. Good night all. Jul 04 07:49:07 night peteru-home Jul 04 07:52:50 nslu2-linux still has an average of 2.25 hours between commits for the last 0.81 years. Jul 04 07:53:21 Ok, I'm on it :P Jul 04 07:54:23 merges will be filtered out now. propagates will remain Jul 04 07:54:31 yeah Jul 04 07:54:48 and the author will be cut at the @ Jul 04 07:55:07 Is it necessary to have the whole hash, or can we do like they do in #oe, cut it off a bit? Jul 04 07:55:18 actually.. the whole hash is a good idea Jul 04 07:55:44 why? Jul 04 07:56:35 Isn't it? I thought it might be useful for looking up stuff in the log. Jul 04 07:57:05 the first 8 digits should give you a fairly unique hash Jul 04 07:57:18 yeah Jul 04 07:58:05 ok, it's now the same as #oe Jul 04 07:58:59 hmm - do we really want propagates listed here? Jul 04 08:00:05 since I'm not sure I know enough python to fix it, then we will :-) Jul 04 08:00:25 (until we start doing it hourly maybe) Jul 04 08:01:04 haha Jul 04 08:01:32 Well.. It is kinda neat to know the number of files modified by a propagate Jul 04 08:02:02 night all Jul 04 08:02:21 Good idea. Think I'll pick up a few ZZzzz's myself. Jul 04 08:05:45 propagating back to #oe Jul 04 08:10:08 now that is one complex graph: http://monotone.vanille.de/viewmtn/revision.psp?id=a96120ae4432f090395e5fa91129bb55062af8ee&ancestry_limit=0 Jul 04 08:10:42 haha Jul 04 08:11:31 I got a yucky orange color this time though. I liked my green much better :P Jul 04 08:11:31 I think we need to start up the propagate-from-oe every hour sometime soon ... Jul 04 08:11:48 hmm Jul 04 08:11:58 Won't that result in unmerged heads? Jul 04 08:12:22 That kinda screws up users trying to build bleeding edge Jul 04 08:13:33 you saw what just happened in #oe? ;) Jul 04 08:13:38 yep Jul 04 08:14:02 That is gonna make us very popular if it comes in large batches Jul 04 08:14:08 NAiL: the monotone user on nudi can do automatic merges Jul 04 08:25:12 monotone question: Is monotone similar in concept to BK? As in, does it keep a local repo and all that? Jul 04 08:25:27 VoodooZ_Work: yep Jul 04 08:25:33 cool. Jul 04 08:25:43 http://venge.net/monotone/ Jul 04 08:26:45 Do you think I could just copy my old openembedded folder in the new monotone tree to get me started with my private changes? Jul 04 08:26:50 Thanks for link. Jul 04 08:27:15 yep, you should be able to copy the files over. make sure you remove the SCCS files, and check for adds and deletes Jul 04 08:27:43 yeah. I was going to write a quick script to clean up all the SCCS junk. Jul 04 08:27:49 check for adds and delete? Jul 04 08:28:38 yeah, you may have added files, and they may have deleted files you use Jul 04 08:30:07 yeah. It's temporary. A lot of my personal changes have been added by either me or other members so there's no that many left. Jul 04 08:30:32 The most imporant being the dead I/O pin reroute and the tweaking of openslug.conf Jul 04 08:47:18 ACTION  Jul 04 08:52:17 rwhitby-asleep: monotone propagate should help reduce the branching, or at least make it more localised. Jul 04 08:52:32 ha, OLS schedule Jul 04 08:55:24 slightly offtopic, but does anyone here have experience/interest in development on Synology DS-101? Jul 04 08:56:07 <[g2]> jaxe, minor but they are not readily available in the states Jul 04 08:56:38 that is an issue indeed Jul 04 08:57:49 we already have full telnet+root access to the box. it has an integral IDE connector, same CPU and twice the memory of nslu2 Jul 04 08:58:06 * [g2] is building custom hw Jul 04 08:59:49 the box runs same binaries as nslu2 Jul 04 08:59:59 <[g2]> it will Jul 04 09:00:14 what are you planning? Jul 04 09:00:15 <[g2]> in addition to the usb2 it'll have a CF slot Jul 04 09:00:36 <[g2]> so the flash on board will be less, but you can boot to the CF Jul 04 09:00:46 <[g2]> plus have 2 USB 2.0 ports Jul 04 09:01:49 what kind of costs are you looking at? Jul 04 09:02:25 <[g2]> it's a little expensive compared to the slug but I think it'll be well worth it Jul 04 09:02:27 <[g2]> $249 Jul 04 09:03:03 <[g2]> IXP422, 64MB, .5MB bootflash, CF, miniPCI, dual etherenet Jul 04 09:03:20 <[g2]> serial port out of the box, JTAG header inside Jul 04 09:03:27 <[g2]> full metal chassis Jul 04 09:03:39 <[g2]> plus other goodies Jul 04 09:06:40 g2: have any pictures yet? Jul 04 09:06:50 anyways, do you gyus think that there would be any interest in the nslu2 community for collaboration with the 'small but emerging' synology community?-) Jul 04 09:07:07 <[g2]> glc_, proto boards will be back in few weeks Jul 04 09:07:59 <[g2]> As the specs are finalized I'll publish the full docs and launch the product Jul 04 09:08:21 <[g2]> I'll be doing a 200 unit production run Jul 04 09:12:47 <[g2]> AHH finally... Apache2 with PHP5 running on OpenSlug Jul 04 09:42:07 hm.. I would want the features of Synology Disk Station DS-101g+, at the price and size of the nslu2 Jul 04 09:44:23 you would need a small 3.5" hard disk then:) Jul 04 09:44:36 [g2]: $249 isn't actually that much, for the DS-101g+ I would have to give $315 in sweden Jul 04 09:44:53 <[g2]> jaxe, or CF Jul 04 09:45:00 hehe, well.. how about a 2,5" hdd, of CF Jul 04 09:45:02 or* Jul 04 09:45:03 <[g2]> I think a little microdrive Jul 04 09:45:23 <[g2]> there's some extra features too Jul 04 09:45:30 dual lan is just great Jul 04 09:45:40 <[g2]> nod. Jul 04 09:45:59 btw, anyone tried 802.1q on the slug? Jul 04 09:46:02 <[g2]> I've done the best I could on features/price/performance Jul 04 09:46:42 <[g2]> DaKa2, I've had 3 interfaces on a slug running but didn't get a chance to play with transparent bridging Jul 04 09:46:45 I'd say you did a good job Jul 04 09:47:04 <[g2]> well thx Jul 04 09:47:07 :) Jul 04 09:47:29 <[g2]> I'll be launching the company and web stte soon Jul 04 09:47:34 <[g2]> site Jul 04 09:49:07 nice, I might actually want one, or two, dunno yet Jul 04 09:49:09 <[g2]-away> bbiab Jul 04 10:23:56 hello Jul 04 10:24:26 any wl-hdd users here, using oleg firmware? Jul 04 10:25:48 i am a bit lost what to do after flashing.... Jul 04 10:55:06 03repvik * r56 10/trunk/openslug/nslu2-linux/packages/openntpd: Added openntpd Jul 04 10:56:14 * NAiL goes "neat". Jul 04 10:57:02 03bzhou * 10unslung/make/ion.mk: upstream upgrade from 20050607 to 20050625 Jul 04 11:15:03 how do i mount the internal hdd on my wl-hdd? Jul 04 11:15:33 mount -t ext3 /dev/ide/host0/bus0/target0/lun0/part1 /tmp/harddisk/part1 Jul 04 11:15:33 mount: Mounting /dev/ide/host0/bus0/target0/lun0/part1 on /tmp/harddisk/part1 failed: Invalid argument Jul 04 11:54:03 03ov2k * 10unslung/make/openvpn.mk: Added -fno-inline to openvpn.mk Jul 04 11:55:59 03ov2k * 10unslung/sources/openvpn/S20openvpn: Jul 04 11:56:00 Changed startup script from Jul 04 11:56:00 grep -q "^tun" `lsmod` Jul 04 11:56:00 to Jul 04 11:56:00 lsmod | grep -q "^tun" Jul 04 11:56:00 as suggested on the mailing list. Jul 04 12:08:07 03ov2k * 10unslung/Makefile: Added openvpn to READY_FOR_TESTING Jul 04 13:47:34 Hi Tiersten? Jul 04 13:52:08 hmm? Jul 04 13:53:04 I asked a question about the wish list on the mail list Jul 04 13:53:14 But nobody responded. Jul 04 13:53:25 erm okay Jul 04 13:53:29 I found out that there are two wish lists? Jul 04 13:54:02 You are in the list as wiki master so maybe you know the reason. Jul 04 13:54:18 There are multiple ones because they're for different things Jul 04 13:54:34 HowTo/WishList is just for general - I wish it could do Jul 04 13:54:53 Unslung/FirmwareWishList for Unslung specific ones Jul 04 13:55:29 etc... Jul 04 13:56:04 There is a packagewishlist as well Jul 04 13:56:08 What did you modify? Jul 04 13:57:07 First of all i find it hard to decide what is really usefully to do with that many lists. Jul 04 13:57:27 I didn't actually start the other ones Jul 04 13:57:40 They just generally appeared by themselves as people decided that we needed specific lists Jul 04 13:57:46 The second problem is that if you made a package you have to correct all the lists that the package is there. Jul 04 13:58:12 Afraid so Jul 04 13:58:26 A wiki isn't the best thing to keep track of these things just because stuff gets out of date quickly Jul 04 13:58:43 Need to go through it at some point and strip out the obsolete or redundant information Jul 04 13:59:38 It would be useful for me if someone decides which packages we really want to support. Jul 04 14:00:03 Some times it's better that someone makes a decission. Jul 04 14:00:17 Ignore the wishlists if so. Stuff in the feeds in theory should be supported Jul 04 14:00:43 Okay. I've found your email Jul 04 14:01:08 hmm. I guess we'd need some sort of voting system or just tell users to put a "Me too" under a specific package Jul 04 14:01:29 I think something like that would be useful. Jul 04 14:02:24 I am prepared to support some more packages but i don't really know what a group of users want. Jul 04 14:02:52 well. the main push so far is that the package maintainer wants it Jul 04 14:02:56 I don't want to put a package in the system beacause some user wrote it in a wishlist a time ago. Jul 04 14:03:07 they're just helpful enough to keep maintaining it for everybody as well Jul 04 14:03:25 There are two problems with that Jul 04 14:04:11 1) Some of those maintainers going to stop supporting those packages. At that moment they are a burden for the project. Jul 04 14:05:21 2) There is no guarantee that packages which a lot users want are build. Jul 04 14:05:51 I hoped to get a discussion about these topics to get a better solution. Jul 04 14:06:14 How do you get around 1 or 2? Jul 04 14:07:00 Unless we can find somebody willing to maintain it, we can't add it in. It's just purely man power issues here. Jul 04 14:07:43 A feedback system which really counts wishes could be useful. Jul 04 14:09:07 I'll have a look into it. If it is really something that will be useful then I'll write it Jul 04 14:09:09 On the other side somebody should decide if it fits in the total solution and if it's word the evert. Jul 04 14:09:46 And the man power issue is something which will work to both sides. Jul 04 14:10:36 If we make better decision's which packages we really want it can be easer to maintain. Jul 04 14:10:52 *shrug* Jul 04 14:11:05 You should really discuss this with rwhitby Jul 04 14:11:21 jacques, rwhitby-asleep: looking at packages/openslug-init/*/kern_header it generates kernel-file-size+16 as the length, but slugimage seems to generate kernel-file-size - is that an error in slugimage? Jul 04 14:11:33 Ok. Jul 04 14:58:21 There were originally 2 lists. Firmware Wishlist and Package Wishlist. Jul 04 14:59:20 03florian 07org.openembedded.dev * r33378305... 10/ (4 files in 4 dirs): Jul 04 14:59:20 Move responsibility for setup links to extra package. Jul 04 14:59:20 Default to osso gtk version in maemo images. Jul 04 14:59:20 03florian 07org.openembedded.dev * r23bb3a52... 10/conf/machine/nokia770.conf: Update to new udev naming. Jul 04 14:59:21 03florian 07org.openembedded.dev * r2d388161... 10/packages/maemo/osso-esd_0.5.bb: Limit RPROVIDES to the binary package. Jul 04 14:59:22 03florian 07org.openembedded.nslu2-linux * r041417bf... 10/packages/maemo/ (nokia770-init/fixup-770.sh nokia770-init_1.0.bb): Jul 04 14:59:25 Add package to do some initialisation and fixups like unmounting stuff Jul 04 14:59:27 from initrd and create some useful links. Jul 04 15:06:39 florian changes - the logs come from my syncing a propagate Jul 04 15:09:06 jaxe: there is interest - maybe we can do an Optware port like we have done for wl500g ... Jul 04 15:12:05 jbowler-away: well, we know that slugimage works, and kern_header works too. So I'd guess that slugimage is correct. Jul 04 15:12:06 jaxe, I split the original firmware for the DS101 a while ago; so there is a little interest, but I cant get my hands on one. Jul 04 15:12:45 rwhitby-asleep: and kern_header is wrong? Jul 04 15:13:33 The last 16 bytes of the image are probably typically not interesting - removing them may not cause an error. Jul 04 15:14:02 reflash currently does what kern_header does, but it's easy to make it do what slugimage does, just slightly worrying to reduce that number by 16... Jul 04 15:15:40 rbwitby: Did you read my discussion with Tiersten? Jul 04 15:22:46 jbowler-away: I guess kern_header is being conservative. slugimage hasn't changed forever, and we know that it works Jul 04 15:23:07 if redboot copies 16 too many bytes, it's not a problem - that's why kern_header works Jul 04 15:24:51 Yes, it looks like I have assume that the flash image file corresponds to what slugimage does anyway. Jul 04 15:25:54 marceln: pick 5 or 6 packages which you are interested in. Start a poll on the yahoo group to see which of those is most wanted. The development of packages on this project really is driven by the needs of the developers. Jul 04 15:26:44 alternatively, put a dividing line on the wiki pages (everything below this line is ignored) and ask people to update the page with their latest wishlist, and ask them to put their name after the wish and multiple names will tell you which are most popular. Jul 04 15:27:36 Ok. We don't need a solution now. But i think we should think about it. Jul 04 15:28:49 I think the project gets better if we gone make some decisions. Jul 04 15:29:13 Otherwise i see some problems which won't be solved. Jul 04 15:29:41 marceln: depends on your point of view :-) I personally am here for the developers first, and then the users are welcome to recieve the fruits of the developers' labour. So if developers have problems, then those get solved first. Jul 04 15:29:46 e.g my problem with aftpd but also busybox 1 which conflicts with to many other packeges. Jul 04 15:30:45 The other source of input is slugbug. If something is reported well in slugbug, then we'll usually discuss it here and fix it. Jul 04 15:34:22 but you are right. the project as a whole would benefit from a group of "second tier" support developers who are focused on the end-user problems instead of focused on driving development forward. Jul 04 15:35:06 Ok. It's maybe it's beacause i am new in this project but it bothers me a litle bit that there isn't enough central managment. Jul 04 15:36:35 marceln: it's intentionally so. The core team manages the direction of the firmware pretty well, but intentionally leaves the management of packages to the core team members who are the package managers. But it is not part of the package managers job to direct which packages should be developed. Jul 04 15:37:25 I guess we have never had anyone before who didn't have a personal itch to scratch, and was looking to develop and maintain packages that other people needed ... Jul 04 15:37:56 So the growth of the packages have happened by developers scratching personal itches, and jp30 managing the collection of those itches. Jul 04 15:38:53 03rpurdie 07org.openembedded.dev * rfdf14f17... 10/packages/sharp-binary-only/sharp-aticore-oss_1.0.1.bb: Update atilib to avoid a segfault Jul 04 15:38:53 03rpurdie 07org.openembedded.dev * r8f8edb0b... 10/packages/minicom/minicom-2.1/configure.patch: minicom: Fix configure to correctly use POSIX_TERMIOS Jul 04 15:38:54 There are a lot of packeges i am not interested at all. Jul 04 15:39:14 I wouldn't develep those packages. Jul 04 15:39:32 But i don't wan't to push a package i only use. Jul 04 15:39:58 when SANE works I could fix/maintain other things, if there is interest Jul 04 15:40:02 And i love to contrib a package everyone loves and is in my expertise. Jul 04 15:41:09 I hope i didn't waste to much time but instead started a useful discusion. Jul 04 15:41:23 But anyway it's sleeping time for me. Jul 04 15:41:26 Bye. Jul 04 15:41:29 marceln: it is a good discussion Jul 04 15:41:33 let's do something about it Jul 04 15:41:54 what do you suggest is the best way to get the user input on which packages are important to them? Jul 04 15:43:00 That's really a problem. Probably most users will never respond unless they have a hard problem. Jul 04 15:43:37 how about a signup page, ie, leave your email on a page for the package/project they want fixed, with the option of beeing notified when things work? Jul 04 15:43:38 Polls might give some input but still miss most users. Jul 04 15:45:57 marceln: your new job is to work out the best way to gather the input using the community infrastructure (wiki, mailing list, yahoo group, slugbug) that we have already. Let me know what you decide, and we'll help you implement it. Jul 04 15:46:37 I will think about it. But first i gone a write the paper for the linux-kongress. Jul 04 15:46:43 if it absolutely can't be done with existing infrastructure, then at that point we can look at developing or installing something else. Jul 04 15:47:17 marceln: right. BTW, we appreciate the work you are doing on the paper. Did you contact ka6sox ? Jul 04 15:47:30 Not yet. Jul 04 15:47:45 he has done a couple of these "Introduction to Unslung" style talks before, and probably has some slides you can start from if you wish. Jul 04 15:47:45 I hope to find time next weekend to write the paper. Jul 04 15:48:24 First off all i have to write an abstract which they use to decide which talks are admitted. Jul 04 15:48:46 feel free to email me if you want me to help proof read it. I will ask that I get to see it before you present it, in case there are any factual errors. Jul 04 15:49:14 I will. Jul 04 15:49:52 Thanks. Jul 04 15:50:09 No, thank you :-) Jul 04 15:50:33 Bye i have to get sleep now. Jul 04 17:23:30 03jbowler 07org.openembedded.nslu2-linux * rbfc49ece... 10/packages/openslug-init/ (openslug-init-0.10/reflash openslug-init_0.10.bb): Jul 04 17:23:30 Allow kernel and root partition to be updated in the flash alone, i.e. Jul 04 17:23:30 just the kernel or just the root partition. Add support for progress Jul 04 17:23:30 indicator (requires new rev of devio). Jul 04 17:29:17 Seems like after upgrading unslung 3.18 to 5.5 the size of disk2/Flash is 0MB in the webpage (df in a shell reports OK) Jul 04 17:30:55 Tried with 2 different disks aswell, same with both of them. Its not really a big error but migth scare some folks after the upgrade (in the check that your disk is recogniced part of the readme). Jul 04 17:32:01 It is in the FAQ. Jul 04 17:34:49 So true, point 43. Smart of me not to reload when I leave the browsertab for weeks on end pointing to pages =/ . Thanks for pointing that out to me. Jul 04 18:00:33 is anyone alive? Jul 04 18:02:52 when i do a make setup, then make i end up with the following error: Jul 04 18:03:03 glib-ectomy.c: In function `g_io_channel_read': Jul 04 18:03:13 glib-ectomy.c:22: error: `LONG_MAX' undeclared (first use in this function) Jul 04 18:03:46 no matter how i do it, ive tried the stepbystep, and now i tried the makefile Jul 04 18:04:37 what *exactly* are you doing Jul 04 18:04:51 you should be able to use the master makefile. Jul 04 18:05:07 mkdir foo; wget http://www.nslu2-linux.org/Makefile; cd foo; make setup; make Jul 04 18:05:20 thats what i did Jul 04 18:05:40 it fails on the exact same place each time, in bluez-utils-nodbus Jul 04 18:06:02 the rest compiles fine Jul 04 18:08:00 do you need to be making openslug-packages right now? Jul 04 18:08:42 lol, i just got home from work, was just curious if someone had seen this before (it had been compiling while i was at work) Jul 04 18:09:14 i can wait till morning ;-) Jul 04 21:52:11 rwhitby-asleep, dyoung-zzzz, good to hear that i'm not the only one:) Jul 04 22:24:34 CHAOSiTEC_away: bluez-utils-nodbus still compiles fine on my box. I did clean build from scratch. Jul 04 22:34:17 back...sorta... Jul 04 22:34:25 must...have....coffee.... Jul 04 22:35:37 rwhitby-asleep, dyoung-zzzz, i would be happy with help in getting dropbear running for starters. optware would of course be fabulous Jul 04 22:35:57 is this on openslug? Jul 04 22:36:13 ka6sox, i'm talking about synology ds-101 Jul 04 22:36:36 jaxe, cool! another optware branch Jul 04 23:03:51 dyoung-zzzz, what exactly have you done so far? do you have some sort of development environment set up? Jul 04 23:18:25 I dont have a DS-101 Jul 04 23:18:51 someone told me it was too difficult to take apart the firmware image. Jul 04 23:28:32 03jbowler 07org.openembedded.nslu2-linux * rc359b835... 10/packages/boost/ (boost-jam-native_3.1.9.bb boost_1.32.0.bb): libboost, builds but untested Jul 04 23:31:06 jbowler-away, this is for monotone? Jul 04 23:31:29 yes Jul 04 23:37:04 dyoung-zzzz, how do you mean? Jul 04 23:38:26 now that there is source code available as well as full root access to the running system Jul 04 23:42:56 jaxe, what method for "root" access? Jul 04 23:43:49 telnet Jul 04 23:45:47 I looked at it the 2nd week after it was released. Jul 04 23:46:13 ka6sox, the board has serial and JTAG headers. Jul 04 23:48:48 dyoung-zzzz, sweet...oh..thats right you have pictures. **** ENDING LOGGING AT Mon Jul 04 23:59:56 2005