**** BEGIN LOGGING AT Mon Jul 11 23:59:57 2005 Jul 12 00:08:35 03rwhitby 07org.openembedded.nslu2-linux * r2613c747... 10/conf/distro/openslug.conf: Bumped OpenSlug version to 2.2-beta. Jul 12 00:09:47 That's because we have moved to a new kernel version, so the modules are incompatible ... Jul 12 00:44:35 Is anyone here? Jul 12 00:48:35 yes Jul 12 00:49:15 you betcha Jul 12 00:51:06 whee Jul 12 00:54:49 BlueSlug is back up and running on OpenSlug 2.2 ... Jul 12 00:54:59 neat Jul 12 00:56:37 now to get the zd1211 drivers working on openslug so I can have a RemoteSlug ... Jul 12 01:01:34 zd1211? Jul 12 01:04:27 its a wireless chipset that appears in some USB dongles. Jul 12 01:05:29 did the zd1211_foo.bb I made up make it through? I wonder if theres a newer version... Jul 12 01:05:37 * dyoung searches Jul 12 01:35:43 hm, progress in my libpam bug Jul 12 05:24:47 03rwhitby 07org.openembedded.nslu2-linux * ra042d7f2... 10/ (3 files in 3 dirs): Jul 12 05:24:47 propagate from branch 'org.openembedded.dev' (head 8350fd2206abaeaa4985840a85d7995323570d78) Jul 12 05:24:47 to branch 'org.openembedded.nslu2-linux' (head 2613c747e0f053ef97adbdc472319898cdd86cb1) Jul 12 06:15:32 I didn't think I would have to face this this early but I just encountered my first 3-way merge in monotone and I'm not too sure how to handle that merging tool. Jul 12 06:15:41 It's not emacs. Jul 12 06:15:47 is it vi? Jul 12 06:15:56 nope. Otherwise I'd be right at home. :)( Jul 12 06:16:15 :( I always get my lovley vi :-) Jul 12 06:16:16 I'll try it again to find out the tool's name Jul 12 06:16:44 I noticed that monotone said it couldn't find emacs, gvim and others. Jul 12 06:16:59 try exporting EDITOR=vi Jul 12 06:17:20 yeah, I'll add it to my setup-env if that works. Jul 12 06:17:21 and then running it, dont know if it folows that, but its worth a try :-) Jul 12 06:17:28 yeah Jul 12 06:18:52 strange, yum can't find gvim Jul 12 06:19:34 * VoodooZ_work hugs yum Jul 12 06:20:20 DaKa2, Actually, it was vim! Jul 12 06:20:25 :) Jul 12 06:20:28 hehe Jul 12 06:20:41 :wq! :q :q Jul 12 06:20:49 the interfaced was a bit different than mine though. I got scared It was emacs!! ;) Jul 12 06:20:54 :-) Jul 12 06:21:16 Nothing wrong with emacs... Jul 12 06:21:19 Still scary though. Having 3 windows and no filenames! Jul 12 06:21:30 I'm at no risk getting emacs.. not installed :-) Jul 12 06:21:38 no, there is nothing wrong iwth emacs Jul 12 06:21:45 with* Jul 12 06:21:49 Tiersten, I know. I just never used it so I feel like a Microsoft WIndows user encountering VI for the first time! :) Jul 12 06:22:03 Nothing wrong with vi either :) Jul 12 06:22:34 How do you force a redraw in vim again? Jul 12 06:22:38 Stop complaining otherwise I'll make you use ed :) Jul 12 06:22:47 hehehe Jul 12 06:23:06 nooooo Jul 12 06:23:35 VoodooZ_work: http://database.sarang.net/study/vim/HOWTO/Vim-HOWTO-9.html Jul 12 06:23:39 I'd rather use Notepad(TM)!!! :) Jul 12 06:23:48 thanks! that was fast! Jul 12 06:23:50 Okay. Soembody kick him now Jul 12 06:23:59 I'm a bit rusty with the multi-window stuff. Jul 12 06:24:16 * VoodooZ_work kicks himself in the head. Jul 12 06:26:30 [STUPID QUESTION WARNING!!] actually, what does the 3 way merge expect me to do? Jul 12 06:26:55 check if you want something from the other two, and fixing it in the one you want and sane that Jul 12 06:26:58 There's 3 files open with tmp* names so It's hard to tell what the hell is what. Jul 12 06:27:07 save* Jul 12 06:27:30 So I would need to update the local one? Jul 12 06:28:01 uhm.. I think you should only save one Jul 12 06:28:03 I"m confused because I would have expected only 2 files has all my changes are commited. Jul 12 06:28:31 ie, make the best out of the three, and just :wq! that one... but I dont know Jul 12 06:29:01 I do prefer's BK's way of merging. It had an option to just use the local/remote one. Jul 12 06:29:10 ok. Jul 12 06:29:17 I really dont know Jul 12 06:29:24 me neither so :) Jul 12 06:29:27 :) Jul 12 06:29:45 * VoodooZ_work approaches the keyboard shaking Jul 12 06:32:42 * VoodooZ_work scratches his head and considers reading the monotone documentation AGAIN! Jul 12 07:14:03 that damn monotone 3way merge thing is driving me nuts! Somebody knows how to deal with it? Jul 12 07:15:42 Can't I just tell it to ignore my changes and take the remote ones instead. Would be easier. Jul 12 07:17:06 just save the remote one then? Jul 12 07:17:39 That's part of the problem. It's hard to tell what's what. They all have tmp names. Jul 12 07:18:08 I read about vim's diff mode but that didn't help. monotone's page doesn't say much Jul 12 07:18:21 I don't want to mess up my repo more. Jul 12 07:18:22 hm.. bad.. for me the first one has been the remote one Jul 12 07:18:37 but Ive only had 3-way merges a few times Jul 12 07:18:50 ok. that helps. Jul 12 07:19:11 So I should be working on the right one then? Jul 12 07:19:30 dunno really.. I usually accept the first one.. Jul 12 07:19:44 ok. I'll try that. Jul 12 07:20:14 but as I said, I know nothing at all about this Jul 12 07:20:24 (that is, dont blame me if something goes wrong) Jul 12 07:20:28 ;)) Jul 12 07:20:38 don't worry. I'm the idiot at the keyboard! :) Jul 12 07:21:03 seems to work. It stopped bothering me. Jul 12 07:21:04 VoodooZ_work: It's efidd Jul 12 07:21:05 diff Jul 12 07:21:15 I'm supposed to write a howto on this Jul 12 07:21:16 what? Jul 12 07:21:22 ediff Jul 12 07:21:26 the window you got Jul 12 07:21:28 probably Jul 12 07:21:36 are you sure? It's vim diff mode. Jul 12 07:21:41 oh Jul 12 07:21:50 oh well.. then I don't know anything about how it works :P Jul 12 07:21:51 I get vim too Jul 12 07:22:14 but a monotone merge howto would be useful nonetheless as I probably messed up my repo big time now. Jul 12 07:23:38 yeah, but I've got no idea how vim diff mode works Jul 12 07:23:50 true. Jul 12 07:24:05 but I'm willing to switch to ediff. Jul 12 07:24:19 To prevent the ordeal I just went through. Jul 12 07:24:27 NAiL: feel like commiting some fixes? :-) Jul 12 07:24:34 DaKa2: Sure ;-) Jul 12 07:24:41 http://david.thg.se/saker/nytt/packages/ Jul 12 07:24:48 [with my best Yoda voice] not obvious at all, that experience was Jul 12 07:24:56 the tar.gz are the full things, the dirs are just the changed files Jul 12 07:25:08 xinetd == fixed init-script Jul 12 07:25:27 sane == added default saned xinetd config file Jul 12 07:25:29 VoodooZ_work: ediff is at best unintuitive. Jul 12 07:25:35 i c Jul 12 07:25:40 but rwhitby told me how to do it Jul 12 07:26:07 I'll take a look at your wiki page once it's done i guess. Jul 12 07:26:14 left window is A, right is B, bottom is result. Press n for next, p for prev, a or b to choose which version to put in the bottom window Jul 12 07:26:59 sounds simple enough. :( Jul 12 07:27:26 yeah, one you know that, it's quite simple actually Jul 12 07:27:38 DaKa2: Does this make sane-backends depend on xinetd? Jul 12 07:27:42 no Jul 12 07:27:45 just recommend Jul 12 07:27:46 good :) Jul 12 07:27:51 nice Jul 12 07:28:11 any chance xinetd could be added to openslug-packages? Jul 12 07:29:16 to unstable atleast, yes Jul 12 07:29:22 :) Jul 12 07:30:41 any chance on everything ending up in stable sometime? :-) Jul 12 07:31:02 You never know ;-) Jul 12 07:31:10 Im thinking about a sane howto Jul 12 07:32:29 but Ill just write that its for unstable, and that they might end up in stable sometime Jul 12 07:33:35 yeah Jul 12 07:35:10 03nail 07org.openembedded.nslu2-linux * r8b01c8dc... 10/packages/ (4 files in 4 dirs): Jul 12 07:35:10 fixed xinetd init-script (DaKa2) Jul 12 07:35:10 added default saned xinetd config file (DaKa2) Jul 12 07:36:43 NAiL: when do you think they will be in the unstable feed? Jul 12 07:37:32 When I rebuild later tonight Jul 12 07:37:56 nice, then Ill try and have some kind of howto up by then Jul 12 07:39:46 neat :) Jul 12 07:40:05 03nail 07org.openembedded.nslu2-linux * r9726cc21... 10/packages/meta/openslug-packages.bb: added xinetd to packages Jul 12 07:40:07 03nail 07org.openembedded.nslu2-linux * r18847e62... 10/packages/ (7 files in 7 dirs): Jul 12 07:40:07 propagate from branch 'org.openembedded.dev' (head 2d8db2772931911d5ac3027499e33219f7a60fe3) Jul 12 07:40:07 to branch 'org.openembedded.nslu2-linux' (head 9726cc2116293dfa15f6158657ae5a2cd4d2fcfd) Jul 12 07:47:32 <[cc]smart> hi Jul 12 07:47:36 hello Jul 12 07:47:46 <[cc]smart> just to make sure that if i'm going to look at /var i'm not doing duplicate work Jul 12 07:47:51 <[cc]smart> somebody on that alrady ? Jul 12 07:48:22 dont think so, NAiL? Jul 12 07:48:26 <[cc]smart> or is there maybe a page even where devs lt others know what they are on ? Jul 12 07:48:53 <[g2]> [cc]smart, yeah... the IRC channel :) Jul 12 07:49:28 [g2]: I feel very strongly for someone doing that change. I'd do it myself if I knew how (and nobody strongly disagreed) Jul 12 07:50:38 I've only messed with packages, not the core system itself Jul 12 07:51:04 I've looked at turnup a bit, its really simple Jul 12 07:51:04 (probably a nice thing to learn, when I've got more time) Jul 12 07:51:31 Is turnup the only thing that needs changing? Jul 12 07:51:41 yes Jul 12 07:51:42 should be Jul 12 07:52:05 <[cc]smart> ? Jul 12 07:52:06 <[cc]smart> nana Jul 12 07:52:24 just needs modification to setup_var() and setup_rootfs()? Jul 12 07:52:25 ? Jul 12 07:52:25 <[cc]smart> the whole topic i'd say is about /var on tmpfs and the magic is in the /etc/init.d Jul 12 07:52:46 <[cc]smart> turnup would just be a little switch i'd guess Jul 12 07:53:01 yeah, and var is setup in turnup Jul 12 07:53:04 [cc]smart: /var when running from flash only, should be tmpfs, right? Jul 12 07:53:45 <[cc]smart> for what we've discussed, the idea was to have /var per se on disk/flash and /var/tmp on ram. then link /var/log into /var/tmp Jul 12 07:54:00 <[cc]smart> my opinion would be to stick to that scheme over all use cases. Jul 12 07:54:13 <[cc]smart> that way, all situations are per se the same Jul 12 07:54:25 <[cc]smart> and tests on disk do not behave different than users on flash Jul 12 07:54:55 <[cc]smart> having all on disk is certainly legit, too. but would in my opinion be for the wizards Jul 12 07:54:56 just nfs and disk have /var/log and /var/run on disk Im happy Jul 12 07:55:03 <[cc]smart> that do their own magic Jul 12 07:55:46 [cc]smart: Having the rest of /var on flash when running from flash isn't such a good idea either. Then we'll end up with users filling up their flash with ipkg update Jul 12 07:55:47 <[cc]smart> it would prolly be nor more than replacink links with directories Jul 12 07:56:08 <[cc]smart> where does ipkg update write to ? Jul 12 07:56:17 /var/cache Jul 12 07:56:19 I think Jul 12 07:56:37 <[cc]smart> nice, just another item on the linklist Jul 12 07:56:40 /var/lib actually Jul 12 07:56:47 <[cc]smart> THATs worse Jul 12 07:56:55 thats the fhs Jul 12 07:56:59 <[cc]smart> does it have an own directory there ? Jul 12 07:57:02 yup Jul 12 07:57:09 /var/lib/ipkg Jul 12 07:57:10 <[cc]smart> THATs better Jul 12 07:58:11 ugh,... I've gotta go take a nap. I'll get back to this a bit later Jul 12 07:58:39 I feel like taking a nap too... its too hot here Jul 12 07:58:49 <[cc]smart> happy reusting Jul 12 07:58:57 33degC... ugh.. Jul 12 07:59:05 <[cc]smart> lucky you Jul 12 07:59:09 Im melting Jul 12 07:59:12 <[cc]smart> :) Jul 12 07:59:27 <[cc]smart> diabel the turbo mod Jul 12 07:59:37 :)) Jul 12 08:00:24 disable my 12 disk fileserver would be better Jul 12 08:00:54 DaKa2, where's that? It's going to be like 38 with humidex here in Ottawa, Canada Jul 12 08:01:10 sweden.. Jul 12 08:01:48 I hate humidity! I'm a hockey player so I don't mind winter one bit. Jul 12 08:02:21 I want maximum 25deg, preferably 18, and winter is perfect Jul 12 08:03:25 and everybody else in sweden just wants it even hotter... Jul 12 08:03:38 <[g2]> [cc]smart, NAiL DaKa2 some thoughs on the /var issue can be written up in an e-mail or on the wiki where ppl can discuss it Jul 12 08:04:14 Ya, e-mail would be nice Jul 12 08:04:28 yup.. Jul 12 08:04:41 well, starting complete rebuild, bbl Jul 12 08:05:02 <[g2]> DaKa2, did you wipe tmp ? Jul 12 08:05:46 yeah. with my allergies summer and humidity sucks but I do like to ride my motorcycle though. Jul 12 08:05:53 [g2]: yep Jul 12 08:06:06 <[g2]> are you capturing the make output to a file ? Jul 12 08:06:20 hm. no... Jul 12 08:06:25 should I? Jul 12 08:06:33 <[g2]> then you could run my tiny perl script an update the openslug packages status page Jul 12 08:07:03 ok, how? make > file ? or &> file ? Jul 12 08:07:28 <[g2]> make openslug-packages 2>&1 | tee build.log Jul 12 08:07:40 ok Jul 12 08:07:59 <[g2]> since the I/O is buffered the screen pauses for a while sometimes Jul 12 08:08:14 <[g2]> then afterwards just script build.log Jul 12 08:08:15 would it matter if I ran it with 'time make openslug-packages 2>&1 | tee build.log' Jul 12 08:08:18 <[g2]> then afterwards just script build.log > foo Jul 12 08:08:36 <[g2]> no that's what I do and the time get noted at the end :) Jul 12 08:08:44 :) Jul 12 08:09:04 <[g2]> I'm going to update the script to pull that out at some point Jul 12 08:09:32 "make: *** No rule to make target `openslug-packages'. Stop.", uhm.. just make? or bb? Jul 12 08:10:20 <[g2]> either make openslug-build or cd openslug; . setup-env; time ..... Jul 12 08:11:10 ok Jul 12 08:11:27 where can I find the script once the build is finished? Jul 12 08:11:42 <[g2]> OH... mgetty is borked! Jul 12 08:12:26 <[g2]> If you're running as a normal user its tries to install into /usr/..... BAD.... Jul 12 08:12:38 very very bad Jul 12 08:12:54 <[g2]> that needs to be moved to broken and the install dir fixed Jul 12 08:13:50 <[cc]smart> hmm... irritating... somebody knows where "/etc/init.d/populate-var.sh" comes from ? Jul 12 08:17:14 <[g2]> [cc]smart, ./base-files-3.0.14-r35/install/base-files/etc/init.d/populate-var.sh Jul 12 08:17:24 <[g2]> that'd be my guess Jul 12 08:17:40 <[cc]smart> smart@center /data/work/nslu2-linux/openembedded $ find . -name populate-var.sh Jul 12 08:17:40 <[cc]smart> smart@center /data/work/nslu2-linux/openembedded $ Jul 12 08:22:37 <[cc]smart> i love this :) Jul 12 08:24:18 [cc]smart: find . -name "populate*" Jul 12 08:24:59 ah, and you're not in the right dir :P Jul 12 08:25:02 aren't you supposed to use wildcards though? "*populate*" Jul 12 08:25:10 <[cc]smart> you tried it ? Jul 12 08:25:25 <[cc]smart> don't need to :) Jul 12 08:25:30 <[cc]smart> 0 hits Jul 12 08:25:56 [cc]smart: tmp/work/base-files-3.0.14-r35/install/base-files/etc/init.d/populate-var.sh Jul 12 08:25:57 <[cc]smart> you can use wildcards Jul 12 08:26:07 <[cc]smart> hm Jul 12 08:26:16 <[cc]smart> so there must be an archive or sth. it is in Jul 12 08:26:21 yup Jul 12 08:26:40 look at it, do a patch, add that in the base-files dir Jul 12 08:27:44 <[cc]smart> it's echoed out from base-files_3.0.14.bb Jul 12 08:27:50 <[cc]smart> this horrifies me a bit Jul 12 08:28:17 That's not exactly a good solution Jul 12 08:28:22 It should be a patch Jul 12 08:28:26 <[cc]smart> seems bad practice, and seem to influence many platforms/distros Jul 12 08:28:42 <[cc]smart> i feel i have not enough mana for this spell Jul 12 08:29:09 <[cc]smart> should be discussed over entire OE then Jul 12 08:29:11 <[cc]smart> i guess Jul 12 08:29:27 talk to mickeyl in #oe Jul 12 08:29:32 <[cc]smart> me ? Jul 12 08:29:37 why not? Jul 12 08:29:46 I'm trying to get a nap :-P Jul 12 08:29:52 <[g2]> he's also in #openslug Jul 12 08:30:03 <[cc]smart> its' not that i had the feeling i would need that from the start Jul 12 08:30:07 [cc]smart: It's posssible to do a change for openslug only Jul 12 08:30:09 <[cc]smart> i'd just have used /data Jul 12 08:30:18 <[cc]smart> now i shall say why I want this change ? Jul 12 08:30:20 <[cc]smart> how can i ? Jul 12 08:30:48 <[cc]smart> who proposed this originally ? Jul 12 08:31:20 good question. might've been me. Or someone else entirely. Jul 12 08:31:27 <[cc]smart> not that i'd argue it again, it hink it's good. but the officer should stay in the first line. that's why hes there :) Jul 12 08:33:15 oh-my-god Jul 12 08:33:23 * NAiL just took a look at the bb Jul 12 08:33:31 <[cc]smart> :D Jul 12 08:33:44 <[cc]smart> hot stuff Jul 12 08:33:48 yes ;) Jul 12 08:35:03 <[cc]smart> i especially fancy the dirsXXX variables Jul 12 08:35:33 <[cc]smart> i mean, besides the method as such Jul 12 08:36:10 <[cc]smart> seems like flyfishing for cows Jul 12 08:36:17 haha Jul 12 08:36:50 <[cc]smart> and *SMACK* hit ya ! Jul 12 08:37:02 What's the default state of /var when booting from flash? Jul 12 08:37:09 All of it is tmpfs, right? Jul 12 08:37:19 <[cc]smart> on tmpfs and populated as does this script Jul 12 08:37:21 <[cc]smart> yes Jul 12 08:37:25 <[cc]smart> all on tmpfs Jul 12 08:38:05 Hmm Jul 12 08:38:52 The simple way would probably be to create the openslug version of populate-var.sh (If it stays the same between our revisions), then just copy that over the echoed one on install Jul 12 08:39:13 (if building for openslug) Jul 12 08:39:39 <[cc]smart> but this would also be an irritating experience for anyone trying to figure once we fell in the war against (t)error Jul 12 08:39:46 The file is way too heavy for me to parse in the state I'm in ;) Jul 12 08:41:01 Well.. I propose someone makes the changes to turnup first, actually. When turnup'ing to disk, it'll stay the same as it is now. When turnup'ing to flash, parts of var is in ramfs Jul 12 08:41:18 s/flash/memstick Jul 12 08:43:12 I don't think fixing /var on flash has high priority, since that will only be used by a few users anyway, and they should be able to figure it out by themselves really.. Jul 12 08:44:27 Anyone strongly disagree? Jul 12 08:45:47 [cc]smart: ? Jul 12 08:47:19 I should have set up irssi to log... Jul 12 08:50:56 I'd say the same, flash users should know what they are doing anyway Jul 12 08:51:40 what needs fixing is memstick, and the right way to fix it is in turnup, where it sets up var for disk and nfs Jul 12 08:51:52 my thoughts exactly Jul 12 08:52:51 just add "setup_var "$new" memstic &&" on line 249 in turnup, and then fix setup_var() to handle that Jul 12 08:52:56 memstick* Jul 12 08:53:37 I'm not *entirely* sure how it is best implemented though. Mounting log,tmp,run as separate filesystems seems a bit overkill (And probably has a tad overhead as well, no?). I'd like the three directories to remain separate though, so we don't end up with one dir containing heaps of files Jul 12 08:53:47 [g2]: ping? Jul 12 08:54:06 is there any big overhead when mounting them separate? Jul 12 08:54:16 I don't know what the mount overhead is Jul 12 08:54:36 tmpfs grows as it needs right? so it should be like a page or two in overhead? Jul 12 08:55:10 looks like it's ~50k per mounted fs Jul 12 08:55:14 <[cc]smart> back Jul 12 08:55:18 <[cc]smart> reading... Jul 12 08:55:34 just linking run and log into tmp/run and tmp/log for memstick then, or something Jul 12 08:56:28 hmm Jul 12 08:56:39 best idea so far, imho Jul 12 08:56:59 I do like to be able to rm -rf * in tmp though :-P Jul 12 08:57:13 hm... thats bad.. Jul 12 08:57:27 <[cc]smart> welll linking is definitely the way Jul 12 08:57:46 <[cc]smart> avoiding to spam /var/tmp is preferable too Jul 12 08:58:04 <[cc]smart> so i'd suggest to go back even more Jul 12 08:58:11 Mounting a tmpfs somewhere (/mnt/.tmpfs?) then adding dirs there, and symlinking from var? Jul 12 08:58:26 <[cc]smart> have /tmp as tmpfs only and link all the others including /var/tmp from there Jul 12 08:58:35 <[cc]smart> or like you said Jul 12 08:58:50 but not in /mnt, lots of people use that for random mounts Jul 12 08:59:05 DaKa2: /media/ram already exists Jul 12 08:59:17 and is tmpfs Jul 12 08:59:26 that would be good Jul 12 08:59:29 yes Jul 12 08:59:45 Providing it is the same when running from usbstick Jul 12 08:59:52 I'm running from disk Jul 12 08:59:57 <[cc]smart> is media RAM in FHS (not that i woud require it, but would be nie to know what it's meant for if there ) Jul 12 09:00:30 filesystem.org/fhs Jul 12 09:00:33 eh Jul 12 09:00:39 pathname.com, even Jul 12 09:00:42 * NAiL messes stuff up Jul 12 09:01:38 Actually, ram shouldn't technically (according to fhs) be mounted in /media, since it's not removeable :P Jul 12 09:01:56 <[cc]smart> er Jul 12 09:02:37 section 3.11 of fhs Jul 12 09:02:41 anyway Jul 12 09:02:55 <[cc]smart> ah, will use it Jul 12 09:03:12 I don't think /media/ram is actually used by anything in OpenSlug Jul 12 09:05:11 a ramdisk is "removable" Jul 12 09:05:43 by desoldering? ;-) Jul 12 09:05:56 no, by unloading the specific kernel module :) Jul 12 09:06:00 heh Jul 12 09:06:13 hm.. played around a bit, something like this maybe: http://david.thg.se/saker/turnup.diff ? just needs to create tmp log and run in /media/ram on startup too Jul 12 09:06:26 okay, i'm off Jul 12 09:10:49 DaKa2: Adding another switch to turnup might not be a bad idea. Make it default to tmpfs, and make it possible to override by providing a switch? That way somewhat knowledgeable users can get it set up automatically Jul 12 09:11:00 heh... Dr�mhus on the radio Jul 12 09:11:08 haha :) Jul 12 09:11:15 Guess which song :-P Jul 12 09:11:40 ja du... they made a lot of bad covers.. Jul 12 09:12:17 Here we never got rid of "Vill ha dig i m�rket hos meg" or whatever it's called Jul 12 09:12:24 haha Jul 12 09:13:02 "vill ha dig", originally by freestyle I think Jul 12 09:13:05 haha Jul 12 09:13:24 Anyway, what do you think about the switch? Jul 12 09:13:36 well, maybe Jul 12 09:13:55 I wasn't going to do anything at all, [cc]smart took on the task Jul 12 09:14:03 I just happened to be looking at turnup Jul 12 09:14:52 The patch you sent is what you wrote, isn't it? :) Jul 12 09:15:03 yes? Jul 12 09:15:13 but.. Jul 12 09:19:24 DaKa2: Could you file a slugbug, and attach that patch? Jul 12 09:19:40 do I have to? Jul 12 09:19:46 heh, no Jul 12 09:20:36 I dont even know if that is working Jul 12 09:21:10 I really should have a memstick to test with Jul 12 09:21:31 Yup.. Jul 12 09:21:33 same here Jul 12 09:21:41 Im running all mine from nfs Jul 12 09:22:24 I'm running one slug off of a disk, and one off of my iPod Jul 12 09:22:44 iPod, thats nice :-) Jul 12 09:24:13 yeah, well.. I can't exactly use it for its original purpose Jul 12 09:24:27 so I'd like to run off something else :P Jul 12 09:28:18 hm.. how do I make the wiki not do edit-links on things it doesnt know? Jul 12 09:30:19 [=BlahBlah=] Jul 12 09:30:23 ahh Jul 12 09:30:35 thanks Jul 12 09:41:57 so.. think this is done: http://www.nslu2-linux.org/wiki/HowTo/SetUpAScannerServerWithSANE , only for openslug, and may contain lots of errors and hard to understand bits Jul 12 09:42:38 who do I talk to to get the links approved? and where should the page be linked from, HowTo or OpenSlug? Jul 12 09:43:26 DaKa2, you can request here... there are a few of us that can do the link approval and the HowTo linkage Jul 12 09:44:07 good, well, http://www.nslu2-linux.org/wiki/HowTo/SetUpAScannerServerWithSANE and Im adding another right now Jul 12 09:44:28 the link approval is caught on a continuous basis, and the formal method for requesting HowTo linkage is via the mailing list Jul 12 09:45:07 by posting to the mailing list the addition of a new HowTo, the whole community has a chance to check it out Jul 12 09:45:31 ahh, should probably do that Jul 12 09:59:44 ahh, there, second page: http://www.nslu2-linux.org/wiki/HowTo/UseVLANsOnTheSlug Jul 12 10:00:04 better wait with mailing about it untill the packages actually are in the feed Jul 12 10:10:40 <[cc]smart> NAiL, DaKa2: just came back and read your posts. Now what is that with this patch. Is this one which already solves the /var issue ? Jul 12 10:11:12 [cc]smart: only a bit Jul 12 10:11:36 just needs to create tmp log and run in /media/ram on startup too Jul 12 10:12:03 and its untested Jul 12 10:12:48 <[cc]smart> ok Jul 12 10:13:00 <[cc]smart> then i just pulled and am merging, promptu got a conflict Jul 12 10:13:13 <[cc]smart> sbdy removed bridge-utils from openslug packages right ? Jul 12 10:13:21 <[cc]smart> that to stay removed ? Jul 12 10:13:30 <[cc]smart> same goes from cron. stay removed ? Jul 12 10:16:17 hmmm, two sec Jul 12 10:16:43 cron stays Jul 12 10:16:49 bridge-utils stays Jul 12 10:16:59 according to my right-now-pull Jul 12 10:17:25 <[cc]smart> i'm baffeld Jul 12 10:19:12 <[cc]smart> and completely lost with this ediff Jul 12 10:19:32 p/n, a/b Jul 12 10:19:33 :) Jul 12 10:19:41 <[cc]smart> ? Jul 12 10:19:50 p for prev, n for next Jul 12 10:20:05 a to choose version from left win, b to choose version from right win Jul 12 10:20:10 result in bottom Jul 12 10:20:19 <[cc]smart> and if i need to modify lines ? Jul 12 10:20:29 no idea Jul 12 10:20:29 <[cc]smart> or better for now, howto quit and start anew ? Jul 12 10:20:43 open a new terminal and kill emacs :P Jul 12 10:20:50 Dunno how to quit cleanly Jul 12 10:20:55 <[cc]smart> argh Jul 12 10:21:14 q, C-c, C-x possibly.. but that might save or something Jul 12 10:22:55 <[cc]smart> ah now i see, lines have been changed to single enries Jul 12 10:25:55 <[cc]smart> mgetty is now broken package or working package ? Jul 12 10:26:23 <[g2]> it was broken this morning Jul 12 10:26:45 <[g2]> durning the install stage trying to install to /usr.... Jul 12 10:27:16 <[cc]smart> and once i resolved all the conflicts how do i accept the result ? Jul 12 10:28:12 q Jul 12 10:28:17 then the mode changes Jul 12 10:28:39 then C-x C-c Jul 12 10:30:04 <[cc]smart> im a vi man. now im supposed to emacs ? dang Jul 12 10:38:42 :) Jul 12 10:39:09 if I had to do emacs I would kill someone, or it Jul 12 10:39:20 but Im not saying there is anything wrong with it Jul 12 10:39:20 03ccsmart 07org.openembedded.nslu2-linux * r19146380... 10/packages/meta/openslug-packages.bb: Added cyrus-imapd, mutt and postfix. Jul 12 10:39:23 03ccsmart 07org.openembedded.nslu2-linux * r9ae647b7... 10/packages/meta/openslug-packages.bb: Added mutt and postfix. Jul 12 10:43:58 Enough with the editor war Jul 12 10:44:00 hahaha. Jul 12 10:44:06 VI RULEZ!! Jul 12 10:44:09 :) Jul 12 10:44:53 Nah. ed... Jul 12 10:46:45 I'm sure we all agree on one thing though: notepad/wordpad sucks! Jul 12 10:47:26 There is one redeeming feature of wordpad and that is it loads a hell of a lot quicker than word Jul 12 10:48:16 What really gets me is some companies distribute documentation for Linux utilities in Word format... Jul 12 10:48:27 LSI Logic for one Jul 12 10:48:43 helllooo... I don't want a Word manual for the Linux admin tool for a RAID card... Jul 12 10:49:05 true. Word sucks even more! Jul 12 10:49:12 <[g2]> [cc]smart, did you pull mgetty ? Jul 12 10:50:18 <[cc]smart> mgetty in the version i sync is marked as not broken, since this is what i pulled Jul 12 10:50:39 <[g2]> have you built it ? Jul 12 10:50:46 <[cc]smart> so... should it be in broken section or not ? Jul 12 10:51:10 <[cc]smart> no, does error happen during building ? Jul 12 10:51:31 <[g2]> yes that's what happend to me, but I don't know if it got fixed Jul 12 10:51:42 <[cc]smart> will try compiling Jul 12 10:52:06 <[g2]> during the do_install it tries installing to /usr Jul 12 10:52:21 <[cc]smart> WRT to /var, these i think are the asprants for being linked into /media/ram: Jul 12 10:52:51 <[cc]smart> /var/cache /var/log /var/lock /var/tmp /var/lib/ipkg Jul 12 10:53:40 <[cc]smart> is this list correct ? Jul 12 10:53:52 <[g2]> dunno Jul 12 10:54:39 <[cc]smart> will assume it Jul 12 11:00:37 <[cc]smart> NOTE: package mgetty-1.1.30: completed Jul 12 11:04:32 <[g2]> you're running as a normal user right ? Jul 12 11:04:42 <[cc]smart> you mean bitbake ? Jul 12 11:05:01 <[g2]> yeah Jul 12 11:05:27 <[cc]smart> sure, read the docs: running bitbake as root bad things can and will happen Jul 12 11:05:46 <[g2]> just checking Jul 12 11:06:20 <[g2]> I hadn't seen any updates in the channel here Jul 12 11:06:22 <[cc]smart> i mean, this doesn't make a bug a good thing (TM) Jul 12 11:06:50 <[cc]smart> but removing such a problem doesn't make running as root a good thing (TM) Jul 12 11:32:20 hehe, NAiL, you added xinetd to openslug-packages, but not oe-symlinks... Jul 12 11:35:50 ogh Jul 12 11:36:37 03repvik * r75 10/releases/OpenSlug-2.0-beta/nslu2-linux/packages/xinetd: added xinetd Jul 12 11:37:43 <[cc]smart> so, lsowly this is gettin somewhere.... if only i could follow it :D Jul 12 11:48:17 <[cc]smart> you guys said you'd like to have a memstick to test that functionality... ? Jul 12 11:48:43 <[cc]smart> just forget that you've got an HD and tell turnup it's a memstick and sic it is a memstick Jul 12 11:49:07 yes.. but I dont have a hd... Jul 12 11:49:22 I run all mine off nfs Jul 12 12:05:26 <[cc]smart> it seems this stuff might even work :) Jul 12 12:07:00 [cc]smart: I use a memstick as disk, I didnt even think of it Jul 12 12:07:36 got one running with a 2GB memstick, and the other with a .. 80(?) GB disk Jul 12 12:07:39 <[cc]smart> this is the other way round... so you have a memstick but turnuooed it as disk ? Jul 12 12:07:49 yes Jul 12 12:07:54 what's the difference? Jul 12 12:08:01 <[cc]smart> you might consider changing that Jul 12 12:08:12 <[cc]smart> i'd expect your /var to be on the stick now Jul 12 12:08:15 <[cc]smart> check with mount Jul 12 12:08:43 <[cc]smart> this owuld cause frequent changes/updates/writes to the memstick Jul 12 12:08:57 <[cc]smart> which is disadvantageous to durability Jul 12 12:08:59 "df ." is broken :P Jul 12 12:09:32 I dont care about durability :) Jul 12 12:09:43 mount noatime Jul 12 12:11:29 <[cc]smart> well, if you care to have a replacement stick around, durability becomes a non issue Jul 12 12:11:37 heh, does laptop_mode work on openslug, I wonder :) Jul 12 12:12:20 noatime,commit=3600 Jul 12 12:12:21 hehe Jul 12 12:12:44 <[cc]smart> should Jul 12 12:12:59 <[cc]smart> if you go memstick noatime will be added par default Jul 12 12:14:06 ok Jul 12 12:14:27 but I wont bother re-turning it, easier to just edit /initrc/linuxrc Jul 12 12:19:59 [cc]smart: this usb stick I bought has 6 years warranty, hehe Jul 12 12:20:06 sorry, 5 Jul 12 12:20:37 <[cc]smart> how much of the time passed ? Jul 12 12:20:49 a week or so :) Jul 12 12:21:11 <[cc]smart> keep runnig it as hd, but remember, the warranty doesn't get extended, so run the last in the series as memstick Jul 12 12:23:19 <[cc]smart> so this should do it. but i have a feeling turnup gathers too much distro logic inside its body, which should in principle be in OE, not in turnup so to say. Jul 12 12:23:28 I run SD card as disk on the zaurus.. this 1GB has been running for close to two years now Jul 12 12:23:44 03ccsmart 07org.openembedded.nslu2-linux * r7e613c4f... 10/packages/openslug-init/ (openslug-init-0.10/turnup openslug-init_0.10.bb): Modified turnup to configure memstick for /var to run splitted in between tmpfs and disk directories. Jul 12 12:23:47 <[cc]smart> if this would gon on this way, one day we only have trunup and dispose OE so to say. Jul 12 12:23:52 and I'm already considering an upgrade.. Jul 12 12:24:26 <[cc]smart> NAiL: you copied the comment ? Jul 12 12:24:50 <[cc]smart> you mean you want more than a gig worth of data ? Jul 12 12:25:07 <[cc]smart> on the Z Jul 12 12:25:09 yes Jul 12 12:25:19 <[cc]smart> MP3 addicts everywhere :) Jul 12 12:25:26 uhm.. no mp3 :) Jul 12 12:25:42 <[cc]smart> OGG addicts everywhere Jul 12 12:25:47 hehe ;) Jul 12 12:26:02 well, I have CF cards with music and movies Jul 12 12:26:12 the SD I only use as system drive Jul 12 12:26:17 <[cc]smart> and an SD card with the index ? Jul 12 12:26:31 index? Jul 12 12:26:35 oh, nah Jul 12 12:26:44 I dont bother indexing what I have :) Jul 12 12:26:45 <[cc]smart> ya, like directory/playlist files :) Jul 12 12:26:56 I prefer to pick a directory and play it Jul 12 12:27:40 never bothered with playlists Jul 12 12:27:46 except for streaming sources Jul 12 12:28:23 <[cc]smart> so, time for sleep once more Jul 12 12:28:32 <[cc]smart> n8 all Jul 12 12:28:50 night :) Jul 12 13:00:40 eno, ping? Jul 12 13:02:25 03jp30 * 10unslung/Makefile: demote xterm Jul 12 13:42:06 jp30, pong. Jul 12 13:43:03 hi, eno - i had trouble rebuilding xterm. the problem is that the url you are downloading is always the latest xterm, and the optware makefiles can't tell whether downloads/xterm.tar.gz is up to date Jul 12 13:43:21 (two problems there really) Jul 12 13:43:54 ok, let me check Jul 12 13:51:37 rebuilding xt xaw right now ... Jul 12 13:52:55 03aabaker * 10unslung/Makefile: Mark Bitlbee and the libraries it needs as ready for testing Jul 12 13:58:48 jeanfabrice, i'm just writing a perl script to do some automated sanity checking on optware packages. it has noticed that php-thttpd builds static binaries. these won't work on the slug, as we have a broken static glibc in our toolchain Jul 12 13:59:05 php-thttpd itself works because it is a library, not a binary Jul 12 13:59:16 doh... scratch that Jul 12 13:59:45 the thttpd binary is built correctly for some reason, the other binaries aren't Jul 12 14:00:22 03bzhou * 10unslung/make/xterm.mk: using version specific URL to download source, stripped binary, changed maintainer Jul 12 14:01:55 jp30, you said there were two problems? Jul 12 14:03:15 you have fixed them both, i think. one was that you were downloading an unpredictable version of xterm, the other was that the version wasn't encoded into the name of the tarball Jul 12 14:04:16 looks like it is working again... Jul 12 14:04:32 great Jul 12 14:55:54 03jp30 * 10unslung/ (unslung-check-package.pl Makefile): package checking magic Jul 12 15:11:22 ello ello Jul 12 15:12:52 so, what does it mean when /dev/sda2 (/share/hdd/conf) gets mounted ro, and when I fsck it it finds and fixes the error, and then when I reboot it mounts /dev/sda2 as readonly again? I just posted to the groups mailing list but then figured #nslu2-linux would be best anyway Jul 12 15:12:59 unslung 4.20 btw Jul 12 15:14:10 hello? quiet here tonight... Jul 12 15:14:25 ~nslu2-general Jul 12 15:14:26 #nslu2-general is the IRC channel specifically set up for end-user questions about the NSLU2 device. Questions that are not related to the nslu2-linux development projects should not be asked in #nslu2-linux, but should be asked in #nslu2-general instead. Jul 12 15:14:44 sigh Jul 12 15:14:54 err.. how is unslung not nslu2-linux related? Jul 12 15:14:55 FAQ34 might be appropriate. Jul 12 15:16:07 *development projects* Jul 12 15:16:23 general is more for end user help Jul 12 15:16:52 -linux is for "I'm packaging up a new ipkg foo, and I need some help" Jul 12 15:16:55 FAQ34 refers to the data partition. I'm talking about the conf parition. But ok, I guess I'll try nslu2-general. (Guess I thought nslu2-general was for stock firmware; guess I was wrong) Jul 12 15:17:05 same thing. Jul 12 15:17:09 they are both ext3 partitions. Jul 12 15:17:14 shrug Jul 12 15:18:22 Well, no... but.. I guess you'll find out if you're also reading nslu2-general Jul 12 15:18:56 what do you mean no. Jul 12 15:19:03 they ARE both EXT3 partitions. Jul 12 15:50:38 Anyone know whether [cc]smart have a logger? Jul 12 16:00:26 AdamBaker, ping? Jul 12 16:00:37 [g2]: I built mgetty as non-root and it didn't fail Jul 12 16:00:53 mgetty is *not* broken and *must* stay in the feed Jul 12 16:01:10 03jp30 * 10unslung/ (Makefile unslung-check-package.pl): unslung-check-package: check packages during make testing; ignore crosstool, ignore missing rpaths on packages that don't require them Jul 12 16:02:07 AdamBaker, bitlbee doesn't configure correctly. it looks like it assumes a host gnutls Jul 12 16:04:14