irclog2html for #nslu2-linux on 20050504

00:01.17krodwhat does busybox use, to load all the services at dosn't seem to use inetd
00:04.50jp30-workkrod, have you looked at the wiki?
00:06.19krodsorta :)
00:06.25krodi will read it fully now..
00:06.31krodthanks for your help
00:11.08jp30-workkrod, do feel free to ask questions, but bear in mind rules #1 and #2 - they are explained on the wiki front page
00:19.25jp30-workdo we have a way of asking jbot to politely remind visitors of rules 1 and 2?
00:22.40jbotrumour has it, nslu2-rules is 1) Those who ask the questions, update the wiki, 2) Those who complain about the doco, update the wiki, 3) Friends don't let friends flash custom firmware without confirmed RedBoot access, 4) Friends don't let friends flash custom bootloaders without confirmed JTAG access.
00:33.45*** join/#nslu2-linux rwhitby-web (
01:07.28*** join/#nslu2-linux krod (~krod@krod.user)
01:10.41krodthere should be a search on the wiki
01:11.02ka6sox-officethere is
01:11.32krodyeah, i say it just as i pressed enter...
01:14.50*** join/#nslu2-linux krod_ (
01:39.27krod#  ipkg install openssh
01:39.28krodipkg: Cannot create directory `/usr/lib/ipkg': File exists
01:39.52krodany ideas?
01:50.56krod# ipkg update
01:50.56krodipkg_download: ERROR: Command failed with return value 1: `wget --passive-ftp    -q -P /tmp/ipkg-3J0vCg http://ipkg.nslu2-linuxAn error ocurred, return value: 1.
02:01.45*** join/#nslu2-linux dyoung-web (
02:02.01dyoung-webso is anyone useing the OptWare DHCP package?
02:16.47*** join/#nslu2-linux ka6sox (
03:24.13*** join/#nslu2-linux dyoung-web (
03:37.32*** join/#nslu2-linux ByronT-Away (~byron-por@ByronT.nslu2-linux)
03:49.26*** join/#nslu2-linux kergoth (
04:23.39*** join/#nslu2-linux Ksilebo^ (
04:23.39*** join/#nslu2-linux kergoth (
04:23.39*** join/#nslu2-linux ka6sox (~ka6sox@ka6sox.nslu2-linux)
04:23.39*** join/#nslu2-linux rwhitby-web (www-data@rwhitby.nslu2-linux) [NETSPLIT VICTIM]
04:23.39*** join/#nslu2-linux dyoung-logger (~dyoung-lo@dyoung.nslu2-linux)
04:23.40*** join/#nslu2-linux glc (~glc@glc.nslu2-linux) [NETSPLIT VICTIM]
04:23.40*** join/#nslu2-linux Ksilebo ( [NETSPLIT VICTIM]
04:23.40*** join/#nslu2-linux rwhitby-away (~rwhitby@rwhitby.nslu2-linux) [NETSPLIT VICTIM]
04:23.40*** join/#nslu2-linux CIA-8 ( [NETSPLIT VICTIM]
04:23.40*** join/#nslu2-linux schula ( [NETSPLIT VICTIM]
04:23.40*** join/#nslu2-linux wbx ( [NETSPLIT VICTIM]
04:23.40*** join/#nslu2-linux zoobab ( [NETSPLIT VICTIM]
04:23.40*** join/#nslu2-linux jf-work (~none@jeanfabrice.nslu2-linux) [NETSPLIT VICTIM]
04:23.40*** join/#nslu2-linux eno-away ( [NETSPLIT VICTIM]
04:23.40*** join/#nslu2-linux Tiersten (~tman@Tiersten.nslu2-linux)
04:23.40*** join/#nslu2-linux senneth (senneth@senneth.staff.irssi) [NETSPLIT VICTIM]
04:23.40*** join/#nslu2-linux d|dge (~didge@2001:960:624:1:0:0:0:1) [NETSPLIT VICTIM]
04:23.41*** join/#nslu2-linux nslu2-log (dyoung@dyoung.nslu2-linux)
04:23.41*** join/#nslu2-linux sp4ce ( [NETSPLIT VICTIM]
04:23.41*** join/#nslu2-linux widrone ( [NETSPLIT VICTIM]
04:23.41*** join/#nslu2-linux jp30-work (~jp30@jp30.nslu2-linux) [NETSPLIT VICTIM]
05:57.17jf-asleepjacques: tinylogin adduser is creating group. Maybe we should patch it not to, as it was patched not co call for passwd ?
05:58.42jf-awaymorning all, sorry, I'm late. going to the office
05:59.20jf-awayjacques: this is just a one line comment in the code. Just have to take care for future version if ever
06:02.08*** join/#nslu2-linux ruDelph (~delph@
06:29.48jbotextra, extra, read all about it, nslu2-camps is 1: want full linksys compatibility (including future releases), can load image through web and telnet in to run ipkg, have no clue about redboot, 2: can telnet into redboot and use linux, 3: power user, but no jtag and no custom redboots, 4: bleeding edge - custom redboots, no need or desire for linksys compatibility
07:25.21*** part/#nslu2-linux rwhitby-web (www-data@rwhitby.nslu2-linux)
07:29.46*** join/#nslu2-linux Christopher (
07:47.39*** join/#nslu2-linux perlguru (
07:47.54perlguruipkg upgrade gives me errors:
07:48.09perlguruupdate-modules and depmod couldnt be found
07:48.40perlgurualso is missing for apache ( exists, typo?)
07:57.55*** join/#nslu2-linux jaro (~ja4aro@
09:43.37*** join/#nslu2-linux dyoung (~dyoung@dyoung.nslu2-linux)
09:48.08perlguruyes, revision 1.500 of the Makefile is upon us!
09:48.39perlgurubtw, CIA is missing in action
09:49.11perlgurui added perl-cgi-application and perl-html-template to native-testing
09:49.21rwhitbydyoung: should have pingu ssh access available soon
10:00.16jacquesi might need to get me a couple of those wl-hdd2.5's
10:00.21jacquesthey look cool
10:00.53jacquesonly about $3 more than a slug too
10:02.22perlguruthat is without a hdd
10:02.31jacqueswell so its a slug
10:02.35jacquesso is a slug
10:04.01jacquesyou have a handy link with the hardware specs? like how much flash and RAM ?
10:04.09jacquesand what proc exactly
10:06.18jacqueslol I love this feature (from the ASUS site): "Real Time Clock: You'll never disorder time even after being unplugged."
10:06.27jacquesthanks jf-work
10:07.31jacquesok so broadcom cpu, 16MB RAM and 4MB flash
10:07.46jf-worksame as wl500
10:08.03jf-workmy optware run on it
10:08.57jbotACTION lights off 3 tons of fireworks
10:10.26jf-workdo we said 'optware packages' or just 'optware' ?
10:10.49dyoungjf-work, you should have that slug in-hand soon.
10:11.06dyoungcan you let me know when you get it?
10:11.26jf-workdyoung: I look in my postal box 3 times a day for 2 day now.
10:11.46dyoungI dont remember when I sent it, but they told me a week.
10:12.02dyoungI think it was mid-last-week ?
10:12.08jf-workdyoung: sure. I will have to look for a correct 220v->5v adapter. Can you give me the polarity of the plug ?
10:12.13dyoungay it was last wednesday.
10:12.20dyoungCenter Positive.
10:12.40dyoungI dunno the precise size t hough...
10:12.52jf-workyep. can you mesure it ? ;)
10:19.22dyoungbest I can guess, its a 2.5mm "short" connector.
10:19.38dyoungthe OD is 5.4mm
10:20.30dyoung"short" = 9.5mm
10:20.38dyoungin length.
10:21.05dyoungjacques, please you are using the english to proper descibing the function of the unit!
10:22.15jf-workdyoung: ok, so 2,5mm center hole (positive) and 5,4mm exterior (negative), that's it ?
10:22.38jacquesdyoung lol
10:22.38dyoungYes; thats about it.
10:23.20dyoungI wonder if someone has made a english -> engrish converter yet.
10:23.50jacquesI'd like to talk like that all the time
10:24.13dyoungI took some pretty good pictures of "engrish" signs.
10:24.20dyoungComplaining phone!
10:24.43dyoung"Please preserve the greenery by no stepping!"
10:25.28dyoungjbot, wiley is the petname for the Asus WL-500 family of products, including the WL-HDD
10:25.29jbotokay, dyoung
10:25.35jbotwiley is probably the petname for the Asus WL-500 family of products, including the WL-HDD
10:26.24dyoung"Walk this way to the Pagoda" in chinese they translated as, "Step on Pagoda Here!"
10:27.01jacqueslol it works
10:27.19jacquesI asked it to translate "Hi there I am your friend" and it gave me "Today as for me it is your friend there."
10:27.52dyoungjacques, youre going to be spending all night translating stuff to "engrish" now.
10:28.18jacqueswell if I wasn't about to pass out from sleepiness I would :-)
10:28.30jacquesI'm glad I installed the japanese fonts now
10:28.34dyoungI can speaking the engrish on the whim!
10:28.49jacquesthat look like anything?
10:28.56dyoungyeah, it looked like:
10:29.25rwhitbyperlguru: wl-hdd is very slow on disk - about half the slug
10:29.31jacquesah well
10:29.44jacquesrwhitby, ouch!
10:29.52jacquesthey claim to have ata100 ide interface
10:30.20jacquessounds more like ata4
10:31.54perlguruspeed is no issue. just want it as a webserver/mediaserver
10:32.39dyoungUsing the WL-HDD reveals only half the required speed of the fresh animal
10:34.23dyoungokay maybe thats not quite right.  I need to work on the engrish translator.
10:35.15dyoungI think a slug is a "namakuji"; but nama can mean fresh.  I'm still thinking of a better misinterpretation of "kuji"
10:36.37dyoungoops, i even got that wrong, its a namekuji.
10:37.21dyoungthe licking ticket.
10:38.05dyoungOkay, who started me on engrish.....
10:38.39jacquesi think it was me
10:38.49jacques<jacques> lol I love this feature (from the ASUS site): "Real Time Clock: You'll never disorder time even after being unplugged."
10:39.12dyoungoh yeah.
10:39.17dyoungits all your fault. ;-)
10:42.56jacquesglad to be of service
11:19.19*** part/#nslu2-linux dyoung (~dyoung@dyoung.nslu2-linux)
11:22.16*** join/#nslu2-linux dyoung (~dyoung@dyoung.nslu2-linux)
11:44.30rwhitbydyoung: tried the latest unslung yet?
11:46.22rwhitbyI'm just about to flash it now ... (4.17-alpha)
11:51.20rwhitbybooted fine.
11:51.44rwhitbynow to touch the flag file and see what happens
11:53.25rwhitbyhmm - endless loop
11:54.56rwhitbyright - glc mentioned that.
11:55.11rwhitbyneed to not run linuxrc on the target rootfs
11:57.47dyoungNope, I'm still in the dark ages of 4.15
11:59.04rwhitbyand fixing the bug in the /linuxrc is as easy as rebooting without the disk, and editing /linuxrc in the jffs2
12:01.01rwhitbyand we're up
12:17.51rwhitbyuploading 4.18-alpha.bin now ..... (will take 15 minutes)
12:18.01rwhitbyCIA is still MIA.
12:25.37*** join/#nslu2-linux caplink811 (
12:25.42glcOr deleting linuxrc on sda1.
12:26.32glcMaybe not executing foreign linuxrc is safer though.
12:26.51rwhitbyyeah, I've deleted those lines.
12:27.20glcWell, off to work for me.
12:27.32rwhitbysince any foreign rootfs is editable, then people can put a shell script in place of /bin/init and call /bin/init from inside it
12:27.52rwhitbybut most people willl never need to know about that
12:28.45glcI did that when tryng to debug before booting into jffs2 was inplace.
12:30.14rwhitbywell glc, congratulations on simplifying the whole Unslung boot sequence!
12:31.53dyoungdid someone propogate the chage to openslug yet?
12:31.56dyoung~praise glc
12:32.01jbotAll hail glc!
12:37.20rwhitbynow, what should I do with the 1 block where switchbox used to be?
12:38.45dyoungis the 2nd block recovered already?
12:39.11rwhitbynot yet, looking at that now.  Should be as easy as just giving slugimage a smaller ramdisk
12:39.16dyoung"Rescue Payload" ?
12:40.45rwhitbywhat does that consist of?
12:41.29dyoungI'm not sure yet.
12:41.57dyoungfacility to cp/boot to ramdisk and reflash the jffs2?
12:42.05dyoungin case you are "inslung" ?
12:42.38rwhitbywe have that already in maintenance mode
12:43.47dyoungokay, so we are keeping that mode?
12:44.40dyoungI dunno then, maybe we should stuff some jpegs in there.
12:45.08rwhitbyanyway, we just need to put the switchbox.ext.gz on and it will be included ...
12:45.13jboti heard nefarious is to be infamous by way of nastyness
12:45.52*** join/#nslu2-linux jp30 (~josh@jp30.nslu2-linux)
12:46.00dyoungSpeak of the devil.....
12:46.11jp30hi, dyoung
12:46.20jp30it's very early in the morning here
12:46.34dyoungI guarantee its even earlier in the morning here. ;-)
12:47.04jp30actually i was hoping to find eno and bug him about samba. looks like he is away though
12:48.14rwhitbywe don't need the payload either now, do we?
12:50.50dyoungnot really, we can cram moer jpegs in there. ;-)
12:51.24dyoungit o nly exists for rescue mode.
12:52.13rwhitbywe don't have recovery mode any more
12:53.53dyoungperfect place for nudi-slugs.
12:54.55jp30is anyone here able to reproduce the samba errors reported on the list? i'd like to figure out whether they are due to the update to samba 3.0.14, or to the move to cross-compilation
12:57.37dyoung , but I havnt tried for myself.
12:59.26dyoung-zzzzwait, isnt uid -1 supposed to be nobody?
12:59.46dyoung-zzzzthat error looks like a bug.
13:01.04jp30trouble is that the two changes happenned at the same time. so the fact that a user was able to back off to 3.0.11 doesn't show which caused the problem
13:01.47dyoung-zzzz"dont break more than one thing at a time."
13:09.39*** join/#nslu2-linux perlguru (
13:32.25rwhitbyok, I've updated slugimage, but will have to wait until anon cvs catches up before I push the BB changes
13:36.48perlgurudoes the new init stuff reduce the boottime?
13:38.09rwhitbywell, the thing is that now you have complete control over the boot sequence, by modifying the linuxrc in jffs2.
13:38.25rwhitbyso you can reduce it to whatever still works
13:46.26rwhitbyok, linuxrc has also been added to openslug-init as well ....
13:46.37rwhitbyand switchbox has been removed from openslug.
13:46.55rwhitbyas soon as slugimage anon cvs updates, I'll push the changes.
13:47.58jp30perlguru, i've just promoted perl-cgi-application and perl-html-template.
13:54.09*** join/#nslu2-linux jaro (~ja4aro@
13:59.50jarojeanfabrice: Yesterday I asked you about compiling thttpd-php with uclibc - you said that you did compilation using uclibc based toolchain. However your package required libc to run.
14:00.05*** join/#nslu2-linux ziL (
14:00.16ziLI just ordered a nslu2 :)
14:02.28jf-workjaro : compiled with unslung_target=wl500g, it seems to me that php-thttpd doesn't depend on libc
14:05.22jarojf-work: It did install correctly using ipkg (bobo.rouselin repository), however it failed during startup. I will try to simulate that situation again. In a moment, I will come back with more detailed report.
14:06.30rwhitbyjaro: there is no thttpd-php package
14:06.36rwhitbydo you mean php-thttpd ?
14:07.34rwhitbythere once was a thttpdphp package, but that has been superceded by php-thttpd
14:08.39rwhitbyziL: welcome to the club.  what are you going to install on it?
14:12.22jarorwhitby: Package: thttpd-php ... Version: 2.25b-5.03
14:13.38jarorwhitby: maybe I've dowloaded just by the time of "superceding" :)
14:13.38rwhitbyInstalling php-thttpd (2.25b-5.0.3-6) to root...
14:14.04rwhitbyjaro: where is that package downloaded from?  I don't believe it's an official Unslung package.
14:15.55jp30morning, ka6sox
14:16.47ziLrwhitby: I'm going to use it with ccxstream and play around with it in general
14:17.13ziLrwhitby: Ordered a  pixelbild
14:17.14ziLMaxtor Onetouch 250GB 7200RPM 8MB USB2.0
14:17.24ziLSorry about the paste :)
14:18.14rwhitbyNAiL: cutting a hole in the side to see it?
14:18.40NAiLrwhitby: I have to cut a hole in the side to make it fit at all
14:18.43rwhitbyyou could mount it on the outside, and just run the wires through a hole in the case
14:18.51ziLI got myself a nice 17" tft too, can't stand my 19" Dell monitor any longer :)
14:18.52NAiLNah, won't look as good ;)
14:19.10ziLNAiL: Looks are everything
14:20.46NAiLBut unless I can fit a usb2 hub inside the case, together with my 512mb flash-stick and a usb->lpt-converter and the display, I'm running out of ports :(
14:21.17NAiLI guess I'll just have to buy another slug ;)
14:21.40ka6sox~seen [g2]
14:21.42jbot[g2] <~g2@g2.nslu2-linux> was last seen on IRC in channel #nslu2-linux, 1d 51m 33s ago, saying: 'rwhitby, hey!'.
14:22.14rwhitbyNAiL: that's two ports - how many more do you need ;-)
14:22.58NAiLrwhitby: Loads! ;)
14:23.22ka6soxrwhitby: show him Slug Central
14:24.48rwhitbyNAiL: <- put one of them inside
14:25.09rwhitbyI've got one of them on the back of my wl500gx
14:26.24perlgurujp30: thanx
14:26.24NAiLrwhitby: Looks a tad fat to fit inside though?
14:26.54rwhitbyjaro: I just installed php-thttpd on my wl500gx and it runs without error
14:27.49rwhitbyNAiL: not after you unsolder all four connectors and connect the wires directly.
14:28.11NAiLah.. yeah.
14:28.37rwhitbyit's only usb 1.1 if that matters to you
14:28.48NAiLNow I only need to find someone who distributes that hub in Norway...
14:29.19NAiLrwhitby: nah, not that much. I can put my stick on that hub, and use the usb2-port it frees up instead. Don't care much about speed with the stick
14:29.28NAiLnor with the lpt-port
14:30.01rwhitbythere you go - now you can have two internal ports spare and one external port spare.
14:30.59NAiLNice, Dell sells them
14:31.39jf-workjaro : thttpdphp package from bobo(rousselin is obsolete (I should have delete all non unslung package form this point :()
14:32.31jf-workrwhitby : if use to be a package from me when I didn't know unslung
14:33.02rwhitbyjf-work: ah, now it makes sense.
14:35.36rwhitbynight all
14:35.43NAiLrwhitby: Thanks for the tip.
14:35.48NAiLand 'nite
14:36.16rwhitby-asleepJust make sure you put up some more pictures :-)
14:36.23NAiLBut of course ;)
14:39.05rwhitby-asleepok, I'm going to push these changes, but they won't work until slugimage anon cvs catches up.
14:39.22rwhitby-asleep(cause I won't have time to push them when I wake up tomorrow)
14:47.27*** join/#nslu2-linux jstueve (
14:47.43jstueveanyone have any ideas how to recover a root password?
14:50.30jp30jstueve - attach your slug's hard drive to a PC and edit the passwd file in the conf partition
14:50.40jarojf-work: thanks. So that's probably the problem :)
14:50.52jarojf-work - I will perform upgrade :)
14:50.54jp30...then you can change the root password to whatever you like
14:52.08jstuevebest to just blank it out, reboot and change it, copy back to /share/*/conf/passwd
14:52.33jstuevecheck... I just read the first part in the readme...
14:53.05jf-workjaro: you should remove first then install. upgrade will not work as packages don't have the same name
14:56.07jarojf-work: Yes. I wrote an inexact thing. I don't have that package on my wlhdd anymore, because I'm experimenting all the time and the disk was formatted many times :-)))
15:00.20ka6sox-awayjstueve, could you please write up a wiki page on the lost password recovery?
15:00.57jbotmethinks nslu2-rules is 1) Those who ask the questions, update the wiki, 2) Those who complain about the doco, update the wiki, 3) Friends don't let friends flash custom firmware without confirmed RedBoot access, 4) Friends don't let friends flash custom bootloaders without confirmed JTAG access.
15:01.20jp30hah, an opportunity to use ~nslu2-rules!
15:03.26jp30for those who are interested in such things, i believe i have PEAR working
15:08.46jstueveka6sox yeah... if/when I get it back.. :D
15:10.25jstueveI'm using it remote now... but the root passwd got changed to something unknonw... so can't do some stuf...
15:29.59*** join/#nslu2-linux ByronT-Away (byrontodd@ByronT.nslu2-linux)
15:43.07ziLDoes the Maxtor drives make a lot of noise?
15:43.33ka6sox-awaysome...and some not.
15:43.39*** join/#nslu2-linux jstueve (
15:44.35ziLka6sox-away: Do you have a good suggestion for a near silent drive?
15:44.53jp30how about a flash key?
15:51.42ka6sox-awayyeah....I was going to say that.
15:52.59jp30if silent counts as "near silent" :)
15:54.03ka6sox-awayoff to the Mountian.
15:54.06ka6sox-awaycya laters.
15:55.07jp30ttfn, ka6sox
15:58.34*** join/#nslu2-linux ByronT-Away (byrontodd@ByronT.nslu2-linux)
16:06.38*** join/#nslu2-linux sparkling (
16:07.07*** part/#nslu2-linux sparkling (
17:02.03*** join/#nslu2-linux pTweety (
17:15.58*** join/#nslu2-linux [cc]smart (~smart@
18:17.55*** join/#nslu2-linux jstueve (
19:06.58*** join/#nslu2-linux jeanfabrice (~jeanfabri@jeanfabrice.nslu2-linux)
19:07.10jp30-workmorning, jeanfabrice
19:07.35jeanfabricegood morning
19:08.25jp30-workwould you care to join me in testing UW imap and PEAR?
19:08.58jp30-work...i was just about to put imap into WL500G_PACKAGES_READY_FOR_TESTING
19:09.39jeanfabriceI would be glad if I can give a hand, np.
19:10.26jp30-workmain thing i'd like checked is that "make imap-ipk php-ipk" works on wiley with current cvs targetting wiley, i should say
19:11.28jeanfabricedyoung-zzzz: no slug today. Tomorrow is off if France.
19:11.47*** join/#nslu2-linux Jacmet (
19:13.02*** join/#nslu2-linux macsat (
19:31.16*** join/#nslu2-linux ByronT_ (byrontodd@ByronT.nslu2-linux)
20:01.55*** join/#nslu2-linux jeanfabrice (
20:10.06VoodooZ_Workjbot: seen odoc
20:10.09jbotodoc <> was last seen on IRC in channel #nslu2-linux, 13d 11h 41m 14s ago, saying: 'hehe'.
20:11.02jeanfabricejp30-work: I did not forget you, still building php dependencies (got network troubles)
20:21.09*** join/#nslu2-linux [g2] (~g2@g2.nslu2-linux)
20:28.07dyoung-zzzzI take it Odoc hasnt gotten back with you eh.
20:28.28VoodooZ_WorkI just did a search on google for his alias but didn't find anything.
20:28.30dyoung-zzzzI waslooking for you the other day because I got the spca5xx working.
20:28.42VoodooZ_Workjbot says he has not been in this channel for 13days.
20:28.46dyoung-zzzzbut I think it is in fact *not* a USB2 camera.
20:29.00VoodooZ_Workyeah, I've narrowed it down to 3 cameras.
20:29.10dyoung-zzzzI was gonna offer to run some basic tests; but its pretty slow.
20:29.43*** join/#nslu2-linux jstueve (
20:29.44VoodooZ_WorkThe Trust spacecam 380 appears to be supported (more or less) with the OV5XX drivers.
20:30.22VoodooZ_Workthe iBot2 is another one and then Aplux Mu2-35 but they are all pretty hard to find in Canada.
20:30.48VoodooZ_Workebay didn't turn out much either.
20:31.46VoodooZ_WorkI fired off an email to the OV511 dude to see what he suggests.
20:32.37dyoung-zzzzAh, okay the spacecam is stupposed to work with the spca5xx.
20:33.19VoodooZ_Workreally? It's supported under the Omnivision OV drivers. Unless it's under both?
20:35.13VoodooZ_WorkI just checked the SPCA5xx project page and there's no mention of support of the 380 model. Just the Trust familycam 300
20:39.38dyoung-zzzzAha.  yes, those model numbers are all so confusing.
20:44.26VoodooZ_Workanyways, got to leave work now. seeya later....
21:19.16rwhitby-asleepok, slugimage changes have propagated, so builds on Unslung and OpenSlug from now on do not include switchbox, and therefore recover one extra block for the jffs2.  The "ramdisk" block is left empty (except for the length).
21:19.32rwhitby-asleepThis has not been tested yet, so those with serial should test it first.
21:20.19rwhitby-asleepjacques: can you look at my slugimage source change, and see if you agree with the logic?  It's a three line change.
21:21.21jacquesrwhitby-asleep, ok
21:21.24jacqueswill look
21:21.30rwhitby-asleeponce someone with serial has tested both Unslung and OpenSlug without switchbox (i.e. the image that will build now), then we'll put out the word for others to test it.
21:21.59rwhitby-asleepback later
21:22.23jacquesrwhitby-asleep, did you submit change to nslu-changesets ? I don't see it
21:22.46rwhitby-awayI did.  Hmm - maybe my new CoLinux mail is broken.
21:26.01ka6sox-away[g2]-away, ping
21:26.45jacquesI can look at the changes in the cvs browser
21:26.59jacquesduh since they wouldn't be in bk anyway
21:27.06ByronT[g2]-away: PING!
21:51.21dyoung-zzzzka6sox, um....
21:52.02dyoung-zzzzka6sox-away, what happens if I accidentally use a 100k instead of a 10k for a pullup?
21:55.17jeanfabricejp30-work: ping
21:56.35jeanfabricejp30-work: php builds fine with PEAR. ipk seems to install pear under /tmp. Sounds strange :/
22:04.13jp30-workhey, jeanfabrice, thanks.
22:05.04jp30-workthe php-pear ipk is supposed to contain only files in /tmp. it is magic
22:05.31jp30-workthe next thing i wanted to ask was whether installing php-pear works on a wiley
22:07.42jp30-workgood way to test it (pear) is to see if "pear list" and "pear list-all" produce sane output
22:11.05*** join/#nslu2-linux jaro (~ja4aro@
22:11.05jeanfabricejp30-work: do you know /tmp is in flahsfs on wiley ?
22:11.21jp30-workulch, that's nasty. it's not a ramdisk?
22:11.56jeanfabriceramfs on /tmp type ramfs (rw)
22:12.37jp30-workyeah, the trouble with pear is that it's complicated to install, and the installation procedure is written in php
22:13.21jeanfabriceany regexp wizard around ?
22:14.25jp30-workwhat level wizard do you need? i am about level 6 :)
22:14.43jeanfabriceI'm near 0 so...
22:15.07jp30-workwell, ask away then.
22:15.09jeanfabricewell did you work on ? I need to trap the real BUSYBOX_VERSION
22:15.41jeanfabricethere are two with a condition in
22:19.22jp30-workit looks like you can "make busybox-ipk USE_BUSYBOX_SNAPSHOT=20050101"
22:19.40jp30-workin fact the busybox version that's in the feeds is 1.00
22:20.56jp30-worklet's look at who wrote those lines...
22:21.35jp30-workit was rwhitby who was responsible for the conditional
22:22.09jp30-worki suspect it can just be removed but it might be worth asking him why it was there to begin with
22:22.14jeanfabriceyep. But I'm rebuilding an package busybox based as I got trouble with tinylogin. I would like to have based on busybox like you did with php-pache based on
22:23.52jp30-workwell, if there's no real reason to have the conditional there, then the complicated USE_BUSYBOX_SNAPSHOT stuff could be removed, and the sed trick that we used with php would work fine
22:24.18jeanfabriceok. will wait for rod approval
22:24.39jp30-workin the meantime, you could make a sed script that extracts the BUSYBOX_VERSION= line that has a digit after the =
22:25.44jeanfabriceI know how to subsitute, but lack some skill on how to extracts
22:26.02jeanfabricereading regexp manpages now...
22:27.13jeanfabriceok, found ;)
22:28.19jeanfabricesed -n -e 's/^BUSYBOX_VERSION *=\([0-9]\)/\1/p' make/
22:28.41jp30-workgood, that looks right.
22:28.52jeanfabricethx *blush*
22:29.41jp30-work500 XP, and you learn the spell "backrefer to subexpression"
22:33.41dyoung-zzzzI had to inspect that statement several times before catching on .
22:38.25jeanfabricejp30-work : what was the other optware you ask me to build ?
22:39.25jp30-workimap; though if you have build php sucessfully, imap will have been built and staged
22:39.55jp30-workcan you check that "ipkg install php-pear" actually has the result of installing pear, though?
22:40.19jp30-work...a lot of work is done in postinst with php-pear
22:40.22jeanfabriceyep it's done. imap-ipk is ok
22:43.53jeanfabriceInstalling php-pear (5.0.3-9) to root...
22:43.53jeanfabriceConfiguring php-pear
22:43.53jeanfabrice[PEAR] Archive_Tar    - installed: 1.1
22:43.53jeanfabrice[PEAR] Console_Getopt - installed: 1.2
22:43.53jeanfabricesh: /usr/bin/cpp: not found
22:43.53jeanfabrice[PEAR] PEAR           - installed: 1.3.3
22:43.55jeanfabriceWrote PEAR system config file at: /opt/etc/pear.conf
22:43.57jeanfabriceYou may want to add: /opt/share/pear to your php.ini include_path
22:43.59jeanfabrice[PEAR] XML_RPC        - installed: 1.1.0
22:44.01jeanfabriceSuccessfully terminated.
22:46.20jp30-workcan you try running "pear list" (expecting to see a short list of packages), and "pear list-all" (expecting to see a long list)?
22:47.14jp30-workhow so?
22:47.18jeanfabrice[jfb@hdd root]$ pear list
22:47.18jeanfabricebefore write 0 : 1115246791
22:47.18jeanfabricebefore read 0 : 1115246791
22:47.18jeanfabricebefore write 0 : 1115246791
22:47.18jeanfabricebefore read 0 : 1115246791
22:47.18jeanfabriceexec: 28: -C: not found
22:47.29jp30-workthat's not so good
22:48.16jeanfabricethat is tipycally waht happen with some package that compile fine but crahes at runtime (sane for example or mysql when I omit libstdc++)
22:49.13jp30-work/opt/bin/pear is a shell script, you might find it helpful to look inside it
22:50.17jp30-workthat "-C not found business" looks like the problem is that /opt/bin/pear fails to find the php binary
22:56.17jeanfabrice=> 'test' (/usr/bin/test) is broken. It is part of wileyware ;)
22:56.49jeanfabriceit is 'test' which produce the errorline, so you job is good !
22:57.03jeanfabricehave to report this to oleg I think
22:57.54*** join/#nslu2-linux rwhitby-web (
22:59.10rwhitby-webNOTE NOTE NOTE: The current OpenSlug image will probably not boot, due to /mnt/tmpmnt being missing from the rootfs, and the new linuxrc depending on that.
22:59.33NAiL*very* nice to know.
23:00.11rwhitby-webalthough it will probably boot fine to jffs2 - the /mnt/tmpmnt is only used when trying to load rootfs from another location (like an external disk)
23:00.37rwhitby-webbefore doing a turnup procedure, you would need to create /mnt/tmpmnt in the jffs2
23:01.17rwhitby-webthe linuxrc is fairly fail-safe
23:03.13rwhitby-web[g2]'s name for the OpenSlug version of unsling
23:08.00ka6sox-officedyoung-zzzz, ping1
23:08.11ka6sox-office(as opposed to ping0)
23:08.26rwhitby-webor pingU
23:10.19ka6sox-officeI think if we can get the necessary zips and tgz's today we can be ready tonight :)
23:11.06ka6sox-officeBerlios and SF both have a problem with what we want to do.
23:11.52rwhitby-webdid you make it clear that source would be there too?
23:11.57ka6sox-officethe TOS are written pretty wierd.
23:12.18ka6sox-officeits the NPE/IXP stuff that is the problem.
23:13.05ka6sox-officebecause of the license the way it reads.
23:13.22jeanfabricehave bumped adduser optware VERSION and update it to use busybox applets instead of tinylogin's ones. tinylogin applets were broken.
23:13.51ka6sox-officejf: optware wiley?
23:14.31jeanfabriceshould be both as rod promote it to nslu2
23:15.36rwhitby-webas soon as jp30 pushes it will be there for the slug
23:17.05ka6sox-officeis Vegetable part of the Openslug Binary package?
23:18.13rwhitby-webjeanfabrice: [root@wl500gx root]$ adduser rwhitby
23:18.18rwhitby-webadduser: /home/rwhitby: No such file or directory
23:18.28rwhitby-web(repeated three times)
23:18.41rwhitby-webcause /home doesn't exist
23:18.42jeanfabriceadduser -H
23:19.04jeanfabriceor >/dev/null 2>&1
23:19.42rwhitby-webso will all usage of adduser (on both nslu2 and wl500g) require the -H flag, cause we want the behaviour to be the same and it cannot be supported on wl500g?
23:19.53rwhitby-webif so, we should hard-code it
23:19.56rwhitby-webas a default
23:20.40rwhitby-webjeanfabrice: what version did you push?  I get
23:20.53jeanfabricewhat if a end user want to add a user for its personal use ?
23:20.54rwhitby-weband it says it's a tinylogin applet
23:21.19jeanfabricerwhitby-away : pushing now... give me 1 or 2 mn :/
23:21.26rwhitby-webok, cool
23:22.09rwhitby-webok, so we may need to write up the package guidelines to make sure that all usage of adduser in installation scripts and rc scripts calls it with -H
23:23.04jeanfabricepackage compile and build but where not pushed :/
23:23.37rwhitby-webor we get Oleg to add a /home mount point to the firmware so we can bind mount it from somewhere
23:23.44jeanfabrice-H AND -D
23:23.57rwhitby-web(but that will not address people with existing firmware)
23:24.04jeanfabrice-D is for non interactive passwd call
23:24.22jeanfabricemay we hardcode /opt/home
23:25.01rwhitby-webcan we make those options default the other way?
23:25.30rwhitby-web(i.e. are there corresponding options to say "do create a home directory" and "do ask for the password interactively")
23:27.35jeanfabricerwhitby-web : nop. you MUST use these options not to create home or ask for passwd. That's the actual busybox scheme. We should document this nicely not changing the busybox behaviour applying many patches, shouldn't we ?
23:27.45jaroDoes anyone understand the boot process of wlhdd (using Oleg's firmware I'm reading linuxrc and noticed that there are pivot_root ; exec chroot commands invoked when booting from alternate device? What is the purpose of running the chroot?
23:27.51ka6sox-officeadduser automatically builds a home directory.
23:28.08jeanfabricephp is broken on nudi and forbid pushing packages
23:28.21rwhitby-webjeanfabrice: then /opt/home may be the best solution.
23:28.46rwhitby-webjeanfabrice: try running make libxml2-dirclean first
23:28.55ka6sox-officesince adduser will only be available with a HD attached anyway then /opt/home is a good choice.
23:28.58jeanfabriceone line patch : I should be abble to do it
23:29.26rwhitby-weband it fits the philosophy of Optware :-)
23:29.34jeanfabriceit broke on imap. sound strange as it compiled fine on my pc
23:29.46rwhitby-webjaro: that's the standard way to change the rootfs
23:30.09rwhitby-webwe do exactly the same thing on the slug
23:30.38rwhitby-webjeanfabrice: let me see if I can build wl500g packages in my area
23:30.53jeanfabricejaro: I think Oleg put it to give the possibility to boot on an external rootfs
23:31.39jarojeanfabrice: pivot_root is not enough?
23:32.30rwhitby-websee the pivot_root manpage
23:34.38jp30-worki am pushing the new adduser now
23:34.50jarorwithby-web: Thank you. I understand now.
23:35.30rwhitby-webjeanfabrice: make php-dirclean
23:35.50jeanfabriceI did
23:36.04jeanfabriceseemes that imap did not stage
23:36.28jeanfabriceI build imap, imap-ipk and imap-stage. rebuilding php now
23:38.16*** join/#nslu2-linux jstueve (
23:39.35jp30-workimap is not an autotooled package, so staging is done manually (and lazily, with cp). could be permissions trouble
23:41.20rwhitby-webjp30-work, jeanfabrice: (from logs): yes, please feel free to remove that conditional in busybox,mk
23:41.33rwhitby-webit was historical from when I copied the file from OpenWRT
23:41.49jeanfabricejp30-work : you don't make non autotooled packages autotooling ?
23:43.08jeanfabricerwhitby-web : It permits me - with the help of josh - to create my first "backrefer to subexpression" regexp
23:43.58rwhitby-webjeanfabrice: my packages build for wl500g just completed.
23:44.59jp30-workjf, it's not worth the effort to autotool packages that aren't designed for it. except in some special cases like the modularised Xlibs
23:46.27jp30-workUW imap has its own complicated configuration system, but doesn't have a usable "make install" target
23:46.41jeanfabricejp30-work : I was just kidding. As my english is very very bad, sometimes (often), it doesn't work
23:46.56jp30-workah, i wondered
23:47.20jeanfabricethat's why I should stop.
23:47.51jp30-worknow i'm embarrassed that i took you literally
23:48.05jeanfabricerwhitby-web : I've pushed optware to the feed.
23:48.27jp30-workjf, imap built and staged ok in the end, then?
23:49.07jeanfabriceseems to, manually done. I should distclean everything one of these days
23:50.47jp30-workalways a good idea. staging gets full of crud, and then you discover that it was only because you had the remains of some long lost version of something staged that anything else would build at all
23:52.19ka6sox-officerwhitby-web, the simplest way for me to amend the ~nslu2-rules is to just overwrite it.
23:52.30ka6sox-officeso "don't panic"
23:52.44jbotextra, extra, read all about it, nslu2-rules is 1) Those who ask the questions, update the wiki, 2) Those who complain about the doco, update the wiki, 3) Friends don't let friends flash custom firmware without confirmed RedBoot access, 4) Friends don't let friends flash custom bootloaders without confirmed JTAG access.
23:54.34jeanfabricePatching adduser to hardcode /opt/home as default top home directory. Should I also make cosmetic change for GECOS as it is now 'Linux User,,,' ?
23:56.27ka6sox-officejbot, no nslu2-rules is 1. You will search the wiki first here before asking questions in channel or on the mailing list.
23:56.28jbotokay, ka6sox-office
23:56.28ka6sox-office2. Those who ask the questions, update the wiki when they get the answers.
23:56.28ka6sox-office3. Those who complain about the documentation, update the wiki to make it better.
23:56.28ka6sox-office4. Friends don't let friends flash custom firmware without confirmed RedBoot access.
23:56.28ka6sox-office5. Friends don't let friends flash custom bootloaders without confirmed JTAG access.
23:56.29ka6sox-office6. You must read and follow the steps in the README file precisely when flashing firmware.
23:56.38jbotmethinks nslu2-rules is 1. You will search the wiki first here before asking questions in channel or on the mailing list.
23:57.27rwhitby-webjeanfabrice: adduser works as advertised on wl500g
23:57.57ka6sox-officejbot, no nslu2-rules is 1. You will search the wiki first here before asking questions in channel or on the mailing list. 2. Those who ask the questions, update the wiki when they get the answers. 3. Those who complain about the documentation, update the wiki to make it better. 4. Friends don't let friends flash custom firmware without confirmed RedBoot access. 5. Friends don't let friends flash custom bootloaders without
23:57.58jbotka6sox-office: okay
23:57.58ka6sox-officeconfirmed JTAG access. 6. You must read and follow the steps in the README file precisely when flashing firmware.
23:58.07jbotnslu2-rules is, like, 1. You will search the wiki first here before asking questions in channel or on the mailing list. 2. Those who ask the questions, update the wiki when they get the answers. 3. Those who complain about the documentation, update the wiki to make it better. 4. Friends don't let friends flash custom firmware without confirmed RedBoot access. 5. ...
23:58.23rwhitby-webjeanfabrice: I wouldn't bother about patching the default home dir, as scripts will need to set that manually anyway
23:58.36jeanfabriceok. less patch
23:59.00ka6sox-officelooks like I'll have to abbreivate...
23:59.17jeanfabriceI can then go to bed as wife is now sleeping for 3 hours...
23:59.35rwhitby-webjeanfabrice: night!
23:59.53jp30-worknight, jf. i'm heading home; i'll drop back into the channel when i get there if anyone needs me

Generated by by Jeff Waugh - find it at! Modified by Tim Riker to work with blootbot logs, split per channel, etc.