| 01:50.40 | *** join/#nslu2-linux rwhitby-away (~rwhitby@CPE-144-136-199-164.sa.bigpond.net.au) |
| 02:06.12 | *** join/#nslu2-linux Christopher (~Christoph@128-193-244-50.resnet.oregonstate.edu) |
| 02:26.12 | *** join/#nslu2-linux kergoth_ (~kergoth@proxyle02.ext.ti.com) |
| 02:33.59 | *** join/#nslu2-linux jimbuzbee (~jimbuzbee@c-24-9-209-249.client.comcast.net) |
| 02:50.10 | *** join/#nslu2-linux [g2] (~g2@rdu57-8-035.nc.rr.com) |
| 02:58.05 | Christopher | [g2] have jtag yet? |
| 02:58.54 | [g2] | Christopher, soon I hope, a slug was being modified a couple days ago :) |
| 02:59.03 | Christopher | cool |
| 02:59.09 | [g2] | I noticed there was an update to the sw |
| 02:59.20 | Christopher | upslug? |
| 02:59.30 | [g2] | couple days ago, bug fix right ? |
| 02:59.34 | Christopher | yep |
| 02:59.43 | Christopher | forgot to initialize a variable to 0 |
| 02:59.47 | [g2] | :) |
| 02:59.50 | Christopher | caused problems for some and not for others |
| 03:00.12 | [g2] | starting from a known state is good in sw (tm) |
| 03:00.13 | Christopher | at least it was easily fixable, but I should have caught it sooner |
| 03:00.40 | [g2] | "with enough eyes all bugs are shallow" |
| 03:01.12 | Christopher | true |
| 03:01.26 | [g2] | horde programming :) |
| 03:03.50 | *** join/#nslu2-linux KsiLaptop (russ@d47-69-99-106.try.wideopenwest.com) |
| 03:04.02 | [g2] | and vica-versa GPL is people wasting and BSD is time-wasting |
| 03:04.17 | [g2] | in general :) |
| 03:06.46 | Christopher | i c |
| 03:08.30 | rwhitby | was that horde programming or hurd programming ... ;-) |
| 03:09.00 | [g2] | we all know hurd programming *never* ships :) |
| 03:09.33 | [g2] | hey 100% of the polled individuals supports after in the path :) |
| 03:14.50 | glc | The poll is based on a false premis. The PATH is the rc is not seen by the shell. |
| 03:15.47 | glc | GO to your serial port without logging in and echo $PATH. /opt/bin and /opt/sbin aree not in the PATH. |
| 03:25.28 | rwhitby | Poll removed, path in /etc/rc.d/rc.sysinit changed. |
| 03:31.01 | [g2] | rwhitby, ever have any luck stracing rc.bootbin ? |
| 03:34.28 | rwhitby | never tried |
| 03:34.38 | rwhitby | works fine for me :-) |
| 03:35.45 | rwhitby | fixed patch - fseeko problem |
| 03:45.01 | *** join/#nslu2-linux jimbuzbee (~jimbuzbee@c-24-9-209-249.client.comcast.net) |
| 03:45.16 | [g2] | jimbuzbee, wb :) |
| 03:46.36 | jimbuzbee | Hey G2, Sorry I've been away so long :-) |
| 03:47.17 | [g2] | jimbuzbee, well you know you have GrandFather status so you're always welcome here :) |
| 03:47.25 | jimbuzbee | Thanks! |
| 03:47.33 | [g2] | How kuro hacking ? |
| 03:47.38 | [g2] | How is |
| 03:47.56 | [g2] | that Kuro HD looks *sexy* |
| 03:48.03 | jimbuzbee | :-) |
| 03:48.12 | jimbuzbee | I've not done much with it |
| 03:48.20 | [g2] | do you have one ? |
| 03:48.52 | jimbuzbee | Yeah, I have one. but I haven't done a whole lot other than bring it up and install Apache, and some other packages |
| 03:49.08 | jimbuzbee | It doesn't need a -whole- lot of hacking :-) |
| 03:49.18 | [g2] | cool! Any idea on throughput on the GigE ? |
| 03:49.35 | Christopher | so they took the fun out of it :-) |
| 03:49.47 | jimbuzbee | I don't have the GigE model ( oops... you were referring to the new model ) |
| 03:49.54 | [g2] | it's running a bunch of ancient apps rigth ? |
| 03:50.12 | jimbuzbee | Yeah, the base system is a bit crude |
| 03:50.30 | [g2] | re: "doesn't need a -whole- lot of hacking :-)" |
| 03:51.01 | jimbuzbee | I've been tied up writing for tomsneetworking... They are keeping me busy |
| 03:51.18 | [g2] | great! anything you can talk about ? |
| 03:51.37 | [g2] | or will you have to nuke are network boxen if you mention it :) |
| 03:51.48 | [g2] | lol |
| 03:52.09 | jimbuzbee | I just got done with a review of a couple of Viewsonic devices : media adaptor and media gateway |
| 03:52.17 | jimbuzbee | They need work :-( |
| 03:52.59 | jimbuzbee | The gateway runs Linux, it is same board as NSLU@ |
| 03:53.01 | jimbuzbee | 2 |
| 03:53.04 | [g2] | :-(, they've got some neat products. I've *wanted* to hack that tablet XP console extender |
| 03:53.35 | jimbuzbee | I haven't played with that one |
| 03:54.07 | [g2] | there's a really cool tablet out I saw mentioned in the comdex write-ups |
| 03:54.27 | [g2] | and the Linux PMPs will be cool |
| 03:54.59 | jimbuzbee | Yeah, there is a lot of cool stuff out there now! |
| 03:55.50 | [g2] | 2005 is going to be a great hacking year |
| 03:56.15 | [g2] | did you see cringleys' predictions ? |
| 03:56.30 | jimbuzbee | Yeah, I read those yesterday. He's big on the wrt54g! |
| 03:56.53 | [g2] | OpenWrt the missed 800K opportunity for 2004 |
| 03:57.10 | [g2] | now that's a story worth writing about |
| 03:57.24 | jimbuzbee | 800k? |
| 03:58.02 | Tiersten | Christopher: You still around? |
| 03:58.09 | [g2] | 44K users @ $20 |
| 03:58.15 | Christopher | yep |
| 03:58.32 | [g2] | here that tablet that I think looks cool http://www.engadget.com/entry/4618484394872473/ |
| 03:58.37 | jimbuzbee | Oh yeah. I did that math too! |
| 03:58.47 | Tiersten | Got some changes for upslug |
| 03:58.54 | Christopher | sure, list em :-) |
| 03:59.03 | Tiersten | Like line 686. you've got && instead of || |
| 03:59.15 | Christopher | 1 sec |
| 03:59.35 | [g2] | jimbuzbee, Happy New Year and have a wonderful year |
| 03:59.45 | jimbuzbee | Hey, you too! |
| 03:59.48 | Christopher | Tiersten: newest version from CVS? |
| 03:59.49 | [g2] | thx |
| 03:59.57 | Tiersten | erm. the public CVS anyway... |
| 04:00.10 | Christopher | hmmm, I have an fflush(stdin); there |
| 04:00.13 | Tiersten | ah |
| 04:00.15 | Christopher | at line 686 |
| 04:00.24 | Tiersten | Find "/*Look for Sercomm signature*/" |
| 04:00.34 | Tiersten | The line after that where you're trying to check for the eRcOmM sig |
| 04:00.34 | Christopher | k |
| 04:00.57 | Christopher | i c |
| 04:01.08 | Christopher | fixed that one |
| 04:01.31 | Tiersten | Using memcmp would probably be nicer looking |
| 04:01.52 | Tiersten | upslug.h in showhelp, for the -b line. defaults is spelt wrong :) |
| 04:02.00 | Christopher | ok |
| 04:02.45 | Christopher | it's commented out :-) |
| 04:02.54 | Christopher | but I'll fix it |
| 04:03.02 | Tiersten | ya. but still a typo :) |
| 04:03.56 | Christopher | fixed that |
| 04:04.20 | Tiersten | Putting -Wall into the compiler flags should make it easier to detect uninitialised vars as well |
| 04:04.35 | Tiersten | You'll need to add a bunch of extra includes however |
| 04:04.57 | Tiersten | stdlib.h, string.h, and netinet/in.h |
| 04:05.33 | Christopher | ok, I've not had much experience with gcc, what does -Wall do? |
| 04:05.43 | Tiersten | Prints out extra warnings |
| 04:05.52 | [g2] | -W (Warning level) all |
| 04:06.06 | Christopher | ok |
| 04:06.16 | Tiersten | That's how I knew about the sercomm sig line |
| 04:06.25 | Tiersten | It was complaining that line with ambiguous |
| 04:06.43 | Tiersten | It'll ask you to change a few while () lines as well |
| 04:06.48 | Christopher | ok |
| 04:07.01 | Tiersten | Apart from that one bug, no other problems :) |
| 04:07.05 | Tiersten | Continue with the good work |
| 04:07.10 | Christopher | alright, thanks for informing me |
| 04:09.48 | Christopher | what happened to CIA? |
| 04:12.12 | rwhitby | hey jimbuzbee! |
| 04:12.28 | jimbuzbee | Hey Rod! How's it going? |
| 04:12.39 | rwhitby | good: www.rwhitby.net/elise.html |
| 04:13.10 | jimbuzbee | Congrats! I guess I can't say anything about myself being busy! |
| 04:15.04 | rwhitby | what's the model number of the viewsonic with the same board as the slug? |
| 04:15.19 | jimbuzbee | WMG120 or WMG80 |
| 04:15.26 | rwhitby | glc: ping |
| 04:15.38 | glc | pong. |
| 04:15.58 | rwhitby | can you send me an rsa1 key please? |
| 04:16.33 | rwhitby | jimbuzbee: twice the flash and ram? |
| 04:17.13 | glc | I'll have to generate one. |
| 04:17.25 | rwhitby | make sure it's rsa*1* |
| 04:17.26 | jimbuzbee | Rod - I don't know. I never got a command line on it - no time :-( I did verify that I could run an executable from the wmg on the slug |
| 04:17.45 | rwhitby | didn't open it up and take pictures? |
| 04:18.51 | jimbuzbee | Yeah, I did, but the chips don't mean much to me :-) My article will have pictures |
| 04:20.36 | rwhitby | got a sneak peek at the pictures for us? |
| 04:21.05 | jimbuzbee | Give me a sec... |
| 04:23.29 | rwhitby | you know how much we love internal shots here :-) |
| 04:23.31 | Christopher | same board with twice flash and RAM? |
| 04:23.44 | Christopher | :-) |
| 04:23.49 | rwhitby | ViewSlug |
| 04:23.57 | Christopher | how much? |
| 04:24.15 | rwhitby | ~google wmg120 |
| 04:24.19 | [g2] | $449 after rebate |
| 04:24.29 | Christopher | ouch |
| 04:24.35 | [g2] | $399 for the 80 Gig |
| 04:24.46 | rwhitby | it's a wireless router too |
| 04:25.06 | [g2] | they were selling wrt54g's for $35 at BB the other day |
| 04:25.13 | rwhitby | those antennae are expensive :-) |
| 04:25.19 | Christopher | i got a belkin wireless g for $10 |
| 04:25.53 | rwhitby | convergence aways costs the consumer ... |
| 04:25.54 | [g2] | when the baby starts eating ceral, they'll be giving APs away in the boxes :) |
| 04:26.06 | rwhitby | that's my triple C theory |
| 04:26.25 | Christopher | make the firmware for the wmg work on a fatslug :-) |
| 04:26.29 | [g2] | 64MB ram, 16MB flash |
| 04:27.18 | [g2] | doesn't mention USB 2.0 |
| 04:27.45 | jimbuzbee | http://wrt54g.homelinux.org/Figure6-full.jpg |
| 04:28.09 | Christopher | maybe we can "steal" the print sever software for those people that want their slug to be a print server |
| 04:28.18 | rwhitby | buzbee's home DSL connection slows to a crawl .... |
| 04:28.44 | [g2] | slugdotted |
| 04:29.18 | jimbuzbee | My uplink is pitifull... |
| 04:29.35 | [g2] | minipci yummy |
| 04:30.00 | rwhitby | P1 and JP4? |
| 04:30.29 | [g2] | and P4 |
| 04:30.37 | [g2] | JTAG and 2 serial ? |
| 04:30.53 | rwhitby | P2 and P4 just look like disk power to me |
| 04:31.02 | rwhitby | actually P2 is something else |
| 04:31.08 | Christopher | what is the via chip? |
| 04:31.09 | rwhitby | P4 is just power |
| 04:31.35 | rwhitby | usb 2.p0 |
| 04:31.37 | rwhitby | 2.0 |
| 04:32.01 | Christopher | yep, just looked it up |
| 04:32.28 | jimbuzbee | I should consult with you guys before I finish an article :-) |
| 04:32.53 | rwhitby | so you could potentially add two more USB ports to it on the VIA chip |
| 04:33.10 | rwhitby | got an underside photo? |
| 04:33.25 | Christopher | what controls the IDE? |
| 04:33.29 | rwhitby | IXP420 chip like the slug, it looks like. |
| 04:33.30 | [g2] | and ix that a ixp420 ? |
| 04:33.38 | jimbuzbee | Rod : no |
| 04:33.45 | [g2] | and is |
| 04:33.51 | [g2] | 425 ? |
| 04:34.12 | rwhitby | looks like a 0 before the BB |
| 04:34.21 | rwhitby | maybe more chips under the wlan card? |
| 04:34.29 | [g2] | your eyes are better than mine :) |
| 04:34.35 | [g2] | you young guy :) |
| 04:35.44 | [g2] | actaully, you could get a wrv54g and a Kuro Box for less |
| 04:36.22 | [g2] | $329 = $180 + $149 |
| 04:36.32 | Christopher | yeah |
| 04:36.34 | [g2] | and still save $70 |
| 04:37.53 | [g2] | jbot 3Cs is Convergence always Costs the Consumer |
| 04:37.55 | jbot | okay, [g2] |
| 04:38.03 | [g2] | ~botsnack |
| 04:38.03 | jbot | [g2]: thanks |
| 04:38.08 | [g2] | ~3cs |
| 04:38.09 | jbot | extra, extra, read all about it, 3cs is Convergence always Costs the Consumer |
| 04:38.45 | [g2] | or should that be TripleC |
| 04:39.30 | rwhitby | back later guys - good to see you again Jim |
| 04:40.05 | jimbuzbee | See 'ya Rod! |
| 04:40.29 | [g2] | jimbuzbee, thx for stopping by, nice seeing you again |
| 04:40.34 | [g2] | nite all |
| 04:40.40 | jimbuzbee | Yeah - good to be back! |
| 04:40.42 | Christopher | nite [g2] |
| 04:40.58 | *** part/#nslu2-linux [g2] (~g2@rdu57-8-035.nc.rr.com) |
| 04:41.06 | rwhitby-away | glc: you now have bk access - please push the patch you sent me, after you check whether the copying of the /lib/ stuff is needed in the stanza where you removed the portmap |
| 04:46.24 | Christopher | bye jimbuzbee |
| 04:47.06 | *** join/#nslu2-linux [g2] (~g2@rdu57-8-035.nc.rr.com) |
| 04:52.35 | Christopher | back [g2]? |
| 04:53.14 | Christopher | Tiersten: if you want you can rewrite that line with a memcmp command, I keep getting segfaults, not sure what I'm doing wrong |
| 04:56.57 | [g2] | Christopher, I need to get help from kergoth on something |
| 04:57.23 | Christopher | i c |
| 05:16.32 | [g2] | rwhitby FYI with some minor changes and including a couple directories, the 3.17 tar is bitbaking (thanks to a little help from kergoth :) |
| 05:45.39 | glc | rwhitby-away: Reading up on BK. Do I need to re-pull or can I reparent, and what are new parents? Will start testing tomorrow after I get the feel for this. |
| 05:49.58 | glc | Good Night. |
| 07:08.10 | *** join/#nslu2-linux LarsAC (~chatzilla@pD9500D47.dip0.t-ipconnect.de) |
| 07:17.18 | ka6sox | if anybody sees rwhitby-away let him know I'm *finally* home. |
| 07:25.51 | Christopher | nite all |
| 07:26.27 | ka6sox | nite Christopher |
| 07:30.48 | rwhitby-away | ka6sox: ping |
| 07:49.53 | ka6sox | Wedhttp://www.opl.ucsb.edu/grace/lc/lc10.html |
| 07:49.57 | ka6sox | oops |
| 07:50.00 | ka6sox | sorry. |
| 08:12.30 | giel | morning |
| 08:20.07 | ka6sox | hi |
| 08:28.45 | *** join/#nslu2-linux perlguru-work (www-data@netblock-66-159-209-60.dslextreme.com) |
| 08:30.28 | perlguru-work | goodmorning |
| 08:30.32 | giel | mogguh |
| 08:35.51 | siddy | morning. |
| 08:36.05 | perlguru-work | morning |
| 08:36.18 | rwhitby | morning Europe |
| 08:36.54 | siddy | evening, down under! ;) |
| 08:37.06 | rwhitby | siddy: there was something I was meant to reply to you, but I forgot what it was |
| 08:37.20 | rwhitby | do you know (to save me looking in logs again) |
| 08:37.32 | siddy | perlguru-work: your torrent-scripts seem to work fine, no problems during the night. |
| 08:38.05 | siddy | rwhitby: I don't know... the howto, my bug-report, gcc ? dunno |
| 08:38.26 | perlguru-work | siddy, ok. i'll tweak them a bit more and then submit the package |
| 08:39.41 | perlguru-work | rwhitby: have you seen inge fix for the mount problem? |
| 08:40.20 | rwhitby | which one? the /opt/bin placement? |
| 08:40.45 | perlguru-work | the missing sda1 mount problem |
| 08:40.56 | rwhitby | yeah, I got the problem, which fix? |
| 08:41.44 | perlguru-work | http://www.nslu2-linux.org/wiki/HowTo/TroubleshootADriveNotFormattedCondition |
| 08:42.46 | rwhitby | that's a bit drastic, when others on the list say that simply moving /opt/bin and /opt/sbin to the end of the path in /etc/rc.d/rc.sysinit (but not in /etc/profile) fixes the problem. |
| 08:44.24 | perlguru-work | I haven't heard that one before :/ |
| 08:44.38 | rwhitby | only happened while you were asleep last night :-) |
| 08:44.46 | rwhitby | ~slugtime |
| 08:44.47 | jbot | slugtime is, like, 40 hour days 10 hours in 4 timezones with overlap |
| 08:44.57 | perlguru-work | ok. found the root-cause? |
| 08:45.30 | perlguru-work | i am about the read yesterdays log |
| 08:46.02 | rwhitby | http://groups.yahoo.com/group/nslu2-linux/message/3313 |
| 08:46.16 | siddy | rwhitby: nearly forgot: could you add http://www.nslu2-linux.org/wiki/HowTo/EnableTrafficShaping to the howto-page? |
| 08:46.23 | perlguru-work | much can happen: http://www.nslu2-linux.org/wiki/Info/BootSequencePage |
| 08:46.59 | rwhitby | perlguru-work: yep, saw that one in the logs |
| 08:47.03 | perlguru-work | i started it with only two point for rc.bootbin and now almost everything is there ;) |
| 08:47.47 | rwhitby | siddy: done |
| 08:47.58 | siddy | perfect, thx! |
| 08:50.01 | rwhitby | siddy: so you're ready to suggest a new defconfig for -able? |
| 08:50.12 | giel | hm... i need to get myself more memory for my main machine |
| 08:50.30 | giel | compiling with openembedded is no fun at all now |
| 08:51.12 | siddy | rwhitby: yeah, the additions are in the howto, running stable so far. |
| 08:51.44 | rwhitby | so QOS is the odd man out? |
| 08:51.56 | rwhitby | BTW, patch now works. |
| 08:52.13 | siddy | QoS is what? |
| 08:52.46 | rwhitby | in the defconfig: CONFIG_NET_QOS is not set ... intentional? |
| 08:53.47 | siddy | ah, yes. QoS is a service we don't really need, (and that is not really supported by ipv4) |
| 08:54.28 | siddy | we need the fair-queueing-part of the kernel, that's what most people mean by "QoS" |
| 08:55.00 | giel | i'm using QoS in my router :) |
| 08:58.05 | giel | and i think the main idea behind the general term QoS is just to rate limit certain queues |
| 08:58.09 | *** join/#nslu2-linux LarsAC (~chatzilla@dagama.lpt.RWTH-Aachen.DE) |
| 09:00.14 | siddy | not really... QoS is needed to have guaranteed rates for a connection, so you say: I need constant 256kbit for the connection to my music-streaming-server, and all routers on the way to the server have to agree on that. But... it's useless as long as the routers of your provider don't use QoS, and I don't know a single provider that would. |
| 09:00.52 | giel | true |
| 09:01.31 | giel | ppl use the same terminology different all the time |
| 09:01.41 | giel | i'm having four separate queues in my router |
| 09:01.54 | giel | best effort, medium priority, high priority and real time |
| 09:02.29 | giel | so, although it's not rate guaranteed, it's QoS anyway IMO |
| 09:02.36 | siddy | but you use it only for your lan, do you? everythiong outside will probably be "best effort" :-\ |
| 09:02.47 | giel | no, those are the send queues to the outside world |
| 09:03.01 | siddy | ah, k |
| 09:03.07 | giel | in my dsl modem |
| 09:04.05 | giel | but that doesn't really matter, if you've got four sendqueues in the router on your LAN and you merge them into one rate limited queue to your modem you should in theory have the same effect |
| 09:05.59 | giel | but the line between fair queueing and QoS is thin ;) |
| 09:07.27 | siddy | hm... the term "QoS" is strictly bound to a bunch of protocols, but people like to mix QoS and fair queueing... I guess QoS just sounds better. ;-) |
| 09:07.52 | giel | and then there's ToS getting in there too |
| 09:07.57 | giel | but that's just a tagging of packets |
| 09:08.18 | siddy | but speaking in the terms of the linux-kernel-config: we need fair queueing and we do not need QoS. |
| 09:09.31 | giel | yes |
| 09:10.15 | giel | i never argued that :) |
| 09:10.50 | siddy | Man, I'm desperately waiting for IPv6 to go out to the "real world". Then we can play around with all those funny features IPv4 doesn't support. |
| 09:11.05 | giel | uhu |
| 09:11.23 | giel | there's this dutch provider which gives you native ipv6 with their most expensive dsl plan |
| 09:12.34 | siddy | yeah, my beloved provider would also give me IPv6 for a lot of money, but I do not see much sence in that atm. |
| 09:12.58 | giel | and i don't like using those tunnels |
| 09:13.09 | giel | the term 'ip broken' just doesn't do it for me ;) |
| 09:13.14 | giel | broker* |
| 09:13.29 | siddy | :-) |
| 09:15.01 | giel | man, i'm so going to rip all the memory out of all the machines i have laying around and put them in my main one |
| 09:15.08 | giel | bitbake is a memory hog |
| 09:15.28 | siddy | there are some trick's on the wiki to reduce the memory-usage. |
| 09:15.46 | perlguru-work | http://www.nslu2-linux.org/wiki/OpenEmbedded/NSLU2PackageSymlinks |
| 09:15.52 | perlguru-work | :) |
| 09:16.00 | siddy | that's the one. |
| 09:16.23 | perlguru-work | giel should read the wiki more often :) |
| 09:16.43 | giel | perlguru-work: i should read all of the wiki :) |
| 09:16.53 | giel | but i'm reading it as much as my limited time allows :) |
| 09:17.11 | perlguru-work | bbl |
| 09:17.19 | giel | and puhlease guys, don't you see i'm just preparing myself for an excuse to buy more memory? :) |
| 09:17.37 | siddy | gg |
| 09:19.17 | siddy | "640K ought to be enough for anybody." Bill Gates, 1981. |
| 09:19.20 | siddy | ;) |
| 09:19.39 | giel | 'uh yeah, i need it so i can compile uh... this stuff for this device i bought while i didn't need it but it was kinda cool... and uh... it works but i'm trying to break it so uh... i need more memory?' |
| 09:19.43 | giel | yeah, that'll fool me |
| 09:22.20 | rwhitby | then you need to buy a second slug so that you can use twice the output of your compilations that you are doing so fast with your new memory ... |
| 09:24.05 | giel | and more harddrives ofcourse! to uh... store all the stuff i download! |
| 09:31.44 | giel | i like the term 'slugfarm' |
| 09:31.45 | giel | :) |
| 09:32.05 | giel | it reminds me of abe's odyssee somehow |
| 09:53.47 | perlguru-work | rwhitby: any problem if I split up the bootsequence page into: bootsequence and LinksysBinaries ? |
| 09:55.13 | perlguru-work | rwhitby: is rc.bootbin divertable now in the repo? |
| 10:03.44 | *** join/#nslu2-linux siddy_ (~sid@p50927687.dip.t-dialin.net) |
| 10:13.17 | *** join/#nslu2-linux inge_arnesen (~bob@254.80-203-34.nextgentel.com) |
| 10:14.57 | rwhitby | perlguru-work: no, no. |
| 10:16.37 | perlguru-work | ok, ok |
| 10:17.30 | perlguru-work | Maybe a new beta would be desired to fix the 'NotFormatted' problem. |
| 10:17.53 | rwhitby | as soon as it is verified as a fix for a majority of people with the problem, then it will be so :-) |
| 10:18.06 | rwhitby | at the moment, it's a theory with a single data point ... |
| 10:19.18 | perlguru-work | so, what is the fix? Get the path straight? |
| 10:21.14 | rwhitby | In /etc/rc - put /opt dirs at the end instead of the start. |
| 10:22.01 | perlguru-work | how many people have to try this? |
| 10:22.30 | rwhitby | we haven't done a diversion in the middle of a script before, so I need to work out the best way to do it whilst getting the same diversion semantics |
| 10:23.24 | perlguru-work | That's why i added the info on BootsequencePage, to make clear when what is called from where. :) |
| 10:24.30 | perlguru-work | The .xls sheet is very helpfull too |
| 10:24.39 | giel | gah! excel! |
| 10:25.04 | perlguru-work | I use openoffice. |
| 10:26.36 | giel | still: gah! excel! |
| 10:49.59 | *** join/#nslu2-linux bob_tm (~bob@254.80-203-34.nextgentel.com) |
| 10:50.31 | rwhitby | perlguru-work: by semantics, I just meant how to do the actual diversion of rc.bootbin, not try and replace it. |
| 10:51.37 | perlguru-work | You mean, is the normal diversion way, the right way for rc.bootbin? |
| 10:53.15 | perlguru-work | the echo before calling rc.bootbin is also very confusing |
| 10:53.38 | rwhitby | no, just simply how to code the diversion for a binary called in the middle of a script, as opposed to diverting a whole script from inside that script |
| 10:54.03 | rwhitby | agreed. I have no idea why SerComm put it in their |
| 10:54.05 | rwhitby | there |
| 10:55.34 | perlguru-work | Calls the diversion like an external program and check $? |
| 10:57.09 | *** join/#nslu2-linux LexDelani-work (~lex@cc571156-a.hooge1.gr.home.nl) |
| 10:57.18 | LexDelani-work | morning all |
| 10:57.29 | perlguru-work | hi |
| 10:57.36 | bob_tm | Would it be possible for a diversion to call rc.bootbin and then add its own functionality, like verifying FAT drives, mounting them and validating them in CGI_ds.conf? |
| 10:59.41 | rwhitby | diversions can do anything. return 1 and the normal stuff happens to (for prepending). return 0 and the normal stuff doesn't happen (for replacing). return 0 and call the orginal from inside the diversion script (for appending, or both prepending and appending). |
| 11:00.28 | bob_tm | What prevents looping? |
| 11:00.51 | perlguru-work | good point ;) |
| 11:01.35 | perlguru-work | rc.1 -> rc.bootbin.divert -> rc.bootbin -> rc.1 resumes |
| 11:02.25 | perlguru-work | btw rc.bootbin does not check diversions |
| 11:02.38 | bob_tm | Of course, it is binary. |
| 11:02.43 | perlguru-work | rc.1 should test for rc.bootbin diversions |
| 11:02.53 | rwhitby | how would a loop occur? |
| 11:03.04 | perlguru-work | it wouldn't |
| 11:03.20 | rwhitby | I was asking bob_tm :-) |
| 11:03.29 | bob_tm | It wouldn't there, but in general if you want to do something after the original, it is not easy since diverts are at the beginning |
| 11:04.00 | rwhitby | bob_tm: Have you read the code which implements diversions? |
| 11:04.16 | rwhitby | You can't get loops |
| 11:04.26 | bob_tm | I know you don't, but I don't get how |
| 11:04.38 | perlguru-work | bob_tm: or test your assumption |
| 11:04.53 | perlguru-work | see if you get it to loop ;) |
| 11:05.03 | bob_tm | I did test, remember, perguru - the rc.local bug - it didn't cause a loop, but I thought it should |
| 11:05.23 | bob_tm | Nevermind - I will figure it out. |
| 11:06.14 | rwhitby | when I say "call the original" I mean do the stuff that the original does, not actually call the original script. Sorry - I should be more clear. |
| 11:06.39 | bob_tm | Ok - I _did_ call the original, but it didn't crash |
| 11:07.13 | rwhitby | what did you return from the diversion script? |
| 11:07.37 | bob_tm | So: rc.local -> rc.local.diversion (with diversion test still in there) which should cause a loop |
| 11:07.48 | bob_tm | But it didn't fail. |
| 11:08.37 | bob_tm | If I would like to append functionality it would be similar: rc.x -> rc.x.divert -> (do original without code duplication) -> rc.x.divert continues |
| 11:08.58 | rwhitby | let's get it straight: /etc/rc.d/rc.local has code in it which calls /unslung/rc.local. If /unslung/rc.local calls /etc/rc.d/rc.local, then you should get a loop |
| 11:09.25 | perlguru-work | . |
| 11:09.35 | bob_tm | Well, it wasn't like that. rc.local.divert called itself |
| 11:10.02 | bob_tm | ...via a diversion script. Anyhow, that is a bug. What is the right way to append functionality without code duplication? |
| 11:10.18 | bob_tm | Normal diversion can either replace or prepend functionality |
| 11:10.36 | rwhitby | there is no way to do it without code duplication |
| 11:10.47 | perlguru-work | We want something like in OO: a subclass want to call the method from its superclass :) |
| 11:10.48 | bob_tm | Ah, ok. Then we agree |
| 11:11.13 | rwhitby | except perhaps put it as a prepend for the following diversion :-) |
| 11:12.37 | bob_tm | The reason I asked was because you said: "return 0 and call the original from inside the diversion script". |
| 11:13.20 | bob_tm | so we agree |
| 11:13.36 | rwhitby | yeah, I misspoke |
| 11:14.02 | bob_tm | Do with have utilities to deal with CGI_ds.conf? |
| 11:14.16 | perlguru-work | not yet |
| 11:14.37 | perlguru-work | or sed /dd and alike |
| 11:14.56 | bob_tm | All CGI_ds.conf access is all single task, so no need for locking? |
| 11:15.14 | bob_tm | (at least in rc it would be) |
| 11:16.18 | perlguru-work | at boottime there are not many users logged in, right? :) |
| 11:16.49 | bob_tm | Prior to start of /usr/sbin/USB_Detect there is no problem, but since it access it... |
| 11:16.55 | rwhitby | bob_tm: we know very little about CGI_ds.conf access. |
| 11:17.04 | rwhitby | a lot of it is in the .cgi binaries |
| 11:17.34 | bob_tm | As perlguru says, we are probably safe prior to start of thttp and USB_Detect. |
| 11:18.20 | bob_tm | (well you didn't say that... :-o) |
| 11:19.27 | rwhitby | I would concur with that assumption |
| 11:20.32 | rwhitby | For those reading the logs: |
| 11:21.12 | rwhitby | TARGET_STRIP has now lost it's options, (so that it's just like any other TARGET_* variable), and has been added to TARGET_CONFIGURE_OPTS |
| 11:21.22 | rwhitby | (this is in Unslung CVS *.mk files) |
| 11:22.10 | rwhitby | usage of TARGET_STRIP in the -ipk targets has been replaced by a new STRIP_COMMAND variable which includes the options --remove-section=.comment --remove-section=.note and --strip-unneeded |
| 11:22.57 | rwhitby | I'm doing a world build now to confirm I made the changes correctly, then I will do a massive checkin of all the *.mk files |
| 11:32.31 | rwhitby | bob_tm: do you know what it is about insmod which is incompatible? |
| 11:32.58 | bob_tm | No |
| 11:33.30 | bob_tm | Should we go for explicit paths everywhere in rc-scripts? |
| 11:34.40 | rwhitby | yes, and we should also have a path in rc scripts which does not allow this thing to break compatibility |
| 11:35.47 | bob_tm | Just did brief testing - seemed to work nicely from the command line |
| 11:36.46 | bob_tm | It can be interaction with the situation at boot time. |
| 11:44.35 | rwhitby | ok, I'm removing the /opt/sbin/insmod symlink from the busybox package. If people want the busybox insmod instead of stock, then they can call "/opt/bin/busybox insmod" or make their own link. This is also what we do for fdisk. |
| 11:45.12 | perlguru-work | good idea |
| 11:45.19 | LexDelani-work | good call rwhitby |
| 11:45.31 | bob_tm | Sounds good. But: Any other verification of insmod being the problem? |
| 11:45.41 | bob_tm | It is a good thing anyhow. |
| 11:46.05 | jacques | they modified insmod??? |
| 11:46.11 | LexDelani-work | my nslu2 is purring quite happy at the moment. only thing missing for me would be a ipkg install gcc :) |
| 11:46.20 | bob_tm | Second that |
| 11:46.36 | rwhitby | bob_tm: are you saying you're not sure it's insmod which is the culprit? |
| 11:46.57 | rwhitby | LexDelani-work: we're working on ipkg install crosstool-native .... |
| 11:47.04 | bob_tm | Jacques: There are no diffs in Linksys' source distro that I can see, so any changes must be BB people. |
| 11:47.31 | rwhitby | or the way we have configured busybox 1.00 |
| 11:47.50 | bob_tm | Could be. |
| 11:48.00 | LexDelani-work | rwhitby: would crosstool-native be the same thing? |
| 11:48.17 | bob_tm | I will do a test - insert explicit paths in all rc-scripts and have BB 1.0 fully in place. |
| 11:48.25 | LexDelani-work | rwhitby: or would it be like crosscompiling to itself :S |
| 11:48.41 | jacques | crosstool-native kinda works |
| 11:48.44 | jacques | but has issues |
| 11:49.08 | jacques | it can build some simple things but not perl |
| 11:49.17 | LexDelani-work | a plain native gcc would be nice |
| 11:49.29 | rwhitby | jacques: you did the manual install? |
| 11:49.29 | jacques | define that please |
| 11:49.37 | jacques | "plain native gcc" |
| 11:50.01 | jacques | rwhitby, yeah, and I had to use gnu tar as the bb one was dropping files |
| 11:50.05 | LexDelani-work | jacques, well, a arm-gcc that generates correct arm code without much fuss |
| 11:50.21 | jacques | lol |
| 11:50.24 | jacques | okaaaay |
| 11:50.38 | LexDelani-work | like most properly written things should ./configure;make;make install |
| 11:51.01 | rwhitby | jacques: so ipkg wouldn't install it? |
| 11:51.07 | jacques | rwhitby, correct in my case |
| 11:51.36 | jacques | ipkg installed maybe 1/4 of it and exited normally |
| 11:51.44 | LexDelani-work | anywy, quite happy with it already |
| 11:51.48 | rwhitby | yeah, that's what I got too |
| 11:52.56 | jacques | rwhitby, it has some issue with libs too - in perl Configure is doesn't find -lc or -lm and that comes back later to bite it |
| 11:53.26 | jacques | since perl Configure apparently doesn't log it's progress like real configure, that's gonna be a pain to track down |
| 11:53.35 | jacques | dammit - its not it's |
| 11:53.36 | rwhitby | yeah, that would be perl being naive about where it's looking for libs |
| 11:53.56 | jacques | rwhitby, absolutely |
| 11:54.38 | rwhitby | dropbear builds with crosstool-native, busybox fails trying to build zconf ... |
| 11:55.07 | jacques | also, hopefully related, even when I manually link miniperl with -lm it segfaults |
| 11:55.13 | rwhitby | (this is running make in the Unslung CVS on the slug, using crosstool-native as the toolchain |
| 11:55.28 | jacques | it would really save some time to see some of these unslung toolchains ppl had that were building perl and passing all the tests |
| 11:56.17 | *** join/#nslu2-linux Koops (~weiner@adsl-19-97.swiftdsl.com.au) |
| 11:56.25 | jacques | rwhitby, I halfway considered trying to build a toolchain on the slug using crosstool, but I suspect it would run out of memory in glibc |
| 11:56.55 | *** join/#nslu2-linux NSLU2-IRC (~NSLU2-IRC@adsl-19-97.swiftdsl.com.au) |
| 11:56.55 | rwhitby | what about just building the constituents one by one ... |
| 11:57.37 | jacques | if we want to build a toolchain natively, we need a standard way to do it - and crosstool already applies all the patches and builds in the right order with somewhat-correct args |
| 11:58.21 | rwhitby | I'll need to attach a bigger disk if I want to build crosstool :-) I'm doing all this on the cornice at the moment ... |
| 11:58.36 | jacques | yeah that won't quite do I suspect :-) |
| 11:58.49 | jacques | I have a 120GB 3.5" on my unslung slug now |
| 11:59.14 | bob_tm | FYI: Rewrote /etc/rc and /etc/rd.d/* to use explicit paths only (as far as I can see) - did full reinstall of BB 1.0 and it works |
| 11:59.42 | LexDelani-work | later all |
| 11:59.48 | jacques | later LexDelani-work |
| 11:59.49 | LexDelani-work | got to work ;) |
| 12:00.10 | rwhitby | I might have to buy a 2.5" enclosure to house the notebook disk which came out of the now-dead allwell stb3036N-CF home.rwhitby.net 24x7 linux server |
| 12:00.40 | jacques | rwhitby, well what cha waiting for then? :-P |
| 12:00.56 | rwhitby | jacques: the shop to open ... ;-) |
| 12:01.00 | jacques | oops I misread - I thought you said you already had the enclosure |
| 12:01.47 | jacques | what the heck - I feel like seeing how far I can get trying to build a toolchain manually (I used to do this *all the time* before crosstool for the ipaq in 2001) |
| 12:02.24 | rwhitby | just put a *big* swap on there, and let crosstool rip ... |
| 12:02.53 | jacques | well I have 128MB, which is what I use on the ipaq which also has 32MB RAM and I emerged gentoo on that :-D |
| 12:03.24 | jacques | tho I think the slug is not handling low memory condition quite as well as the ipaq - I will maybe investigate overcommit setting in proc |
| 12:03.25 | rwhitby | I still want our native toolchain to have the same configuration and patches as our cross-chain, to make reproducing bugs easier ... |
| 12:03.48 | jacques | yeah, maybe I'll try using crosstool to build a native toolchain first after all |
| 12:03.49 | rwhitby | so a natively built crosstool would be ideal |
| 12:04.24 | jacques | too bad we never got fatslug working yet :-( |
| 12:05.14 | rwhitby | We need ka6sox to build us a FatSlugFarm now!!! |
| 12:05.48 | jacques | yep! |
| 12:10.05 | rwhitby | So for ipkg-build to run on the slug, we either need to recompile /bin/sh, or to find a standalone getops (which I am not sure is possible), or to patch ipkg-build to use /opt/bin/bash if it's on the slug. |
| 12:10.21 | rwhitby | s/getops/getopts/ |
| 12:13.51 | jacques | on a devel system I generally install bash |
| 12:14.04 | rwhitby | as /bin/sh ? |
| 12:16.29 | jacques | well it links itself to /opt/bin/sh automatically |
| 12:17.42 | rwhitby | yeah, but I'm assuming that perhaps /bin/sh has mods done to it, so we can't just replace it at will |
| 12:18.12 | rwhitby | anyway, I'm outta here. |
| 12:18.14 | rwhitby | night all |
| 12:18.45 | jacques | `night rwhitby-asleep |
| 12:30.22 | bob_tm | Jacques: Do you know whether USB CD-ROMs should be accessible manually? |
| 12:31.15 | perlguru-work | plug it in and see what dmesg gives you and chech /proc |
| 12:31.45 | bob_tm | It is recognized in proc/scsi, but there is no /dev/sdcX devices |
| 12:32.22 | perlguru-work | may need the ide-scsi module then. |
| 12:32.40 | jacques | bob_tm, I dunno - would depend on if you were running unslung -able maybe |
| 12:33.16 | jacques | yeah the ide-scsi module is also a necessity |
| 12:33.20 | bob_tm | Does one still have to use mknod to created devices. |
| 12:33.21 | *** join/#nslu2-linux [g2] (~g2@rdu57-8-035.nc.rr.com) |
| 12:34.02 | [g2] | rwhitby-asleep, hmmmm early to bed, early to rise, makes a man happy, wealthy and wise :) |
| 12:34.07 | perlguru-work | bob_tm: how would you do it on a normal linux system? |
| 12:34.37 | bob_tm | On my Gentoo box it works out of the box - haven't needed to do anything manually with that for years |
| 12:35.10 | bob_tm | Hmm... /dev isn't persistent, is it? |
| 12:35.21 | rwhitby-asleep | [g2]: still here :-) |
| 12:35.39 | [g2] | quick question |
| 12:35.50 | [g2] | 3.17 tar.gz |
| 12:36.47 | [g2] | NOTE: Executing mkfs.jffs2 --root=/home/tom/dev/unslung/build/tmp/rootfs --faketime --output=/home/tom/dev/unslung/build/tmp/deploy/images/unslung-able-nslu2-20050111050757.rootfs.jffs2 --pad --big-endian --eraseblock=0x20000 -D |
| 12:36.47 | [g2] | ERROR: function image_cmd failed |
| 12:36.47 | [g2] | ERROR: see log in /home/tom/dev/unslung/build/tmp/log.image_cmd.8595 |
| 12:36.47 | [g2] | "~/dev/unslung/build/tmp/work/unslung-able-image-1.0-r3/temp/log.do_rootfs.14138" 131L, 6896C |
| 12:36.48 | rwhitby-asleep | bob_tm: use /dev.state |
| 12:37.02 | [g2] | the -D option is missing the device...txt file |
| 12:37.06 | bob_tm | thnx rwhitby-not-quite-asleep |
| 12:37.22 | rwhitby-asleep | [g2]: yep, looks like it |
| 12:37.46 | rwhitby-asleep | dunno why |
| 12:38.43 | [g2] | ok, I'll keep looking at the bb files |
| 12:38.55 | [g2] | sweet dreams! |
| 12:41.07 | jacques | rwhitby-asleep, since you;re not asleep yet - there might be a problem with the patch makefile |
| 12:41.23 | jacques | make patch-ipk tries to execute patch for some reason |
| 12:42.14 | [g2] | jacques, that should happen for all non-working patches :) and good morning |
| 12:42.30 | rwhitby-asleep | I just did a build of the Unslung CVS world, and it worked fine |
| 12:42.45 | rwhitby-asleep | are you talking cross or native? |
| 12:42.49 | jacques | ok dunno what's going on with that then. I have other things to worry about |
| 12:42.58 | jacques | i just scpd the binary to my slug anyway |
| 12:43.07 | jacques | i dont need no stinking ipks |
| 12:43.38 | rwhitby-asleep | too good for ipkg upgrade, hey? |
| 12:43.45 | jacques | i just did a cvs up patch.mk too so maybe something else was not updated |
| 12:44.04 | jacques | I'll try a full world build of latest late |
| 12:44.04 | jacques | r |
| 12:44.05 | rwhitby-asleep | or slug never in one configuration long enough for an ipkg upgrade to do anything ? |
| 12:44.14 | jacques | heh |
| 12:44.32 | jacques | BTW bash is pissing me off - it doesn't have rehash but apparently needs it |
| 12:44.51 | rwhitby-asleep | I'm slowing converging the Unslung CVS to be able to build both cross and native. |
| 12:45.05 | jacques | I had two wgets installed, and it was picking up the wrong one so I removed it and then it just kept saying /opt/bin/wget not found |
| 12:45.15 | jacques | even tho there was another wget in the path |
| 12:45.24 | jacques | in csh one does rehash in this case |
| 12:45.38 | jacques | rwhitby-asleep, extremely worthwhile goal that |
| 12:45.48 | jacques | dammit |
| 12:45.53 | jacques | now it wants bzip2 |
| 12:46.00 | jacques | crosstool is very demanding |
| 12:46.45 | jacques | at least we already have that ipkg :-) |
| 12:47.19 | jacques | by the time I'm done setting up my unslung devel env there probably wont be much of busybox left |
| 12:49.57 | [g2] | rwhitby-asleep, got another missing dir |
| 12:50.31 | rwhitby-asleep | are they missing from the tar file? |
| 12:50.43 | [g2] | yes there's a few |
| 12:51.00 | [g2] | I'm building the images now and I'll generate a diff |
| 12:51.27 | rwhitby-asleep | wow - I never actually expected anyone to be able to easily build from those. |
| 12:51.59 | [g2] | I was going to generate a tiny shell script that'll tweak the conf files and then ppl would be able to build assuming they had python, etc.... |
| 12:52.33 | [g2] | so basically, untar, path-adjust.sh, bb unslung-image |
| 12:52.50 | [g2] | so basically, untar, path-adjust.sh, . setup-.env, bb unslung-image |
| 12:52.59 | rwhitby-asleep | cool |
| 12:53.39 | rwhitby-asleep | well, my native build of bash just took about half an hour, so I'll let the rest run overnight and see how far it gets before it fills the 1,5Gb cornice. |
| 12:53.40 | [g2] | it might be nice for 4.x/3.18 to add the capability to build the native toolchain too |
| 12:54.09 | [g2] | from with OE or at least up to perl |
| 12:54.46 | rwhitby-asleep | [g2]: what do you mean? |
| 12:55.36 | rwhitby-asleep | we have crosstool.mk (which builds the cross toolchain) and crosstool-native.mk (which builds the bootstrap native toolchain) |
| 12:55.49 | [g2] | ok. |
| 12:55.57 | rwhitby-asleep | jacques is testing to see whether crosstool.mk can be built on the slug using the bootstrap native toolchain |
| 12:56.15 | jacques | rwhitby-asleep, so your native build of unslung packages is chugging along? no toolchain problems so far? |
| 12:56.16 | [g2] | great |
| 12:56.23 | rwhitby-asleep | I am upgrading the Unslung packages CVS repo to allow native building of unslung packages |
| 12:56.37 | rwhitby-asleep | jacques: "chugging" is a bit of an overstatement. |
| 12:57.56 | [g2] | NOTE: package unslung-image: build completed |
| 12:57.56 | [g2] | NOTE: build 200501110750 completed |
| 12:57.56 | [g2] | Build statistics: |
| 12:57.56 | [g2] | <PROTECTED> |
| 12:58.01 | rwhitby-asleep | dropbear, zlib, miau, termcap build |
| 12:58.25 | rwhitby-asleep | busybox doesn't get very far at all |
| 12:58.43 | rwhitby-asleep | bash builds, but the makefile fails due to the lack of "size" |
| 12:58.53 | jacques | size is in the toolchain |
| 12:59.09 | rwhitby-asleep | yeah - that's why I said the makefile fails |
| 12:59.44 | rwhitby-asleep | it just calls 'size' |
| 13:00.14 | jacques | I make symlinks |
| 13:00.17 | rwhitby-asleep | (I have been attempting to build Unslung packages without gcc, ld, etc in the path directly, as this is how it is done for cross) |
| 13:00.27 | jacques | oh |
| 13:01.27 | rwhitby-asleep | it helps spot bad cross-compilation assumptions in the makefiles |
| 13:01.32 | jacques | good point |
| 13:01.42 | jacques | i guess a bash makefile patch is in order |
| 13:01.51 | rwhitby-asleep | yeah |
| 13:01.58 | [g2] | gotta run, sweet dreams catch you guys later |
| 13:03.33 | rwhitby-asleep | but I'll deal with those later |
| 13:03.48 | jacques | sure. get some sleep. |
| 13:03.56 | rwhitby-asleep | just categorising packages at the moment. |
| 13:04.07 | rwhitby-asleep | those that work out of the box, and those that don;t |
| 13:04.25 | jacques | nod |
| 13:05.18 | rwhitby-asleep | I'd really like someone who know about these things to track down why fseeko exists but doesn't work on the slug ... |
| 13:05.56 | jacques | probably just standard brokenness and needs a glibc patch |
| 13:06.09 | jacques | heck it might already be fixed in later glibc |
| 13:06.21 | jacques | this *is* 2.2.5 after all |
| 13:06.51 | jacques | maybe sg or someone who still cares about 2.2.5 has a patch |
| 13:07.34 | rwhitby-asleep | good point |
| 13:40.34 | *** join/#nslu2-linux wipmac (DrFonk@J909d.j.pppool.de) |
| 13:40.57 | wipmac | hi. any solutions to the drive missing condition on 3.x? |
| 13:44.41 | bob_tm | You mean missing or unformatted? |
| 13:46.09 | wipmac | yep |
| 13:46.32 | bob_tm | If you mean "unformatted", go to /opt/sbin, delete insmod and reboot. |
| 13:47.04 | bob_tm | Please don't tell me you don't have BB 1.0 installed :-) |
| 13:47.27 | *** join/#nslu2-linux mrb51 (~blah@wa-pullman-cuda1k-244.losaca.adelphia.net) |
| 13:48.09 | bob_tm | wipmac: Please report success or failure. |
| 13:48.47 | wipmac | bb 1.0? |
| 13:49.00 | bob_tm | Busybox 1.0 |
| 13:49.24 | bob_tm | Installed manually using ipkg |
| 13:50.03 | wipmac | how can i check if i have it installed? :) |
| 13:50.18 | bob_tm | Well - if the file /opt/sbin/insmod is there, you do. |
| 13:50.21 | perlguru-work | ls -l /opt/bin/busybox |
| 13:50.29 | bob_tm | tnkx |
| 13:50.40 | wipmac | thx |
| 13:51.50 | perlguru-work | bob_tm: have you updated the http://www.nslu2-linux.org/wiki/HowTo/TroubleshootADriveNotFormattedCondition page (no) |
| 13:51.54 | wipmac | or rename insmod? :) |
| 13:52.05 | perlguru-work | rm /opt/sbin/insmod |
| 13:52.18 | bob_tm | Just delete it, you can alway reestablish the link later |
| 13:52.37 | bob_tm | Perlguru: Waiting for confirmation - wipmac is the perfect confirmer :-) |
| 13:52.55 | wipmac | rm: cannot remove `insmod': Read-only file system |
| 13:54.00 | perlguru-work | are you root? |
| 13:54.12 | wipmac | yep |
| 13:54.13 | bob_tm | Hmmm... RO ? |
| 13:54.38 | bob_tm | Can he unmount, run fsck and remount the conf partition? |
| 13:54.58 | *** join/#nslu2-linux VoodooZ (~VoodooZ@CPE00c0f021712c-CM014110003785.cpe.net.cable.rogers.com) |
| 13:55.05 | wipmac | hehe..would have been too easy just ermoving one file ;) |
| 13:55.35 | bob_tm | Well, seems to work for those with read-write /dev/sda2 |
| 13:55.54 | bob_tm | What does df say? |
| 13:56.28 | *** join/#nslu2-linux tjyang (~tj_yang@c-67-175-74-12.client.comcast.net) |
| 13:56.44 | bob_tm | - /dev/sda2 mounted on /share/hdd/conf? |
| 13:57.18 | giel | VoodooZ_work: i was wondering (as I still didn't have the opportunity to look at openslug) if openslug gets built with glibc or uclibc? |
| 13:57.26 | wipmac | bob: yes |
| 13:57.47 | wipmac | and /opt/sbin means /share/hdd/conf/opt/sbin |
| 13:57.55 | *** part/#nslu2-linux tjyang (~tj_yang@c-67-175-74-12.client.comcast.net) |
| 13:57.58 | jacques | giel, it can be built with either, default is glibc as there is a bug with ipkg and uclibc |
| 13:58.26 | giel | hm... okay... well i should stop asking questions until i actually have looked at it, but i'm so curious :) |
| 13:58.27 | bob_tm | "umount /dev/sda2 ; fsck.ext3 /dev/sda2; mount -t ext3 /dev/sda2 /share/hdd/conf" |
| 13:58.57 | bob_tm | Probably a reason why it is not mounted read-write |
| 13:59.15 | wipmac | WARNING!!! Running e2fsck on a mounted filesystem may cause |
| 13:59.15 | wipmac | SEVERE filesystem damage. |
| 13:59.15 | wipmac | Do you really want to continue (y/n)? yes |
| 13:59.25 | wipmac | fsck.ext3: Bad magic number in super-block while trying to re-open /dev/sda2 |
| 13:59.25 | wipmac | e2fsck: io manager magic bad! |
| 13:59.25 | wipmac | mount: Mounting /dev/sda2 on /share/hdd/conf failed: Device or resource busy |
| 13:59.26 | bob_tm | no - didn't umount work |
| 14:00.19 | bob_tm | ah - you have something running. Well. Shut down the slug, bring it up without drives and reconnect the drive - telnet in and retry. |
| 14:00.23 | perlguru-work | bob_tm: wipmac you may have processes running |
| 14:00.29 | bob_tm | Perlguru: Any better suggestion? |
| 14:00.32 | siddy | ;) |
| 14:00.38 | perlguru-work | goto maintenance mode |
| 14:00.56 | perlguru-work | telnet in, and run the check command |
| 14:02.57 | wipmac | ok, now i turn the hdd off, reboot, telnet into the box and type "check" |
| 14:04.10 | siddy | I would suggest: shutdown, unplug hdd, boot, plug in hdd, fsck /dev/sda2 |
| 14:04.38 | bob_tm | You probably have to unmount it, since USB_Detect will do its nasty stuff and mount it. |
| 14:04.44 | perlguru-work | Or as I said go in to maintenance mode: http://www.nslu2-linux.org/wiki/HowTo/TroubleshootADriveNotFormattedCondition |
| 14:04.53 | perlguru-work | *ggrrrr* |
| 14:05.55 | bob_tm | ? |
| 14:05.57 | siddy | you could also do an 'lsof' now and kill all processes, which access files on /share/hdd/conf..., then umount etc |
| 14:07.11 | perlguru-work | i have wipmac now in a private chat to help hoim |
| 14:07.26 | bob_tm | perlguru: The concept "Maintenence mode" is not explained on the wiki and not refered to on that page. It is mentioned in http://www.nslu2-linux.org/wiki/HowTo/InstallUnslungFirmware, but never explained. Where should we put an explanation? |
| 14:07.55 | bob_tm | s/Maintenence/Maintenance/ |
| 14:09.04 | perlguru-work | whatever. |
| 14:12.22 | bob_tm | perlguru: Added in InstallUnslungFirmware - ... "maintenance mode" (that is: boot without drives connected) - is that ok? |
| 14:13.28 | bob_tm | wipmac: What is your status? |
| 14:13.39 | jacques | hmm OK I'll swap you a cardboard box for your slug anyday. |
| 14:13.47 | wipmac | lol |
| 14:14.03 | wipmac | but hdont you believe the movie? |
| 14:14.24 | jacques | i think the device is a cardboard box with stickers taped to it |
| 14:15.04 | jacques | one of the screenshots shows a manual that says philbot on it |
| 14:16.04 | wipmac | bob: sec |
| 14:16.47 | jacques | see Photo-0013 |
| 14:18.31 | wipmac | jacques: which url? /dev/ram1 on / type ext2 (rw) |
| 14:18.31 | wipmac | /proc on /proc type proc (rw) |
| 14:18.31 | wipmac | usbdevfs on /proc/bus/usb type usbdevfs (rw) |
| 14:18.31 | wipmac | /dev/ram1 on /dev.state type ext2 (rw) |
| 14:18.31 | wipmac | ramfs on /dev type ramfs (rw) |
| 14:18.32 | wipmac | /dev/ram1 on /var.state type ext2 (rw) |
| 14:18.34 | wipmac | ramfs on /var type ramfs (rw) |
| 14:18.36 | wipmac | none on /tmp type ramfs (rw) |
| 14:18.38 | wipmac | oops |
| 14:18.40 | wipmac | sorry! |
| 14:24.31 | *** join/#nslu2-linux LarsAC (~chatzilla@dagama.lpt.RWTH-Aachen.DE) |
| 14:25.10 | giel | i think the iHome is just a stack of paper |
| 14:25.26 | giel | the ihome and the apple logo are printed on one sheet and taped to the top |
| 14:25.28 | giel | it's so fake |
| 14:26.09 | wipmac | can well be, you peeps are right. when is that speech today? |
| 14:27.20 | jacques | giel, Photo-0013 says philbot on it - how much more proof does anyone need? |
| 14:44.47 | giel | jacques: maybe slapped in the face with it :P |
| 14:45.04 | jacques | heh |
| 14:48.48 | wipmac | SORRY |
| 15:02.33 | wipmac | ..unslinging gain, no success |
| 15:03.40 | bob_tm | What have you done in the mean time? |
| 15:06.26 | wipmac | in the meantime i got rid of my fxxxxing conf |
| 15:07.53 | wipmac | ..accidentally somehow ;) |
| 15:07.57 | bob_tm | ok - so you wiped the conf-partition and reformatted it? |
| 15:08.10 | wipmac | nono |
| 15:08.20 | wipmac | i NEED my date.. not gonna reformat |
| 15:08.44 | bob_tm | Ok - did you do a "fsck.ext3 -n /dev/sda2"? |
| 15:09.02 | bob_tm | Data is on /dev/sda1 |
| 15:09.18 | pnk_ | anyone know whats up with the tinyproxy pkg .. |
| 15:09.34 | *** join/#nslu2-linux glc-work (~glc@www.thelabelservice.com) |
| 15:10.21 | wipmac | fsck.ext3 -n /dev/sda2 -> cleam |
| 15:10.33 | perlguru-work | we just did this |
| 15:10.51 | perlguru-work | 1st go wipmac had to repair a lot |
| 15:11.14 | wipmac | but /sbin/unsling tells me now "Cannot locate target disk" in the telnet window |
| 15:11.33 | bob_tm | Ah, ok - I'll stay out of it then. |
| 15:11.56 | perlguru-work | no bob_tm take over. i am going home now.... |
| 15:12.00 | wipmac | appreciate your help a lot ppl, dont get me wrong, am just a bit ..hmm..angry on my nslu2 ;) |
| 15:12.01 | perlguru-work | please |
| 15:12.07 | wipmac | lol |
| 15:12.07 | bob_tm | ok |
| 15:12.14 | wipmac | thx perl!! |
| 15:12.22 | perlguru-work | thanx, see you guys later in about an hour |
| 15:13.21 | giel | wipmac: stop being angry, it doesn't help :) |
| 15:13.47 | giel | don't see it as a problem, see it as a challenge |
| 15:13.52 | giel | gah, i can't believe i said that |
| 15:28.37 | bob_tm | oh my - are we gonna have Dr. Phil speak here? |
| 15:29.17 | giel | well... slugs are way too nice to get angry at :) |
| 15:29.34 | bob_tm | Well, wiped partitions are not. |
| 15:30.07 | LexDelani-work | ~emulate drphil |
| 15:31.21 | LexDelani-work | ~emulate drphil |
| 15:31.22 | jbot | it has been said that emulate drphil is Well, look at it from _their_ perspective... |
| 15:33.01 | *** join/#nslu2-linux mrblack51 (~blah@wa-pullman-cuda1k-244.losaca.adelphia.net) |
| 15:34.25 | jacques | jbot, emulate drphil is <reply> Well, look at it from _their_ perspective... |
| 15:34.26 | jbot | ...but emulate drphil is already something else... |
| 15:34.38 | jacques | jbot, no emulate drphil is <reply> Well, look at it from _their_ perspective... |
| 15:34.39 | jbot | okay, jacques |
| 15:34.48 | jacques | ~emulate drphil |
| 15:34.49 | jbot | Well, look at it from _their_ perspective... |
| 15:35.20 | LexDelani-work | ~celebrate |
| 15:35.21 | jbot | ACTION detonates a small nuclear device to celebrate the end of the world |
| 15:35.35 | giel | ~praise jacques |
| 15:35.37 | jbot | All hail jacques! |
| 15:37.18 | LexDelani-work | ~praise NSLU2 and Unslung |
| 15:37.22 | jbot | All hail NSLU2 and Unslung! |
| 15:43.22 | *** join/#nslu2-linux mrb51 (~blah@wa-pullman-cuda1k-244.losaca.adelphia.net) |
| 15:59.04 | wipmac | any1 having a conf-pattion and willing to send me the files? |
| 16:00.54 | Tiersten | wha? |
| 16:01.40 | wipmac | Conf-directory on the nslu2: /share/hdd/conf |
| 16:02.07 | *** join/#nslu2-linux Obelisk (russ@d47-69-99-106.try.wideopenwest.com) |
| 16:02.08 | Tiersten | It gets generated though. Why do you need a copy? |
| 16:03.34 | wipmac | becuase that dir is emoty on my slug! |
| 16:03.40 | wipmac | empty |
| 16:05.45 | bob_tm | I don't know if it works, but he has a slug with data on the data partition, but an empty conf partition. It is not recognized and I wonder if there is something that MUST be in the conf dir. |
| 16:06.18 | Tiersten | I thought it was already established that it did check? |
| 16:07.21 | bob_tm | It checked to lost+found. |
| 16:07.31 | bob_tm | Perlguru knows more. |
| 16:08.05 | bob_tm | But the only thing in conf is lost+found |
| 16:08.18 | Tiersten | Yeah but it must check for other files |
| 16:08.20 | bob_tm | With a nice inode list in it. |
| 16:08.26 | Tiersten | inode list in it?? |
| 16:08.42 | Tiersten | borked partition. restart... |
| 16:08.46 | bob_tm | Yes - you know - what you get after fsck'ing a messed up file system |
| 16:08.52 | bob_tm | Data partition is ok |
| 16:08.52 | Tiersten | yeah I know |
| 16:09.08 | Tiersten | I'm just saying that you should reformat the conf partition |
| 16:09.21 | Tiersten | Unless you want to dig through lost+found to work out what files are what |
| 16:09.27 | kergoth_ | bleh |
| 16:09.45 | Tiersten | Morning kergoth |
| 16:09.46 | siddy | wipmac: what the f*** did you do to trash a complete ext3-partition? Should be quite impossible to brake a journaling fs, you are my hero!! ;) |
| 16:09.47 | bob_tm | wipmac: Listen to Tiersten - I must run. Forget about the lost+found - nothing of value |
| 16:10.04 | wipmac | thx bob. cya! |
| 16:10.41 | wipmac | siddy: hold, fsck did that i guess :) |
| 16:10.53 | kergoth_ | morning |
| 16:11.00 | wipmac | tiersten: can i just reformat Conf-Partition without danger for my data? |
| 16:12.02 | Tiersten | assuming you only broke the conf partition yes |
| 16:12.06 | Tiersten | back up your data anyway |
| 16:12.54 | wipmac | wish i hadd space to backup the data.. is it really dangerous? i mean i just unslung before and data was still there |
| 16:13.20 | Tiersten | yeah but you managed to break the conf partition some how |
| 16:14.34 | wipmac | :) |
| 16:14.41 | wipmac | you *might* be right |
| 16:15.32 | Tiersten | it's always prudent to backup anyway |
| 16:20.16 | VoodooZ_work | OT: what would you use to burn a bin/cue CD image in linux? |
| 16:20.26 | wipmac | but lemme tellyou, the purpose of my nslu2 was initially to backup.. but then i started playing.. |
| 16:20.53 | siddy-away | VoodooZ_work: cdrdao |
| 16:21.22 | VoodooZ_work | thanks man. I'll give that a try. I'm tired of using windows for simple tasks. :) |
| 16:21.32 | siddy-away | np |
| 16:22.39 | Tiersten | How do you want these files then? |
| 16:22.56 | Tiersten | They're off my standard Linksys one however |
| 16:23.25 | VoodooZ_work | siddy-away: Would you have a quick example for me? the manpage doesn't mention anything obvious about it. |
| 16:24.12 | wipmac | tiersten you mean me? |
| 16:24.20 | Tiersten | Yes |
| 16:24.32 | Tiersten | you'll probably have to edit them as well since they've got my MAC address int it etc... |
| 16:25.05 | VoodooZ_work | siddy-away: I think I just found something. |
| 16:26.07 | wipmac | can u dcc? |
| 16:26.35 | Tiersten | No. Behind a firewall |
| 16:27.05 | wipmac | tiersten: email in /msg |
| 16:31.46 | *** join/#nslu2-linux KsiLaptop (russ@d47-69-99-106.try.wideopenwest.com) |
| 16:35.15 | *** join/#nslu2-linux perlguru (~maurice@cc444495-a.hnglo1.ov.home.nl) |
| 16:37.39 | wipmac | hot wo put files on the nslu2 via telnet? |
| 16:38.41 | *** join/#nslu2-linux kergoth (~kergoth@ip71.churchill-park.dfw.ygnition.net) |
| 16:43.07 | *** join/#nslu2-linux droemel (ident@p5081FB31.dip.t-dialin.net) |
| 16:43.28 | perlguru | wipmac, you cannot via telnet |
| 16:44.48 | perlguru | but wget is installed on the slug |
| 16:45.14 | perlguru | but, then you need a webserver on the host. |
| 16:46.36 | wipmac | perl: u have a webserver online? |
| 16:46.47 | perlguru | what do you need |
| 16:47.10 | wipmac | i have the content of the Conf-dir here. Tiersten suggested copying it to my conf dir |
| 16:47.35 | perlguru | so you need a conf.tar.gz ? |
| 16:48.10 | wipmac | i got one here from Tiersten |
| 16:48.24 | wipmac | and he told me to put it on the nslu2 .. |
| 16:49.48 | perlguru | dcc it to me |
| 16:50.22 | wipmac | coming? |
| 16:50.46 | perlguru | status connect |
| 16:50.59 | wipmac | email? |
| 16:51.19 | perlguru | ok it is not big |
| 16:51.25 | wipmac | icq |
| 16:51.28 | wipmac | ftp |
| 16:51.29 | wipmac | whatever |
| 16:51.44 | perlguru | you have my email |
| 16:53.38 | wipmac | sent. 1kb :) |
| 16:53.59 | *** join/#nslu2-linux NSLU2-IRC (~NSLU2-IRC@69-164-178-43.pittpa.adelphia.net) |
| 16:59.35 | *** join/#nslu2-linux NSLU4me (~knoppix@69-164-178-43.pittpa.adelphia.net) |
| 16:59.58 | wipmac | perl: got it? |
| 17:01.09 | perlguru | wipmac, got it, but setting it up. please stand by |
| 17:02.09 | wipmac | THANK YOU |
| 17:02.50 | perlguru | wget this on the slug |
| 17:03.20 | *** join/#nslu2-linux NSLU4me (~knoppix@69-164-178-43.pittpa.adelphia.net) |
| 17:06.42 | wipmac | tiersten, perl: content of the conf.tar.gz on the nslu2..reboot? |
| 17:08.19 | wipmac | ok reboot :) |
| 17:09.10 | *** join/#nslu2-linux tking_ (~tking@adsl-66-159-198-116.dslextreme.com) |
| 17:11.11 | wipmac | no change after reboott |
| 17:16.53 | Tiersten | What does the harddisk section in /etc/CGI_ds.conf say |
| 17:18.55 | *** join/#nslu2-linux wipmac (DrFonk@J909d.j.pppool.de) |
| 17:19.00 | Tiersten | I don't care about the rest of it |
| 17:19.04 | wipmac | Tiersten: where shoul i post that? |
| 17:19.05 | Tiersten | JUST THE HARDDISK SECTION |
| 17:19.06 | Tiersten | jeez |
| 17:19.13 | wipmac | sorry |
| 17:19.45 | Tiersten | There should be a validhd and a invalidhd line in that section |
| 17:19.47 | Tiersten | What do they say |
| 17:20.19 | wipmac | [harddisk] |
| 17:20.20 | wipmac | validhd=0:0 |
| 17:20.20 | wipmac | invalidhd=1:0 |
| 17:21.13 | Tiersten | Change it to to "validhd=1:0" and "invalidhd=0:0" |
| 17:21.20 | Tiersten | Change it to "validhd=1:0" and "invalidhd=0:0" |
| 17:23.44 | *** join/#nslu2-linux glc-work (~glc@www.thelabelservice.com) |
| 17:25.40 | Tiersten | Morning glc |
| 17:25.51 | wipmac | reboot tiersten? |
| 17:26.13 | Tiersten | Reboot properly yes. Don't just yank the power |
| 17:26.26 | wipmac | entering"reboot" in telnet |
| 17:27.55 | wipmac | no change |
| 17:28.29 | Tiersten | Don't know then. You sure it's not the problem that other people have been having regarding disks being detected as being unformatted? |
| 17:29.31 | wipmac | Tiersten: it should be that one i guess.. |
| 17:29.41 | wipmac | you thought it was something different? |
| 17:30.05 | Tiersten | No. I'm just asking if it's the same problem |
| 17:30.12 | wipmac | yes it is |
| 17:30.23 | wipmac | and then conf got lost during trying to rstore things |
| 17:34.28 | wipmac | thx for help ppl .. nogo.. have to buy a new hd and reformat probably the whole thing... thx again and cya |
| 17:35.08 | *** join/#nslu2-linux kergoth_ (~kergoth@proxyle01.ext.ti.com) |
| 17:38.08 | *** join/#nslu2-linux NSLU4me (~knoppix@69-164-178-43.pittpa.adelphia.net) [NETSPLIT VICTIM] |
| 17:38.08 | *** join/#nslu2-linux bob_tm (~bob@254.80-203-34.nextgentel.com) [NETSPLIT VICTIM] |
| 17:38.08 | *** join/#nslu2-linux cbrake (~cbrake@mail.accelent.com) [NETSPLIT VICTIM] |
| 17:45.20 | *** join/#nslu2-linux NSLU2-IRC (~NSLU2-IRC@c-efdb71d5.02-85-73746f13.cust.bredbandsbolaget.se) |
| 17:46.48 | RickardDahlstran | I need help with a compile error I get when I try to compile iperf on the nslu2 |
| 17:47.37 | RickardDahlstran | When i run configure I get a question "What type is the 3rd argument of accept a pointer to? [int]" What should i answer? |
| 17:47.58 | Tiersten | It's probably int |
| 17:48.09 | Tiersten | Give me a minute and I'll check |
| 17:48.41 | RickardDahlstran | Then I get "Socket.cpp: In member function `int Socket::Accept()':" "Socket.cpp:195: error: invalid conversion from `int*' to `socklen_t*'" |
| 17:48.59 | Tiersten | ah. in that case it's socklen_t :) |
| 17:49.35 | Tiersten | yeah. sorry, it's socklen_t |
| 17:49.42 | Tiersten | Just checked the headers |
| 17:50.43 | Tiersten | Well. BYE THEN! Rude people... |
| 17:51.12 | *** join/#nslu2-linux NSLU2-IRC (~NSLU2-IRC@c-efdb71d5.02-85-73746f13.cust.bredbandsbolaget.se) |
| 17:51.30 | Tiersten | It work? |
| 17:51.50 | jacques | seems to be having connection issues |
| 17:52.20 | Tiersten | Possibly. He's using the Java IRC client and most people forget that you can't navigate away from that page without losing the connection |
| 17:52.20 | *** join/#nslu2-linux NSLU2-IRC (~NSLU2-IRC@c-efdb71d5.02-85-73746f13.cust.bredbandsbolaget.se) |
| 17:52.40 | Tiersten | RickardDAhlstran: It work? |
| 17:52.57 | RickardDAhlstran | Sorry, backspace sends me back one page, Will get a real irc-client. =) |
| 17:53.02 | jacques | :-) |
| 17:53.13 | RickardDAhlstran | Still compiling, should it really take this long?? |
| 17:53.36 | Tiersten | Yeah. The NSLU2 is quite slow compared to a recent PC |
| 17:53.44 | jacques | lol |
| 17:53.45 | Tiersten | Not much RAM as well so it has to swap for anything big |
| 17:54.34 | jacques | dont talk to me about long - I'm building a toolchain on me slug right now |
| 17:55.10 | kergoth_ | sick bastard |
| 17:55.12 | Tiersten | Uh huh... masochist... ;) |
| 17:55.16 | RickardDAhlstran | OK, have been using some embedded systems, but I have never compiled anything. =) |
| 17:55.33 | jacques | kergoth_, they said something similar when I emerged gentoo on my 32MB ipaq |
| 17:55.38 | kergoth_ | yes |
| 17:56.13 | jacques | I just don't look forward to the locales :-( |
| 17:56.27 | Tiersten | jacques: The question that comes to mind. Is WHY?!?! |
| 17:56.39 | Tiersten | you really like staring at a compile screen or something? :) |
| 17:56.54 | RickardDAhlstran | Now I get "/share/hdd/data/tools/gcc-3.3.4-glibc-2.2.5/lib/gcc-lib/armv5b-softfloat-linux/3.3.4/include/limits.h:11:23: syslimits.h: No such file or directory" |
| 17:56.58 | kergoth_ | if you're really that bored, just have your desktop do a bitbake world |
| 17:57.20 | jacques | Tiersten, we don;t have a good way of making an unslung toolchain right now |
| 17:57.38 | Tiersten | No. I'm referring to you doing emerge on an iPaq |
| 17:57.49 | Tiersten | desktop? jacques would probably do it on his NSLU2 |
| 17:57.51 | jacques | Tiersten, ah heh. it only took three days |
| 17:58.05 | Tiersten | I might have a 386 somewhere if you want to do bitbake world... |
| 17:58.25 | jacques | RickardDAhlstran, touch /share/hdd/data/tools/gcc-3.3.4-glibc-2.2.5/lib/gcc-lib/armv5b-softfloat-linux/3.3.4/include/syslimits.h |
| 18:07.58 | *** join/#nslu2-linux RickardDahlstran (~root@puh.tilde.se) |
| 18:09.29 | *** join/#nslu2-linux RickardDahlstran (~rd@puh.tilde.se) |
| 18:10.12 | RickardDahlstran | Ok, back with a real irc-client. |
| 18:10.45 | jacques | cool |
| 18:11.01 | RickardDahlstran | The compile worked and iperf starts, but the performance is not the best. It's probably the cpu that is the limit. |
| 18:11.11 | RickardDahlstran | bash-2.05b# ./iperf -c 10.1.1.1 -r |
| 18:11.11 | RickardDahlstran | ------------------------------------------------------------ |
| 18:11.11 | RickardDahlstran | Server listening on TCP port 5001 |
| 18:11.11 | RickardDahlstran | TCP window size: 42.7 KByte (default) |
| 18:11.11 | RickardDahlstran | ------------------------------------------------------------ |
| 18:11.13 | RickardDahlstran | Client connecting to 10.1.1.1, TCP port 5001 |
| 18:11.15 | RickardDahlstran | TCP window size: 16.0 KByte (default) |
| 18:11.17 | RickardDahlstran | ------------------------------------------------------------ |
| 18:11.19 | RickardDahlstran | [ 6] local 10.1.1.80 port 1036 connected with 10.1.1.1 port 5001 |
| 18:11.21 | RickardDahlstran | [ ID] Interval Transfer Bandwidth |
| 18:11.23 | RickardDahlstran | [ 6] 0.0-10.0 sec 977 KBytes 800 Kbits/sec |
| 18:11.25 | RickardDahlstran | [ 7] local 10.1.1.80 port 5001 connected with 10.1.1.1 port 41737 |
| 18:11.27 | RickardDahlstran | [ 7] 0.0- 9.9 sec 977 KBytes 809 Kbits/sec |
| 18:12.35 | Tiersten | The way the NPEs work is a bit clunky |
| 18:12.48 | Tiersten | I don't think it's that amazingly fast. Especially since there are so many interrupts |
| 18:13.42 | RickardDahlstran | Thanks for all the help, what you are doing with this device is truly amazing. |
| 18:16.46 | RickardDahlstran | I wish I could help with anything, but I'm really more in to BSD coding and don't know that much about Linux. Sorry. |
| 18:40.46 | *** join/#nslu2-linux RickardDahlstran (~rd@puh.tilde.se) |
| 18:41.46 | RickardDahlstran | Just another question. How do I start inetd during startup without modifing the rc-script? |
| 18:44.52 | bob_tm | The standard inetd _is_ started on boot automatically. |
| 18:45.07 | bob_tm | ./etc/rc.d/rc.xinet |
| 18:45.12 | RickardDahlstran | Ok, just figured out that inetd isn't started from rc. =) |
| 18:45.39 | RickardDahlstran | But my inetd still doesn't start when I start /etc/rc.d/rc.xinetd |
| 18:46.07 | bob_tm | Check /var/log/messages |
| 18:46.24 | bob_tm | It should - maybe you've done something interestin in /etc/inetd.conf |
| 18:47.10 | bob_tm | Note that this inetd uses non-standard format on /etc/inetd.conf |
| 18:47.31 | bob_tm | argv[0] is _not_ to be specified. |
| 18:47.53 | RickardDahlstran | If i type inetd & it starts allright, but if I type /etc/rc.d/rc.xinetd it doesn't |
| 18:50.13 | bob_tm | Hmmm... "which inetd" |
| 18:50.40 | RickardDahlstran | \/bin/inetd |
| 18:51.01 | bob_tm | Only way this could happen I can think of is due to the presence of /unslung/rc.xinetd |
| 18:51.28 | bob_tm | If you look at /etc/rc.d/rc.xinet it is not rocket science. |
| 18:52.56 | RickardDahlstran | Aha, I missed the return 0 part. Ok, so if I add a rc.xinetd in one of the unslung dirs it overides the startup as well. |
| 18:53.08 | bob_tm | Yes |
| 18:53.56 | RickardDahlstran | The howto at http://www.nslu2-linux.org/wiki/Unslung/Vsftpd then should be updated, don't you think?+ |
| 18:54.14 | bob_tm | In what way? |
| 18:55.08 | RickardDahlstran | When I create the /unslung/rc.xinetd I stop inetd from starting. |
| 18:55.34 | bob_tm | Depends on the return value - returning 1 makes the rest of the original run. |
| 18:56.36 | bob_tm | Personally I find the inetd.conf part a bit of a kludge. |
| 18:57.14 | bob_tm | ... but that is the prize you pay for being and extension of Linksys. And there will come a time when you will be happy for that :-) |
| 18:57.37 | RickardDahlstran | Does "echo "ftp stream tcp nowait root /opt/sbin/vsftpd /opt/etc/vsftpd.conf" >>/etc/inetd.conf" return 1 |
| 18:59.17 | bob_tm | You are so right. |
| 18:59.34 | bob_tm | I will edit it. |
| 18:59.48 | RickardDahlstran | Also, if I runt /etc/rc.d/rc.xinetd a couple of times during the same boot I will add multiple lines to inetd.conf and that maked inetd put out error. |
| 19:02.08 | bob_tm | That is true: As I said, I never liked that thing. To be honest I don't even use the standard inetd (*blush*) |
| 19:03.04 | RickardDahlstran | Is there anything else that will add things to /etc/inetd.conf?? Can I use a cp /opt/etc/inetd.conf /etc/ ; return 1 ? |
| 19:03.33 | bob_tm | Mom - I will take a look at my setup now after upgrading to 3.17. |
| 19:04.51 | bob_tm | Since /etc is persistent in 3.16/17, I think it is only the Linksys software that will alter it (maybe when enabling telnet) |
| 19:07.14 | bob_tm | So just drop the /unslung/rc.xinetd and edit /etc/inetd.conf manually until the package maintainer have come up with a better solution. Check the "slug bug database" for entries |
| 19:07.35 | bob_tm | ... about this. |
| 19:08.44 | RickardDahlstran | I just tried enabling telnet and it restarted inetd but it didn't update the /etc/inetd.conf file. And telnet is not working. |
| 19:09.42 | bob_tm | That is with the disks inserted? |
| 19:10.31 | RickardDahlstran | yes. |
| 19:11.23 | bob_tm | On my machine enabling telnet appends the telnetd line in /etc/inetd.conf (I have 3.17). I will check telnet |
| 19:11.34 | bob_tm | It works |
| 19:11.39 | bob_tm | I will try to disable. |
| 19:11.42 | RickardDahlstran | It seems to replace /etc/inetd.conf with /opt/etc/inetd.conf when I enable it. |
| 19:12.45 | bob_tm | The two are similar on my pc, but there is an extra newline |
| 19:13.31 | bob_tm | When I disable it again, the telnet line is removed from /etc/inetd.conf, but it remains in /opt/etc/inetd.conf |
| 19:13.47 | bob_tm | ..and telnet stops working |
| 19:13.58 | bob_tm | You have 3.17? |
| 19:14.07 | RickardDahlstran | Yes 3.17. |
| 19:14.56 | bob_tm | For me it works nicely with vsftp inserted /etc/inetd.conf, but note: I have no diversion scripts. |
| 19:15.16 | bob_tm | What other scripts do you have in /unslung |
| 19:15.41 | RickardDahlstran | When I removed rc.xinetd it started working again. |
| 19:15.50 | bob_tm | The /unslung one? |
| 19:15.54 | RickardDahlstran | Yes |
| 19:16.01 | bob_tm | Ok - then you have the same setup as I |
| 19:18.35 | bob_tm | Happy? If so, can you edit the wiki page to reflect your setup - I have some work to do. |
| 19:19.07 | bob_tm | Please state that your setup is 3.17 based as the 2.x things still need to append form diversion on boot. |
| 19:21.06 | RickardDahlstran | It seems like /etc/rc.d/rc.xinetd is run every time I enable or disable telnet and since I copied a new inetd.conf every time that meand that the telnet line got overwritten. |
| 19:21.42 | RickardDahlstran | I use 3.17 and 2.3R25 |
| 19:22.18 | RickardDahlstran | We need to find a way to add stuff to /etc/inetd.conf while still having telnet working. Any ideas? |
| 19:23.34 | *** join/#nslu2-linux [g2] (~g2@66.57.8.35) |
| 19:24.30 | jacques | re [g2] |
| 19:26.05 | [g2] | hey jacques |
| 19:26.15 | [g2] | what's up ? |
| 19:26.31 | jacques | trying to build toolchain still |
| 19:26.35 | jacques | it's still going |
| 19:26.49 | [g2] | THX for all your help with that! |
| 19:27.08 | jacques | hmm? |
| 19:27.42 | [g2] | it's great that you've been working on that and you are still plugging away |
| 19:28.08 | jacques | ah yeah well it takes a certain kind of OCD to do these things |
| 19:28.23 | [g2] | ~ocd |
| 19:28.39 | [g2] | ~3c |
| 19:28.58 | jacques | jbot, OCD is Obsessive Compulsive Disorder |
| 19:28.59 | jbot | jacques: okay |
| 19:29.20 | [g2] | ~3C |
| 19:29.48 | bob_tm | Rickard - U there? |
| 19:29.59 | [g2] | well thx anyway :) |
| 19:30.13 | bob_tm | Don't copy every time - just one time. |
| 19:31.08 | [g2] | does this look normal for 3.17 unslung-able ? |
| 19:31.11 | [g2] | # df |
| 19:31.11 | [g2] | Filesystem 1k-blocks Used Available Use% Mounted on |
| 19:31.11 | [g2] | rootfs 6400 5184 1216 81% / |
| 19:31.11 | [g2] | /dev/root 6400 5184 1216 81% / |
| 19:31.11 | [g2] | /dev/root 6400 5184 1216 81% /dev.state |
| 19:31.12 | [g2] | /dev/root 6400 5184 1216 81% /var.state |
| 19:31.28 | bob_tm | No |
| 19:31.34 | jacques | looks like mine |
| 19:31.48 | [g2] | it's booted without any disks attached |
| 19:32.00 | jacques | even with a disk it will look like that |
| 19:32.10 | jacques | since now it's booting to internal flash jffs2 |
| 19:32.35 | [g2] | well I could see 1 or 2 mounts but 4 seems a little OCD :) |
| 19:32.49 | bob_tm | Mine has /dev/mtdblock4 as device instead of /dev/root |
| 19:33.10 | jacques | [g2], we should als an unslung developer :-) |
| 19:33.12 | bob_tm | But that is probably not important. |
| 19:33.14 | jacques | ask |
| 19:34.00 | [g2] | jacques, nod. I'm just checking because I built this from the tarball + some additions on SF |
| 19:34.18 | bob_tm | Is /dev/root major 31, minor 4 ? |
| 19:34.23 | RickardDahlstran | Bob, I have a couple of lines that will do the trick. Should I update wiki?? |
| 19:34.31 | bob_tm | Yes please. |
| 19:34.41 | bob_tm | Again, please mark version 3.X |
| 19:35.52 | RickardDahlstran | what do you mean? Where should I add it?? |
| 19:36.35 | RickardDahlstran | This is basicly what I did: |
| 19:36.37 | RickardDahlstran | #!/bin/sh |
| 19:36.37 | RickardDahlstran | if ( !(grep vsftp /etc/inetd.conf -q) ) then |
| 19:36.37 | RickardDahlstran | <PROTECTED> |
| 19:36.37 | RickardDahlstran | fi |
| 19:36.37 | RickardDahlstran | return 1 |
| 19:36.42 | bob_tm | Oh - please update the wiki page, but note that your fix is relative to Unslung version 3.x |
| 19:36.57 | bob_tm | :-) |
| 19:37.29 | bob_tm | *happy, happy* *joy, joy* |
| 19:37.40 | RickardDahlstran | Should I put it at the end, as a comment. |
| 19:37.49 | *** join/#nslu2-linux glc-work (~glc@www.thelabelservice.com) |
| 19:39.14 | bob_tm | Please put it in troubleshooting or as a note where the author instructs the append solution. Your solution should work on all versions! |
| 19:40.39 | bob_tm | It is better than his (or maybe hers). |
| 19:42.04 | *** join/#nslu2-linux LexDelani (~lex@cc571156-a.hooge1.gr.home.nl) |
| 19:44.31 | siddy-away | VoodooZ_work: Did you succeed with cdrdao? |
| 19:48.06 | RickardDahlstran | Ok, I added it under troubleshooting. Thanks for all the help. |
| 19:48.18 | bob_tm | Thank you. |
| 19:53.46 | VoodooZ_work | siddy: I didn't get a chance to try but it looks fairly straightforward now that I've read the man page. THanks. |
| 20:14.10 | *** join/#nslu2-linux ka6sox (~tking@netblock-66-159-209-52.dslextreme.com) |
| 20:22.14 | jacques | re ka6sox |
| 20:29.03 | *** join/#nslu2-linux PTweety (~Miranda@dialin-145-254-198-187.arcor-ip.net) |
| 20:29.44 | *** part/#nslu2-linux PTweety (~Miranda@dialin-145-254-198-187.arcor-ip.net) |
| 20:30.18 | *** join/#nslu2-linux PTweety (~Miranda@dialin-145-254-198-187.arcor-ip.net) |
| 20:31.59 | *** join/#nslu2-linux LarsAC (~chatzilla@pD9500723.dip0.t-ipconnect.de) |
| 20:46.50 | VoodooZ_work | damn, I still get the no root found bug with latest openslug. |
| 20:47.45 | VoodooZ_work | Strange, it does mount the jffs2 (/dev/mtdblock4) partition to / though. which root is it talking about??? |
| 20:48.08 | VoodooZ_work | I think someone might have broken the switchbox somehow or played with linuxrc. |
| 20:49.06 | [g2] | VoodooZ_work, can you post the messages to the wiki under OpenSlug ? |
| 20:49.22 | VoodooZ_work | There's not much more than a single line really: |
| 20:49.39 | VoodooZ_work | Root filesystem cannot be found - dropping into shell ... |
| 20:49.39 | VoodooZ_work | tar: Invalid gzip magic |
| 20:50.02 | VoodooZ_work | It did that after initialising all the hardware. |
| 20:50.29 | VoodooZ_work | Where do you want me to put that exactly? |
| 20:50.54 | [g2] | anywhere under OpenSlug is fine |
| 20:51.08 | *** join/#nslu2-linux [te] (~Miranda@dialin-145-254-198-187.arcor-ip.net) |
| 20:55.30 | VoodooZ_work | damn. how to you prevent the formatting again? |
| 20:56.21 | VoodooZ_work | I want to post the whole log but it looks like crap. |
| 20:56.42 | ka6sox | [= =] |
| 20:56.55 | ka6sox | means literal |
| 20:58.10 | VoodooZ_work | I did that but it still looks like a single line. |
| 20:58.23 | VoodooZ_work | Check it out at NoRootFoundErrata |
| 20:58.32 | VoodooZ_work | under Openslug |
| 20:58.45 | VoodooZ_work | unless I can't copy and paste? |
| 21:04.49 | jacques | only thing that works for me is to put a space at beginning of every line I dont want formatted |
| 21:05.10 | VoodooZ_work | wow. tedious. I'll do that for now. |
| 21:05.47 | jacques | someone told me a way to do it in emacs but I don't recall it ATM |
| 21:06.04 | [g2] | s/^/ / |
| 21:06.08 | ka6sox | voodooZ: use [= in the front and =] at the end of the section |
| 21:06.22 | ka6sox | oh okay..I understand |
| 21:06.36 | ka6sox | use either emacs or vi to fix it globally |
| 21:07.15 | VoodooZ_work | nah. I just did the space trick. all done. |
| 21:07.47 | jacques | ka6sox the [= =] thing has never worked for me |
| 21:07.59 | ka6sox | okay |
| 21:08.27 | jacques | does it work for you? |
| 21:09.03 | jacques | maybe I was doing it wrong when I tried |
| 21:11.10 | jacques | man building glibc takes forever |
| 21:12.12 | VoodooZ_work | I know! the longest thing in the world! Are you building openslug too? |
| 21:12.37 | *** join/#nslu2-linux Ximinez (~kandre@115.80-202-23.nextgentel.com) |
| 21:12.38 | [g2] | VoodooZ_work, so you're in the shell right now ? |
| 21:13.03 | VoodooZ_work | yep |
| 21:13.03 | VoodooZ_work | mount gives me: /dev/root.old on / type ext2 (rw) |
| 21:13.06 | [g2] | you're in switchbox |
| 21:13.13 | VoodooZ_work | No idea where I am. |
| 21:13.17 | [g2] | ls / |
| 21:13.21 | [g2] | cat linuxrc |
| 21:13.38 | VoodooZ_work | mount: Mounting /proc on /proc failed: Device or resource busy |
| 21:13.42 | VoodooZ_work | then the same as before |
| 21:14.04 | [g2] | ummmm |
| 21:14.16 | VoodooZ_work | Unless my switchbox is outdated somehow? |
| 21:14.29 | VoodooZ_work | Is it recompiled by default? |
| 21:14.42 | [g2] | I take that back you've booted and loaded modules or are you modules static ? |
| 21:14.54 | VoodooZ_work | most of them are static. |
| 21:15.07 | VoodooZ_work | Have you checked my wiki page? |
| 21:15.32 | [g2] | the new one sure.... I see the insmod errors |
| 21:15.57 | VoodooZ_work | yeah, but more importantly: why do we get there. |
| 21:16.06 | [g2] | cat you cd / |
| 21:16.11 | [g2] | can you cd / |
| 21:16.18 | VoodooZ_work | yep. |
| 21:16.21 | [g2] | cat linuxrc |
| 21:16.23 | VoodooZ_work | but I'm already there |
| 21:16.28 | VoodooZ_work | ... |
| 21:16.41 | VoodooZ_work | you want to see the whole hting? |
| 21:16.54 | [g2] | can you paste it in the wiki ? |
| 21:17.01 | VoodooZ_work | ok.. |
| 21:19.10 | VoodooZ_work | done |
| 21:20.41 | VoodooZ_work | [g2]: Thanks for looking at this. |
| 21:22.32 | jacques | VoodooZ_work, right now I'm just building the toolchain... on the slug :-) |
| 21:22.57 | VoodooZ_work | toolchain? really? |
| 21:23.15 | jacques | trying to. using crosstool |
| 21:23.30 | [g2] | VoodooZ_work, it looks like it's trying to load the ixp drivers |
| 21:23.33 | jacques | it's been on glibc for hours |
| 21:23.34 | VoodooZ_work | you looking for something to do or something? :) |
| 21:23.44 | jacques | sigh why does everyone say that? |
| 21:23.45 | jacques | :-) |
| 21:23.50 | VoodooZ_work | hehehe |
| 21:23.57 | jacques | self-hosting is an important milestone |
| 21:24.01 | VoodooZ_work | [g2]: before it crash you mean? |
| 21:24.16 | jacques | at least rwhitby-away agrees with me on that :-P |
| 21:24.32 | [g2] | well it doesn't really "crash" it exists to the shell |
| 21:25.09 | VoodooZ_work | damn. I'll have to change that later as I really don't want tons of useless crap (for me) running at startup. |
| 21:25.41 | VoodooZ_work | I noticed it pauses for a few second at that stage. |
| 21:26.07 | [g2] | yes, I'm thinking that switchbox is broken for OpenSlug right now |
| 21:26.27 | jacques | I'm trying to remember - I think it was dyoung who worked on it a lot during christmas |
| 21:26.35 | jacques | (I wasn't around I just heard about it) |
| 21:26.37 | VoodooZ_work | Do you know who made the last change? I didn't see anything in bk changes. |
| 21:27.13 | [g2] | I haven't pulled OpenSlug in a while, I've been tweaking my development box a bunch |
| 21:27.17 | jacques | unfortunately he's not here to defend himself :-) |
| 21:27.22 | VoodooZ_work | :) |
| 21:27.44 | jacques | yeah me too - I haven't updated the flash on my openslug box in a while - I've been chrooting into my openslug devel env |
| 21:28.03 | jacques | that I have been updating but not too often as I have my own changes to worry about |
| 21:28.41 | [g2] | however, the good news that I like my setup and I've been building the Unslung 3.17 from the source tarball |
| 21:28.44 | [g2] | on sf |
| 21:28.45 | jacques | changes which I need to get into shape to commit but haven't because they are kinda intrusive - doing it right requires changing openslug.conf etc |
| 21:29.44 | VoodooZ_work | rwhitby made switchbox changes on the 26th. 1.2667.39.1 and 1.2667.1.60 |
| 21:29.56 | [g2] | so I'll try to finish testing the Unslung 3.17 build stuff today, and then we can do the same and have a OpenSlug Tarball |
| 21:30.33 | [g2] | on the bright side you could probably nfs boot :) |
| 21:30.38 | VoodooZ_work | Any suggestions of stuff I can try in the meantime? |
| 21:31.08 | jacques | VoodooZ_work, the tricky thing is, even tho switchbox has been updated, you would have to look at the openslug stuff in bb to see which version of switchbox it's building into the image (since it uses a binary version) |
| 21:31.09 | VoodooZ_work | yeah, probably but that's just overhead for me. I want the fastest boot possible. no frills. |
| 21:31.30 | VoodooZ_work | yeah. I was wondering about that. |
| 21:31.48 | VoodooZ_work | I'm confused as to how it fits in with BB and all. |
| 21:31.57 | [g2] | We should decide what we want to do for OpenSlug |
| 21:32.11 | jacques | when bb builds the final image, it uses a binary switchbox from the web |
| 21:32.18 | VoodooZ_work | hopefully I won't be forced with something big and slow though. |
| 21:32.38 | [g2] | VoodooZ_work, it's OpenSlug :) |
| 21:32.53 | VoodooZ_work | yeah. I guess. |
| 21:32.55 | jacques | there's nothing stopping you from building your own switchbox tho |
| 21:33.11 | VoodooZ_work | yeah, eventually I guess. |
| 21:33.19 | [g2] | I've been thinking about removing switchbox from OpenSlug |
| 21:33.21 | jacques | you'd just have to use slugimage and assemble the image manually |
| 21:33.37 | VoodooZ_work | sounds like an idea. |
| 21:33.49 | VoodooZ_work | I could probably save a few seconds that way too. |
| 21:34.10 | jacques | i think for most openslug users it's worth it to have switchbox |
| 21:34.20 | [g2] | why ? |
| 21:34.28 | VoodooZ_work | yeah, I don't see switchbox as necessary for openslug. Openslug is for developers not script-kiddies. |
| 21:34.59 | VoodooZ_work | Unless you can have OE toggle it as an option? |
| 21:35.00 | jacques | it adds flexibility in the boot - where you boot to |
| 21:35.28 | [g2] | I guess, for OpenSlug, booting to JFFS with the kernel in a big partition makes sense |
| 21:35.32 | jacques | you saying no opensluggers will want to boot to network or directly to disk? |
| 21:35.50 | ka6sox | VoodooZ_work, do we just ignore those who are unable to build a full development system and relagate them to Unslung? |
| 21:35.50 | VoodooZ_work | yeah, but it's not really a desktop. I see openslug for people like me who have a specific app in mind. unslung covers the rest. |
| 21:35.52 | jacques | personally I want to boot directly to disk at some point |
| 21:36.27 | VoodooZ_work | disk? As in usb hard disk? perhaps. |
| 21:36.33 | jacques | I see openslug for ppl who want to use their slug as a desktop |
| 21:36.34 | [g2] | however, there's real reason we can pivot_root on the hd right ? |
| 21:36.44 | jacques | so we are at odds |
| 21:36.45 | [g2] | or boot directly there |
| 21:37.08 | VoodooZ_work | I guess I'm ok with whatever you guys decide as long it's not adding more than 1 second to my boot and it work. |
| 21:37.36 | ka6sox | how about those who want to be able to do things that unslung is not invisioned for...like a special app |
| 21:37.42 | jacques | there should be an easy way for you to bypass switchbox I agree |
| 21:37.53 | jacques | but I don't want the same things you do |
| 21:38.07 | jacques | somehow we have to reach a compromise |
| 21:38.10 | ka6sox | since OE gives us the ability to have thousands of apps why must we be developers? |
| 21:38.18 | VoodooZ_work | openslug is supposed to be flexbile so it's hard to decide. |
| 21:38.39 | VoodooZ_work | like I said. If it works I'm happy. |
| 21:39.04 | [g2] | jacques, I don't think we're at odds |
| 21:39.11 | VoodooZ_work | I admit my application is "special" so I'll go with whatever you decide. |
| 21:39.37 | [g2] | I'd like to fully understand what ppl are doing and come up with the best solution(s) |
| 21:39.48 | jacques | I'm not at all happy with how long my openslug boot takes currently. but I don't think getting rid of switchbox is the answer for me. |
| 21:40.00 | jacques | most of my boot time is still in hotplug |
| 21:40.01 | VoodooZ_work | I agree. |
| 21:40.05 | [g2] | nod. |
| 21:40.11 | VoodooZ_work | yeah. That still pisses me off. |
| 21:40.24 | VoodooZ_work | Not an issue for now but it will be for my competition. |
| 21:40.33 | [g2] | that needs to be fixed, but I'd label it as Major |
| 21:40.34 | ka6sox | I look at OpenSlug is a GP distro. not as specific as Unslung and certianly not as specific as Stock. |
| 21:40.58 | jacques | ka6sox yeah I think of it as the "use your slug as a GP computer" distro |
| 21:41.20 | ka6sox | exactly...and as such it takes longer to boot as it has more "options" |
| 21:41.21 | [g2] | nod. I think we're all in agreement |
| 21:41.45 | ka6sox | unless you compile a special kernel/boot combo then it will boot faster |
| 21:42.19 | jacques | we could make a dozen different dists - all specialized, but we would have developer dilution |
| 21:42.30 | VoodooZ_work | yeah but it's still not a full pc though. |
| 21:42.56 | [g2] | jacques, right and we need to get OpenSlug beta out first |
| 21:42.58 | ka6sox | no video is a big stumbling block. |
| 21:43.06 | ka6sox | for a gp distro |
| 21:43.08 | [g2] | a headless GP |
| 21:43.08 | jacques | so the tricky thing is only making as many dists as we can actually develop with the resources we have while keeping as many ppl as possible happy |
| 21:43.23 | [g2] | jacques, fully agreed |
| 21:43.46 | VoodooZ_log | well, that's normal. It's an embedded board. |
| 21:44.13 | VoodooZ_log | anyways. I have to go guys. Stupid dentist appointment. talk to you later. Thanks for your help. |
| 21:44.21 | jacques | later VoodooZ_log |
| 21:44.27 | [g2] | see you |
| 21:45.21 | [g2] | So right now I'm working on a working Unslung 3.17 from source |
| 21:45.32 | jacques | i gotta go get a haircut |
| 21:45.41 | ka6sox | cya jacques |
| 21:45.55 | jacques | if I'm lucky my crosstool build will still be going when I get back (as in not errored out) |
| 21:45.58 | [g2] | see you, will you be around later ? |
| 21:46.30 | jacques | oh yeah - haircut sholdn't take long - tho I may go out after that to some shops |
| 21:46.42 | jacques | I'll check back in before I do that anyway |
| 21:47.09 | [g2] | I'd like to get consensus on a OpenSlug Beta plan and you're a Key man |
| 21:49.43 | jacques | yeah I think we need to get a beta out soon |
| 21:50.18 | *** join/#nslu2-linux pTweety (~private_t@dialin-212-144-092-249.arcor-ip.net) |
| 21:51.03 | *** join/#nslu2-linux glc-work (~glc@www.thelabelservice.com) |
| 22:03.02 | rwhitby-away | what do people think getting rid of SwitchBox will gain them? |
| 22:03.38 | [g2] | rwhitby-away, well for VoodooZ_log a working OpenSlug :) .... just kidding |
| 22:03.51 | rwhitby-away | are you sure about that? |
| 22:04.17 | [g2] | he may be using an old baseline |
| 22:04.27 | [g2] | I was looking at the kernel command in the log |
| 22:04.32 | rwhitby-away | OpenSlug is using SwitchBox 3.8 - dyoung has nothing to do with that - he has been working on SwitchBox 4.x in a separate repo |
| 22:05.04 | [g2] | I haven't rebuild OpenSlug from the repo in a couple weeks |
| 22:05.19 | *** join/#nslu2-linux pTweety (~private_t@dialin-212-144-092-249.arcor-ip.net) |
| 22:05.40 | [g2] | I've built Unslung 3.17 from the source tarball and I'm just about done with creating a working version |
| 22:06.36 | [g2] | there's a minor issue to iron out, but I'm creating the patch script now and after testing a virgin rebuild we can think about releasing it |
| 22:08.04 | [g2] | I'd really like to release the same setup for OpenSlug when we go Beta |
| 22:08.23 | rwhitby-away | cool - it should be distro independent |
| 22:08.46 | [g2] | well the package files are distro dependent |
| 22:08.59 | [g2] | but the concept is really distro independent |
| 22:09.22 | rwhitby-away | I mean the procedure for packaging and releasing a BB/OE based source tar file |
| 22:09.36 | rwhitby-away | I'm agreeing that it is :-) |
| 22:09.39 | [g2] | nod. |
| 22:10.26 | [g2] | it may be easy to automate the release too |
| 22:10.56 | [g2] | but first lets get Unslung 3.17 going and well move from there |
| 22:11.45 | *** join/#nslu2-linux siddy (~sid@p50927687.dip.t-dialin.net) |
| 22:12.15 | rwhitby-away | [g2]: you need to look at Switchbox 4.x to make sure it will work for OpenSlug too |
| 22:13.23 | [g2] | rwhitby-away, thx, we need to have a OpenSlug stakeholders meeting about what the community wantes |
| 22:13.27 | [g2] | wants |
| 22:16.54 | rwhitby-away | yep |
| 22:17.10 | giel | i want beer! |
| 22:17.11 | rwhitby-away | you know what I want, right? |
| 22:17.15 | giel | (excuse my french) |
| 22:18.21 | [g2] | rwhitby-away, WANTS IT ALL :) |
| 22:18.45 | giel | i want the time to play with openslug |
| 22:18.55 | giel | but this job of mine is so mentally demanding |
| 22:19.05 | *** join/#nslu2-linux droemel (ident@p5081FB31.dip.t-dialin.net) |
| 22:33.45 | *** join/#nslu2-linux pTweety (~private_t@dialin-212-144-092-249.arcor-ip.net) |
| 22:36.30 | jacques | back |
| 22:36.41 | [g2] | wb, clean-cut guy |
| 22:36.46 | jacques | heh |
| 22:36.59 | jacques | i do look better now. not necessarily good, but better :-) |
| 22:37.35 | jacques | hmm, what *does* rwhitby-away want? |
| 22:38.03 | jacques | man, my crosstool build is *still* going |
| 22:38.13 | *** join/#nslu2-linux glc (~glc@user-119ah2f.biz.mindspring.com) |
| 22:38.23 | jacques | which is better than the alternative I suppose |
| 22:40.13 | [g2] | rwhitby-away, Ok I've got the Unslung 3.17 additions all ready |
| 22:40.28 | [g2] | there's a build running right now |
| 22:41.45 | [g2] | basically it's untar Unslung...3.17...., untar sf_extras, . fix_paths; . setup-env; bitbake unslung-image |
| 22:42.58 | [g2] | it assumes you have sed and pwd and the stuff needed to run bitbake etc... |
| 22:45.31 | *** join/#nslu2-linux nikki_chumakov (~nikki@clusterfw.beelinegprs.net) |
| 22:45.58 | rwhitby-away | glc: you need to subscribe to that group too |
| 22:46.04 | rwhitby-away | hey Nikki |
| 22:46.15 | nikki_chumakov | hi! |
| 22:46.22 | rwhitby-away | (Nikki is an awesome wrt54g developer) |
| 22:47.30 | rwhitby-away | jacques: for openslug, I want it to read sysconf settings, and boot onto my network after flashing |
| 22:48.34 | nikki_chumakov | rod, I'm looking docs on compiling unslung-3. could you point me to it? |
| 22:48.58 | rwhitby-away | the wiki page you reference is the latest |
| 22:49.36 | [g2] | greets nikki! |
| 22:50.50 | rwhitby-away | http://www.nslu2-linux.org/wiki/OpenEmbedded/GettingStarted |
| 22:51.26 | [g2] | are you the nikki that was working with Kaloz on the 2.6 port ? |
| 22:51.47 | nikki_chumakov | hmm... is there any docs on how to setup DEBIAN_MIRROR, GNU_MIRROR, HANDHELDS_CVS? |
| 22:51.57 | rwhitby-away | local.conf.sample is there |
| 22:52.20 | rwhitby-away | dunno - maybe on hh.org wiki |
| 22:55.25 | nikki_chumakov | I downloaded Unslung-3.17-beta-source.tar.gz from sourceforge, there is no local.conf.sample |
| 22:56.24 | [g2] | nikki_chumakov, I've got an update with all you need build with that stuff |
| 22:56.35 | nikki_chumakov | [g2]: never worked on 2.6 port. |
| 22:56.44 | [g2] | nod. |
| 22:57.52 | [g2] | nikki_chumakov, you're looking to build Unslung 3.17 right ? |
| 22:58.02 | nikki_chumakov | yes |
| 22:58.26 | [g2] | do you want me to dcc/email/post this tarball ? |
| 22:58.49 | [g2] | you can be the first tester after me ? :) |
| 22:59.01 | giel | ok, it's late, i'm off |
| 22:59.06 | giel | bye y'all |
| 22:59.16 | nikki_chumakov | yes, please. |
| 22:59.40 | [g2] | let me just gzip it and I can dcc you |
| 22:59.48 | nikki_chumakov | could you put it on some ftp or http where i can take it? |
| 23:00.17 | [g2] | can you pull from cvs ? |
| 23:00.31 | nikki_chumakov | yes. |
| 23:00.47 | [g2] | I can add something on the cvs server |
| 23:03.05 | nikki_chumakov | OK |
| 23:06.00 | [g2] | rwhitby-away, got any preference on a temporary place for the extra tarball for Unslung 3.17 until we've tested a little more and can post to the world ? |
| 23:15.34 | rwhitby-away | nslu.sf.net/downloads |
| 23:16.11 | rwhitby-away | nikki_chumakov: we normally build directly from bitbake. no-one ever builds from the source tarball - [g2] is just getting that working now |
| 23:17.26 | [g2] | nikki_chumakov, ka6sox is going to post it |
| 23:17.41 | [g2] | I didn't do a readme yet so.... |
| 23:18.02 | [g2] | It's untar the Unslung-3.17-beta-source.tar.gz from sf |
| 23:18.17 | [g2] | then dl and untar the Unslung-3.17-beta-source-extra.tar.gz |
| 23:18.29 | [g2] | a1529eb4d4f424d8f5f718f7a997891e Unslung-3.17-beta-source-extra.tar.gz |
| 23:18.36 | [g2] | is the MD5sum |
| 23:19.09 | [g2] | then . fix-paths |
| 23:19.12 | *** join/#nslu2-linux NSLU2-IRC (NSLU2WikiI@host217-44-237-4.range217-44.btcentralplus.com) |
| 23:19.32 | [g2] | which updates the path info in two conf files to be where you are at |
| 23:19.37 | [g2] | . setup-env |
| 23:19.57 | [g2] | which setups some OE/BitBake vars and puts you in the build directory |
| 23:20.06 | [g2] | and bitbake unslung-image |
| 23:20.25 | [g2] | then wait a couple hours based on how fast your computer/network connection is |
| 23:24.26 | nikki_chumakov | rwhitby-away: did not get it: did you build directly from bitbake or bitkeeper? |
| 23:25.43 | rwhitby-away | we build according to that GettingStarted page. |
| 23:26.02 | rwhitby-away | bitbake is the tool used to build. bitkeeper is where the repo is stored |
| 23:26.20 | jacques | they both start with bit but aren't related |
| 23:26.39 | rwhitby-away | openembedded is the name of the metadata repository that is our parent repo |
| 23:26.48 | rwhitby-away | we have nslu2-specific packages in openembedded. |
| 23:27.06 | rwhitby-away | openembedded uses the bitbake tool for building |
| 23:27.09 | nikki_chumakov | [g2]: I'm going to compile staff remotely on fairly fast comp/network. hope it will end in less than 2 hours. |
| 23:27.54 | rwhitby-away | we dump the source tree on sf.net at release time, but no-one every tries building from it (and it doesn't include the local.conf - something which [g2] is rectifying) |
| 23:28.02 | rwhitby-away | s/every/ever/ |
| 23:28.42 | rwhitby-away | hey memelalauk1985 |
| 23:29.25 | nikki_chumakov | rwhitby-away: I understand it in general. Unfortunatly I have no knowlegde of bitbake/bitkeeper in details. |
| 23:29.57 | rwhitby-away | memelalauk1985: if you run 2.x or 3.x, you can put a .ext3flash in /share/hdd/conf/ and then unslung will do it's best to reduce writes on that disk |
| 23:30.22 | rwhitby-away | we folded that functionality into the standard release using the /.ext3flash flag file. |
| 23:30.42 | jacques | i have no idea what crosstool is doing. it seems to have build glibc like four times now |
| 23:30.57 | rwhitby-away | jacques: at least it is still going .... |
| 23:30.58 | jacques | and now it's configuring libiberty yet again |
| 23:31.18 | rwhitby-away | my native Unslung packages build failed running awk |
| 23:31.18 | jacques | rwhitby-away, yeah, but I suspect it may be confused and looping (tho I have no idea how that would happen) |
| 23:31.30 | rwhitby-away | I doubt that crosstool.sh can loop |
| 23:31.41 | jacques | rwhitby-away, bb awk? try gawk |
| 23:32.25 | jacques | well I think one thing confusing it is that there are binaries in the path with the exact same names as the "crosstool" binaries it's making |
| 23:32.27 | rwhitby-away | yeah, I remembered that and am doing that now |
| 23:32.46 | rwhitby-away | jacques: ah - you didn't change prefix to something else? |
| 23:33.06 | jacques | no I just started it up. I never expected it to get this far |
| 23:33.23 | jacques | oh prefix |
| 23:33.36 | rwhitby-away | I figured that the natively built native toolchain would just install directly under /opt ... |
| 23:34.03 | jacques | well that's not the issue - I dunno if it changes its own path when it rebuilds things with the gcc it builds |
| 23:34.08 | rwhitby-away | memelalauk1985: did you get that? |
| 23:34.30 | jacques | the gcc it builds has the exact same name as the native toolchain - thus the potential for confusion |
| 23:35.15 | jacques | I had to install flex. a couple of configures complain about missing bison, but it hasn't stopped it yet |
| 23:36.51 | jacques | it really looks like it's starting over |
| 23:37.08 | rwhitby-away | jacques: right, but neither is in the path, so if you give a different PREFIX then it should be fine |
| 23:38.01 | jacques | it had been building everything with arm-linux-softfloat-gcc and now it's back to building libiberty with gcc |
| 23:38.20 | jacques | sigh. wtf. |
| 23:39.45 | jacques | also, I need to find out if crosstool knows how to build a native toolchain - the right way - that's building at least stage 1 and 2 compilers |
| 23:40.10 | jacques | it doesn't do that with cross chains (for obvious reasons) |
| 23:40.33 | rwhitby-away | ah - yeah. |
| 23:40.36 | memelalauk1985 | hey rwhitby. got it :) thanks for that, il give it a go when im back at work with the nslu2. |
| 23:40.44 | memelalauk1985 | (need one @ home too) |
| 23:41.02 | rwhitby-away | you're using it at work, or it just happens to be at work? |
| 23:43.02 | jacques | lol it's been going > 9 hours so far |
| 23:43.44 | jacques | and looks like it has far to go - looks like maybe it build binutils, gcc-core, glibc, and now is building the "real" gcc with g++ :-\ |
| 23:43.57 | jacques | geez I hope it doesn't want to build glibc yet again after that |
| 23:44.55 | jacques | took 6 hours to build glibc alone |
| 23:46.48 | jacques | I'd like to continue using crosstool to build the toolchain as it handles the patching well, but it depends how deep the cross assumtions run |
| 23:47.04 | memelalauk1985 | manager bought a couple, he likes buying new gadgets. 1 spare, and found this site, so thought id have a go. |
| 23:47.24 | [g2] | jacques, you can use crosstool to build a native compiler |
| 23:47.26 | rwhitby-away | jacques: sequence is: binutils, then glibc-headers, then gcc-core (including libiberty), then glibc and linuxthreads, then final gcc, and that's it. |
| 23:47.45 | nikki_chumakov | [g2]: Unslung-3.17-beta-source-extra.tar.gz is still absent on nslu.sf.net/dl. Not that I especially hurry, but... :-) |
| 23:47.48 | *** join/#nslu2-linux tjyang-away (~tj_yang@c-67-175-74-12.client.comcast.net) |
| 23:48.01 | jacques | rwhitby-away, thanks. that makes me feel somewhat better |
| 23:48.13 | rwhitby-away | nikki_chumakov: is there a reason you don't want to build from the live repo? |
| 23:48.16 | [g2] | nikki_chumakov, ka6sox was going to post it to a web site |
| 23:48.22 | jacques | [g2], have you looked into it? is that documented somewhere? |
| 23:48.27 | *** part/#nslu2-linux tjyang-away (~tj_yang@c-67-175-74-12.client.comcast.net) |
| 23:48.39 | rwhitby-away | he came, he slapped, he left ... |
| 23:48.39 | jacques | because I'd really like it to build multi-stage gcc |
| 23:48.50 | [g2] | jacques, I got instructions from Lennart a couple months ago |
| 23:48.56 | [g2] | iirc |
| 23:49.04 | jacques | [g2], whooo! find and email please! |
| 23:49.23 | [g2] | that could take a little digging |
| 23:49.54 | jacques | but I neeeeeed them |
| 23:50.04 | [g2] | I cleaned up 50GB last month, I think I've got it but it may be offline |
| 23:50.40 | jacques | doing google search. |
| 23:50.47 | jacques | "Did you mean: krysstal native toolchain " lol |
| 23:50.50 | rwhitby-away | jacques: the gawk ipk seems to be broken - it links awk->gawk in /opt/bin but gawk don't exist, just gawk-3.1.4 |
| 23:51.03 | jacques | rwhitby-away, oops yeah I forgot to mention that |
| 23:51.05 | [g2] | don't bother, just ask Lennart |
| 23:51.10 | [g2] | I'll check |
| 23:51.14 | jacques | rwhitby-away, other than that it appears to work fine |
| 23:51.32 | [g2] | and crosstool does distcc too |
| 23:51.54 | nikki_chumakov | rwhitby-away: no real reson. I don't know bitkeeper, thus don't like it, but I can handle it. |
| 23:52.00 | jacques | "Your search - krysstal native toolchain - did not match any documents. " rofl |
| 23:52.04 | [g2] | I'd like go get crosstool running deja-gnu and running the checks on the slug |
| 23:52.37 | jacques | [g2], well... |
| 23:52.53 | jacques | in my experience, one does not really want to know all the toolchain tests that fail |
| 23:53.03 | jacques | one ends up wondering how anything works at all |
| 23:53.45 | jacques | I'm not *that* much of a masochist :-) |
| 23:54.21 | jacques | google found some good looking stuff |
| 23:54.34 | jacques | krysstal notwithstanding |
| 23:55.07 | [g2] | nikki_chumakov, do you have dev access to sf ? |
| 23:55.44 | nikki_chumakov | g2: no |
| 23:55.45 | jacques | [g2], does Lennart frequent irc? |
| 23:56.06 | [g2] | very rarely, but sometimes, he's on the yahoo groups much more often |
| 23:56.15 | [g2] | nikki_chumakov, yahoo groups access ? |
| 23:56.21 | nikki_chumakov | yes |
| 23:56.23 | jacques | hmm I've never posted to yahoo IIRC |
| 23:56.32 | jacques | and now it's like a tradition |
| 23:57.59 | *** join/#nslu2-linux bob_tm (~bob@254.80-203-34.nextgentel.com) |
| 23:58.09 | *** part/#nslu2-linux bob_tm (~bob@254.80-203-34.nextgentel.com) |
