irclog2html for #nslu2-linux on 20050603

00:21.43*** join/#nslu2-linux [g2] (~g2@g2.nslu2-linux)
01:03.01CIA-803jp30 * 10unslung/ (7 files in 2 dirs): upstream upgrade php to 5.0.4
01:28.39[g2]jp30 ping
01:46.31jp30-workhi, [g2]
01:47.09jp30-workguess what, unslung exposed a bug in php
01:47.55CIA-803jp30 * 10unslung/ (make/ sources/php/zend_strtod.patch): workaround for zend_strtod bug
01:50.54jacquesjp30-work, :-)
01:53.31jp30-workthe php guys had forgotten about big endian arm. hah!
01:58.06jp30-work[g2], i must have missed you - may catch you later from home
02:03.37jacqueshow dare they!  :-)
02:10.39[g2]jacques, did you connect the VCC for the JTAG, is it required ?
02:15.53jacques[g2], not on the usb one
02:16.23jacquesbut yes on the MAX based one
02:16.31jacquessorry was in other room
02:16.42jacquessorry misread
02:16.47jacquesthought you were talking about serial
02:16.52[g2]no JTAG
02:17.21jacquesI connected my JTAG just as it has on the wiki - I dunno if it's requireed, but I got VCC from bottom of board where serial port is (and GND)
02:17.43jacqueshmm I guess that's not exactly on the wiki
02:17.50jacquesbut that's what I did and it worked :-)
02:18.04[g2]hmmmm maybe I need to connect that over
02:18.12[g2]you used a digilent cable right ?
02:19.33jacquesyep, the 2.8V one
02:19.50[g2]yeah... that's the one I've got
02:20.01[g2]I think all the pins are connected but the VCC
02:20.24[g2]I could jumper that over from the serial line
02:25.02jacquesyep, that's what I did
02:25.30[g2]OK thx
02:37.54[g2]jacques, did you add the 10K pulldown ?
02:39.44[g2]Hey Tiersten
02:46.54jacques[g2], nope, I added no parts
03:21.09*** join/#nslu2-linux dyoung-web (
03:21.16dyoung-web[g2] ping
03:21.41[g2]dyoung-web, morning :)
03:22.28[g2]dyoung-web, when you hooked up the digilent, did you use the 10K pull-down ?
03:22.32dyoung-webI need to couble check, but I think the vcc is required, because it provides the vccio voltage refernce for the chip inside the JTAG cable.
03:22.51dyoung-webso for a slug, use 3.3V.
03:23.00dyoung-webits all documented on my JTAG pictures page.
03:23.06dyoung-webI didnt use any additional parts.
03:23.24[g2]maybe I just need to hookup VCC better
03:24.24[g2]dyoung-web, I've got Good News / Bad News on the FAT slug
03:25.04dyoung-webjust checked, the NL17SZ125 is powered by the pin1 VCC, so you need to connect it.
03:25.41dyoung-webor is it pin6.... well you get the idea.
03:26.07dyoung-webso whats the good news?
03:26.42[g2]serial, 64MB, APEX work GREAT
03:26.49dyoung-webokay, whats the bad news?
03:27.17[g2]The rework looks good to me, but it's 2X or 3X the cost I had hoped for :(
03:27.37dyoung-weboh, like $50   ?
03:27.54dyoung-webstupid cgiirc.
03:27.54[g2]Plus :( and that's just the labor
03:28.56dyoung-webMakes sense to me, minimum .5hr charge right?
03:29.29dyoung-webmaybe if you do 20 of them the per unit price will be less?
03:29.37[g2]well I'll be following up but right now it looks like it takes 2 hours
03:29.49[g2]I got 10, 25, and 50 piece pricing
03:29.59dyoung-web2hrs per unit?!
03:30.07dyoung-webwhat does that include?
03:30.29[g2]serial, JTAG, and memory upgrade
03:30.40jacqueshow well do you know these ppl?
03:30.56dyoung-webI'm rather suprised at the time estimate.
03:31.00[g2]brand new, just met them
03:31.11jacquesmethinks they may be taking you for a ride
03:31.29[g2]I doubt that
03:31.40[g2]the did the first unit for free :)
03:31.56jacqueswhat takes so long?
03:31.58dyoung-webIts free the first time.
03:32.05jacquesare they stacking chips?
03:32.13[g2]no replace chips
03:32.17[g2]no replaced chips
03:32.21jacquesthat's faster
03:32.23dyoung-webafter that, its $60 for a bag. ;-)
03:32.25[g2]8M16s with 16M16s
03:32.32jacqueswell, I say ask around
03:32.41jacquessee what other ppl think about that time
03:32.47dyoung-webAsk ka6sox how long it took him to replace the chips.
03:32.51jacquesppl who should know
03:32.56dyoung-webhe just did one for me.
03:33.05jacqueswell he's not a professional board reworker is he?
03:33.11jacques(or is he?)
03:33.29dyoung-webI dunno, he has a rework station.
03:33.38dyoung-webbut he may not be an expert at it.
03:33.51jacquesfor ppl that rework boards all day long, I think 0.5 hours sounds about right
03:33.53[g2]ka6sox has always been very busy for me
03:34.21[g2]the SLUG I sent him in Jan/Feb is still AWOL
03:34.29dyoung-webI think 30min is not out of line for the chip replacement.
03:34.50[g2]I was thinking in the .5-1 hour range also
03:34.59[g2]but hell I'm SW guy
03:35.14[g2]it takes me like 3 hours to do the 4-pin header
03:35.16dyoung-webWell, someone was teaching how to stack.
03:35.39dyoung-webwith a rework station removing the old parts should take around 30 seconds.
03:35.59[g2]then you should clean the holes
03:36.00dyoung-webpasting /heating the new one on shouldnt take more than 15min each.
03:36.06dyoung-webwhat holes?
03:36.21[g2]then inspection
03:37.13dyoung-webokay 10minutes for inspection.  so maybe an hour.
03:38.45dyoung-webI think if I were spending that much, I'd spend a little more and get it with 128MB.
03:39.10jacquesat those prices I'd do the serial and jtag myself
03:39.11[g2]I think if we're spending that much we should roll our own boards
03:39.45[g2]I think the avila boards would be cheaper
03:40.13dyoung-webaround the same price right?
03:40.20dyoung-webexcept the avila has a minipci
03:40.33dyoung-webinsetad of USB2
03:40.36[g2]and no usb2
03:41.14[g2]to me $150 is a hard limit for developers
03:41.59[g2]I mean I spend $400 for a AMD64, Mobo and 512MB of 400Mhz DDR
03:42.14[g2]with a fast disk that does nearly 400Mbs
03:42.55[g2]especially with the new C7's and all the stuff around the corner
03:47.29dyoung-webwow, was just reading about it.
03:47.37dyoung-webthe low power mode isnt too bad.
03:48.27[g2]my take is that this market is going to be limited on the upper-end by mini/nano/... -itx and even the PC market
03:49.03dyoung-weband power.
03:49.21dyoung-webunless someone starts making low power PC perpheral devices
03:49.24[g2]so except for custom applications or fanless low-power it's gonna get squeezed
03:50.00dyoung-webSo, The Man wins again.
03:52.33[g2]dyoung-web, so what's your fatslug look like memory config wise ?
03:52.50[g2]are you running stacked 8M16 with an extra CS ?
03:54.24dyoung-webI just sent him 16M16's since stacking didnt seem to work so well.
03:54.52dyoung-webso it will be the single bank of 16M16.
04:02.36[g2]hey dyoung-web and jacques can you run the following for me ?
04:02.40[g2]dd if=/dev/zero of=/dev/null bs=8k count=125000
04:02.46dyoung-webon what target?
04:03.01[g2]125000+0 records in
04:03.01[g2]125000+0 records out
04:03.01[g2]real    0m 16.07s
04:03.01[g2]user    0m 0.23s
04:03.01[g2]sys     0m 15.85s
04:03.38[g2]it runs about 2.5 seconds faster on the avila
04:04.29dyoung-webUnslung: real 5.007s, 3.330s, sys=5.730s
04:04.52dyoung-web4.730 even
04:04.54[g2]Huh ?
04:05.11dyoung-webtime dd if=/dev/zero of=/dev/null bs=8k count=125000
04:05.12dyoung-web125000 0 records in
04:05.12dyoung-web125000 0 records out
04:05.12dyoung-webreal    0m5.007s
04:05.12dyoung-webuser    0m0.330s
04:05.14dyoung-websys     0m4.730s
04:05.31[g2]you've got the bs=8k in there right ?
04:05.38dyoung-websee above.
04:07.23[g2]Sounds like something isn't setup right
04:07.43dyoung-webwheres your rootfs?
04:08.08dyoung-webhard drive for me.  wonder if that makes a differnce.
04:08.29[g2]try swapoff
04:09.17dyoung-webThe wiley does this:
04:09.19dyoung-webreal    0m 14.72s
04:09.19dyoung-webuser    0m 0.68s
04:09.19dyoung-websys     0m 14.04s
04:10.10[g2]P4 does this :)
04:10.13[g2]real    0m0.267s
04:10.13[g2]user    0m0.047s
04:10.13[g2]sys     0m0.219s
04:10.49[g2]I think the amd64 will be faster too
04:13.06dyoung-webafter turn off swap I get similar results.
04:13.38[g2]hm... Interesting
04:13.52dyoung-webI dont have a openslug box to try this on right now.
04:13.59dyoung-webnot one I can get to anways.
04:16.15*** join/#nslu2-linux rwhitby-web (
04:16.31dyoung-webrwhitby-web: Can you re-forward me that zd1211 stuff please?
04:16.35dyoung-webI never received it.
04:16.49dyoung-webbut you should recieve a serial cable in around 5 days.
04:20.35*** join/#nslu2-linux [g2]-amd64 (
04:21.12[g2]-amd64real    0m0.191s
04:21.12[g2]-amd64user    0m0.032s
04:21.12[g2]-amd64sys     0m0.156s
04:23.02rwhitby-webdyoung: msg me the email address to use
04:23.22rwhitby-webdyoung-web: how much do I owe you?  msg me the paypal address too
04:28.08rwhitby-webdyoung-web: resent to your address
04:28.18rwhitby-web(previous one was to
04:39.34*** join/#nslu2-linux jp30 (~josh@jp30.nslu2-linux)
04:40.09jp30[g2], ping?
04:40.17[g2]jp30, pong
04:40.29jp30you wanted to talk to me earlier?
04:40.41[g2]I was wondering about running some apache/thttp/php benchmarks on a 64MB slug
04:41.28jp30hmm... apache, mysql, and php could probably benefit from more real ram
04:42.10jp30the only testing i've down on them has been pretty informal though
04:44.02jp30what would you like to know? i can suggest some applications that might benefit from the extra ram if that would help
04:44.39[g2]well you've worked with it a bunch more than have
04:45.18[g2]just chatting about an approach and usnlung versus openslug testing I think would be a good start
04:46.16dyoung-webAh, okay I havnt check the gmail for a while..
04:46.32[g2]It's a big area IMHO, maybe we could start a wiki page and ppl could contribute ideas and test scenarios ?
04:46.48jp30seems like a good idea.
04:47.15[g2]I'd also like to benchmark the avila IXP422 with 64MB
04:47.47jp30one thing i've found is that mediawiki's mmcache is useless because of lack on memory on unslung / slug. would be interesting to know if it would work better on a fat slug
04:48.20[g2]nod. I'd think we'd want to start with some basic disk benchmarks
04:49.02[g2]surely flashstick, cornice drive, 7200 RPM drives are going to perform differently
04:49.04dyoung-webWas there interseted in a fatslugged unslung?
04:49.44[g2]dyoung-web, a funny thing happened with my fatslug testing today
04:50.01dyoung-webIt sucked too much power and the brick cooked?
04:50.22[g2]I booted with APEX and the kernel wasn't passed a memory parameter and it knew to use 64MB
04:50.28jp30[g2] agree it would be nice to know about disk vs flash stick performance. slug as diskless webserver is a great application
04:50.45dyoung-webthat sounds about right....
04:51.02dyoung-webits inline with what dsaxena said earlier.
04:51.06[g2]which means with Unslung and no paramter unslung would run with 64MB right and APEX
04:51.43dyoung-webalso supposedly for the dyoung modifed redboot.
04:51.48*** join/#nslu2-linux d|dge (~didge@2001:960:624:1:0:0:0:1)
04:52.02dyoung-webif people wish to keep redboot recovery as an option.
04:52.48[g2]it's kinda funny to me because I've flashed APEX 50+ times and I STILL don't have JTAG on the slug
04:53.09*** join/#nslu2-linux Helius (
04:53.25[g2]I did have RedBoot replacment on the avila and it was all tested out there first
04:53.29jp30[g2] you must be not your own friend. "friends don't let friends" and all that
04:53.58[g2]jp30 I had it on the avila and all was tested there first, and then in memory on the slug
04:54.51[g2]I may be just a VCC pin away from it right now.. there's some more checking needed
04:55.21rwhitby-webdyoung-web: how much for the cable and shipping?  Was it $15 or so ?
04:57.20enojp30, heads up: although not ready for testing, i got erlang cross built
04:57.45jp30eno, good news. that will save 16 hours of processor time on the official build machine
04:58.02jp30let me know when it's ready to test
04:58.21enoit's not as difficult as i thought
04:59.11enobasically build a host version first. I'm wondering this general technique can be used for other hard to cross-build packages
04:59.55jp30if you can make it work for emacs and xemacs, i will be very thankful
05:00.53enoso the next long weekend is july 4th, i may take a shot at these then
05:00.54jacqueseno, that technique is widely used, for example by oe
05:01.07jacquesemacs however is another matter entirely
05:01.23jacquesit uses dump
05:01.30rwhitby-webyeah, the whole dumping thing
05:01.57jacquesthank rms
05:02.04rwhitby-webAnyone have good or bad reports about Powerware UPS's ?
05:02.07jacquesim sure it seemed like a good idea at the time
05:02.28enonot an emacs user myself, what is the "dump"
05:02.58rwhitby-webI'm looking at this one:
05:03.21rwhitby-webto sit near the floor of slug-central and have all the power packs plugged into it.
05:03.36[g2]jp30 so if you get a minute can you jot down and note or two about the web server benchmarking and configs ?
05:04.16[g2]rwhitby-web, winner-take-all  they all stand or all fall
05:04.41jacquesrwhitby-web, I'm not familiar with that brand
05:05.00jp30[g2], i'm not totally sure what you want. advice about where the bottlenecks are i can give; mechanical ways of testing for speed i know little about
05:05.32[g2]jp30, ok.  
05:05.47rwhitby-webjacques: the form factor is the big thing.  I need something with normal outlet sockets that can handle plug packs, not the standard IEC sockets that you get on most UPS's.
05:05.47jp30eno, i think what emacs is doing is storing an image of its internal state in full machine dependent glory to disk
05:05.50[g2]I'm sure will get it worked out over time
05:06.24jp30[g2], would it help if i wrote up a bit on the wiki about how i configured apache, php, eaccelerator and why?
05:06.41[g2]jp30, *absolutley*
05:06.46jacquesrwhitby-web, it looks nice in the photo - and yes, mine have 4 regular and 2 "wall wart" sockets
05:06.47jp30ok, will do
05:08.00enojp30, thx for the explanation
05:09.14jacquesit does it to reduce list file parsing at startup
05:10.13jp30jacques, i take it there's no way to get an emacs to write out a dump in a foreign byte-order / word-size / alignment?
05:10.39jacquesI've never heard of one
05:10.50jacquesthere is an explanation here:
05:11.01jp30hmm. it is like nethack then. "great minds think alike"
05:11.35jacqueshmm wan't that one of stallman's too?
05:11.45jacqueshere's a thing
05:12.03jacquestoo much trouble probably, but you could cross-build temacs, and then only do the dump stel on the slug
05:12.09jacquesdump steP
05:12.13jp30i was just going to say
05:12.28jp30that's the technique i used for pear, too
05:13.43jp30PHP Extension and Application Repository
05:14.31jp30emacs takes a long time to compile and it's x11 support needs work, so it would be useful to have it cross-compiling
05:16.14rwhitby-webjacques: what UPS do you have?
05:24.17jacquesrwhitby-web, I have three APC Back-UPS XS 1500
05:25.18jacquesI've always bought that brand (APC) not like thei'r the best in the worls or anything, but they've worked fine - I've probably have about 7 of their devices - I bought these because I wore the batteries out on my last (smaller) ones
05:25.49jacquesdamn I can't type tonight
05:28.09jacquesthe batteries wore out after about 5 years
05:28.14jacquesof constant use
05:28.28jacquesthe 1500's are new as of last year
05:28.31jacques(to me)
05:29.41rwhitby-webyeah, I have a height restriction, so form factor is the main thing
05:29.44rwhitby-webok, back later
05:29.48*** part/#nslu2-linux rwhitby-web (www-data@rwhitby.nslu2-linux)
05:30.25[g2]nite all
06:37.50*** join/#nslu2-linux ka6sox (
06:54.47*** join/#nslu2-linux Christopher (
08:18.48*** join/#nslu2-linux eno-away (
09:45.33*** join/#nslu2-linux rwhitby-nslu2 (~rwhitby@rwhitby.nslu2-linux)
09:47.16*** join/#nslu2-linux mr_claus (
10:07.09dyounganyone around?
10:07.17dyoungokay lets qualify that.
10:07.23dyoungjacques or rwhitby around?
10:30.08copperbeechI have just been surprised by getting an email from Andrew Tridgell in my mail box...
10:30.45copperbeechunfortunately not as exciting as I thought when I realised it was a post to the appWeb mailing list
10:31.31copperbeechapparently looks like they are using appweb for some part of samba 4.
10:35.32*** join/#nslu2-linux rwhitby-away (~rwhitby@rwhitby.nslu2-linux)
11:06.24*** join/#nslu2-linux rwhitby-nslu2 (~rwhitby@rwhitby.nslu2-linux)
11:11.58*** join/#nslu2-linux rwhitby-nslu2 (~rwhitby@rwhitby.nslu2-linux)
11:24.53*** join/#nslu2-linux rwhitby-nslu2 (~rwhitby@rwhitby.nslu2-linux)
11:48.04*** join/#nslu2-linux rwhitby-nslu2 (~rwhitby@rwhitby.nslu2-linux)
12:11.30mr_claushmm, i finished working with my nslu2 2 days ago and now i cannot login with ssh again
12:11.43mr_clausno web frontend
12:11.44rwhitbyjbowler-zzz: we'll also have to handle the update of /boot with update-kernel too, as ipkg will only do it on the external rootfs, and it needs to be updated in the jffs2
12:11.46mr_clausno ssh
12:19.25*** join/#nslu2-linux cbrake (
12:21.18mr_clauswhich could be the right password, is there a default password which is being used with dropbear ssh?
12:21.26mr_clausi have set the password but i cannot login
12:30.17copperbeechmr_claus: have you read through
12:32.25mr_clauscopperbeech: yes i have read a few weeks ago as i started but perhaps something has changed with the last upgrade
12:33.17copperbeechok - these howtos set-up dropbear and open-ssh not to accept password verification, only key verification
12:33.27copperbeechthat might be something woth checking.
12:33.50mr_clauscopperbeech: hmm, but it worked with password verfication 2 days ago
12:34.18copperbeechok - what exactly did you upgrade?
12:41.03*** join/#nslu2-linux rwhitby-away (~rwhitby@rwhitby.nslu2-linux)
12:42.54mr_clauscopperbeech: i upgraded nothing, i run passwd and it worked, today its not working anymore
12:43.11mr_clauscopperbeech: but i think anything was changed but i cannot say what i changed
12:43.37mr_clauscopperbeech: is there a way use passwords instead of keys?
12:43.59copperbeechchanging passwords is a bit tricky - I assume you followed the ins and outs described in the wiki at
12:44.52copperbeechI think dropbear and openssh default to use of passwords... the two howtos for dropbear and openssh above change the configuration to use keys for security reasons when opening up the slug to the internet.
12:53.51mr_clausbut should dropbear not take the passwords from /opt/etc/password instead of /usr/local/passwd and /share/hdd/conf/passwd?
12:53.52*** join/#nslu2-linux k81caplin1_log (
12:54.13mr_clausor is /opt/etc/passwd == /etc/passwd if unslung was done?
12:55.24mr_clausor is /share/hdd/conf/passwd linked to /opt/etc/passwd?
13:01.54copperbeechif you look at it shows which are individual files and which are links... either way they all need to be the same.
13:02.19copperbeechsorry I meant :
13:05.16mr_clausis the FAQ 4.20 ready? it seems that with 4.20 the files are in /share/hdd/data instead of /share/hdd/conf
13:08.52copperbeechThe FAQ may not detail that but it was a change which was made in 4.20 to default the /opt folder onto the data partition where there was more space for installing packages.
13:09.33copperbeechhave you done a > find / -name 'passwd' ?
13:09.58mr_clauscopperbeech: at the moment i have no running nslu2
13:10.19mr_clauscopperbeech: but i will switch off the disk and reboot and switch it on again
13:10.22copperbeechbecause the password issue?
13:12.31mr_clausyes, and the web interface is not running too, i dont know why
13:12.40mr_clausso i cannot telnet into my box
13:14.38mr_clausfind: /proc/9/fd: No such file or directory
13:14.55copperbeechcan you even see your box ? is the ip address correct?
13:15.10mr_clausyes because i can use ssh but i cannot login
13:15.44copperbeechhave you tried booting it without disks attached and trying the web interface then?
13:16.23copperbeech(with no disks attached it will be the default passwords "uNSLUng" for telnet and "admin" for the web site IIRC
13:17.00mr_clausyes of course, booted now without disks attached, activated telnet with the webinterface and login
13:17.10mr_clausi plugged in the disk after the boot process
13:17.22mr_clausnot i have to repair the things are not working
13:17.44copperbeechok - and then can the lines you pasted above gibe the results of the find / -name 'passwd' ?
13:17.51copperbeechgibe = give
13:18.06mr_clausyes, but i did a mistake with posting
13:18.13mr_clausfind: /proc/9/fd: No such file or directory
13:18.21mr_claushmm, something wrong here
13:18.45mr_claushmm, yes the / at the beginning perhaps :)
13:19.12mr_clausdoesn't matter, so i will copy the /etc/passwd to all other locations now?
13:19.44copperbeechI would say so - as per the wiki
13:19.46mr_clausbut what i need to know, which version will be used by dropbear?
13:21.48copperbeechgood question - not sure. But should not matter if they are all the same.
13:21.56copperbeechThey all should be the same.
13:22.23copperbeechAs I mentioned before I have never faced this problem because I use public/private key pairs.
13:29.43*** join/#nslu2-linux VoodooZ (
13:35.41mr_clausok, copied the password and rebooted without a disks connected
13:35.51mr_clausno web interface, no telnet
13:36.04mr_clausbut i can ping the machine
13:39.08mr_clausok, i will continue that work on weekend, thx for your help
14:35.06*** join/#nslu2-linux phedny (
15:41.22*** join/#nslu2-linux phedny (
17:10.57*** join/#nslu2-linux ka6sox-office (~tking@ka6sox.nslu2-linux)
17:12.43*** join/#nslu2-linux [cc]smart (
17:15.30*** join/#nslu2-linux [cc]smart (
17:25.24*** join/#nslu2-linux Jacmet (
18:06.38*** join/#nslu2-linux Christopher (
18:19.58*** join/#nslu2-linux llagendijk (
18:20.15*** join/#nslu2-linux llagendijk (
18:30.12*** join/#nslu2-linux Christopher (
19:01.14*** join/#nslu2-linux giel (~giel@
19:07.50copperbeech~jbot, seen [g2]
19:07.57jbot[g2] <~g2@g2.nslu2-linux> was last seen on IRC in channel #nslu2-linux, 13h 37m 32s ago, saying: 'nite all'.
20:14.40*** join/#nslu2-linux caplink811 (
20:18.29*** join/#nslu2-linux caplink811 (
20:24.05*** join/#nslu2-linux Christopher (
20:33.29*** join/#nslu2-linux Christopher (
20:55.57copperbeechI've just managed to get linux installed on my laptop and have got the build tree from CVS... but...
20:56.16copperbeechwhen I type make toolchain it is failing with an error:
20:57.11copperbeechtar: /home/mcneillm/unslung/toolchain: cannot chdir: No such file or directory
20:57.18copperbeechanyone out there got any ideas?
20:58.06llagendijkare you using the same userid as when you retieved the build tree from cvs?
20:58.22dyoungmake directories
20:58.27copperbeechno - but I did a chown so it was equivalent.
20:58.46dyoung"make directories" rather.
20:58.47llagendijkjust to be sure: cd  /home/mcneillm/unslung/toolchain
20:58.47copperbeechaaaahhh make directories!!!
20:59.03copperbeechdir mkdir toolchain and it seems to be off now.
20:59.16copperbeechwhat should I make first before the toolchain?
20:59.29dyoung"directories" . :-)
20:59.39copperbeechwhat about the make ipkg ?
21:00.01dyoungI think it makes that as a dependency later.
21:02.05copperbeechgosh it's so fast compared to the slug!
21:02.45dyounguh huh.
21:03.11copperbeechThis experience of tweaking a .mk file ha been a real "horse-shoe nail" job!
21:03.24dyounglearning is fun.
21:03.56copperbeechI've been having to learn GRUB, Knoppix, Debian, XP multiboot etc etc etc
21:04.29copperbeechyeah it's been kind of fun in a sadistic way.
21:05.11*** join/#nslu2-linux Christopher (
21:06.11Christophermy second slug came yesterday :-)
21:06.22copperbeechwhat you going to do with it?
21:06.36Christopheruse it with openslug, possibly make a fatslug out of it
21:06.49Christopherthis way I can run openslug and unslung at the same time
21:07.12copperbeechI'm still on unslung 3.18 - I haven't had the time to go back to basics and install 4.20
21:07.25ChristopherI think i'll run most of my applications on one box and then have only apache running on the other
21:07.34*** join/#nslu2-linux [g2] (~g2@g2.nslu2-linux)
21:07.45copperbeech[g2] hey!!
21:07.56Christopheryeah, i haven't installed 4.20 yet
21:08.02[g2]copperbeech, hey!
21:08.33copperbeechI've just managed to get Knoppix installed to a harddisk and dual booting - the toolchain is making itself as we speak...
21:08.43copperbeech[g2] what's the news
21:08.49[g2]copperbeech cool!
21:09.06[g2]you probably don't need to dual boot
21:09.15[g2]but that's nice too :)
21:09.25copperbeechhe he - I do. It's not my laptop and work needs XP.
21:09.46copperbeechI was booting off a CD but of course it's non-persistent
21:10.00[g2]no I meant you only would need one or two linux partitions and no Linux install :)
21:10.30[g2]there's an option for having persistent info (on the said partition) :)
21:10.57jp30-workcopperbeech, congrats on getting the cross build system working
21:11.10copperbeechhmmm yes but i just thought I might as well install it to the partition. I didn't have to change the MBR because I am using grldr and adding the Grub bootloader to the Windows XP boot.ini which is a really nice option
21:11.49jp30-workwe'll be seeing a cross-built php enabled appweb 2.0.0 with singing and dancing and doing your dishes for you soon?
21:12.20[g2]jp30-work, I was looking on the wiki for those apache notes :)
21:12.28copperbeechhe he - I need to pick your brains about that php one day when I get back up the "call-stack" from GRUB to appWeb again.
21:13.41[g2]do you guys know if there's a way to securely tunnel the ipk stuff ?
21:13.44jp30-work[g2], i'll get to them
21:14.08[g2]jp30-work, nod. I'll be gone for most of the weekend
21:15.30copperbeech[g2] if you want any help with testing of the knoppix iso - let me know. I'm also intending to write some wiki stuff alongside it when it becomes available.
21:16.33*** join/#nslu2-linux CIA-8 (
21:17.06[g2]copperbeech, ok.  Right now I'm unsure what the best way to create the swap and data area is
21:17.11[g2]on an NTFS system
21:17.41copperbeechwhat are the options
21:18.17[g2]there's some chant at creates and image in a single file which can be used for swap
21:18.41[g2]we could basically do that twice (one for swap the other for data)
21:18.48[g2]that's option a)
21:19.22[g2]option b) is tell'm to use ghost/norton and make some space on a partition avialable
21:19.34copperbeechisn;t there a problem getting write access to NTFS?
21:19.39[g2]option c) is plug in an extrnal USB 2.0 drive and do it all there
21:19.56[g2]I think with option a) you can do it to a single file or two
21:20.03copperbeechI went for (b)
21:20.24[g2]you went b+ where you actually dual boot
21:20.26copperbeechOne thing I have been doing is booting from the ISO on an NTFS drive
21:21.15copperbeechWell I booted from the ISO and did an HD install to the partition and then sussed out grub to boot that partition... calling grub from XP boot.ini
21:21.28copperbeechI think the problem is that there are too many options.
21:21.52[g2]right, you could just boot from the iso, install the persistent data and on subsequent boots just use that
21:21.52copperbeechIs our intention to select one and then leave the other options open to the expert user?
21:22.08[g2]then you wouldn't need the dual-boot stuff
21:22.47[g2]an olive
21:23.04llagendijkjp30-work, did you make any progress with apache?
21:23.09copperbeechI think that the baseline should be to boot from an ISO - either from a CD or from an image - I can write a wiki for that
21:23.54copperbeechthe question is about the dynamic data and how best to manage that... how realistic is it to have data partiton files on NTFS?
21:24.17[g2]it's a single file that's loop mounted
21:24.40copperbeechyep - but what about the write access to NTFS?
21:24.59[g2]I don't think that's ready for prime-time yet
21:25.17[g2]if it was/is then there'd be no need to create the file/partition
21:25.22copperbeechno that's my underatanding too.. so that crosses out option (a)
21:25.49[g2]I think there's a way to do it with a single file or two
21:25.53copperbeechso now the question is between option (b) or option (c) and to be honest I think we need to support both
21:26.06jacqueshmm, i just read something about NTFS write the other day
21:26.09[g2]meaning for swap or for a loop mount partition you'd create 1 or 2 files
21:26.19copperbeechyes? how do you see it?
21:26.21[g2]and it's only accessing information associated with those files
21:26.26jacquesit was a FAQ - apparently there are two completely separate NTFS drivers
21:27.17copperbeechI think I remember - one has some read-write capabilities - but immature and only limited to already existing files, the other was a wrapper around a windows file
21:27.34[g2]my guess at how it works is it's just accessing the data of the file and not chaning the filesystem at all
21:27.50jacquessomeone understands NTFS, because ntfs-resize works perfectly
21:27.57jacquesyou dont even have to defrag anymore
21:28.16copperbeechHow reliable is it?
21:28.26jacquesfrom everything I have read, extremely
21:28.52jacquesas in no known failures other than human error
21:28.55copperbeechWhat about having a 0.1 release which supports options (b) and (c) and leaving option (a) for 0.2 ?
21:29.00[g2]if that's avialable then if the user has space and a free partition then they can resize ntfs and create the new partition
21:29.22jacquesubuntu has the resize capability as part of their install
21:29.23[g2]this is all a work-in-rapid-progress
21:29.27jacques(using ntfs0resize)
21:30.12copperbeech[g2] don't you use an external USB drive for the persistent stuff?
21:30.26[g2]copperbeech, it can be on a local drive
21:30.28jacquesI used the ubuntu install disc to resize the NTFS partition on my T42, then installed gentoo
21:30.51[g2]jacques, somehow that doesn't surprise me :)
21:30.57copperbeechUbuntu live is quite cool - I tried it out a couple of days ago.
21:31.21jacquesyeah, live worked great on the laptop too
21:31.21[g2]I ran ubuntu live from inside qemu
21:31.50[g2]it's fun to try out the distro's without actually burning the isos
21:32.51copperbeechhave you seen this:
21:33.18copperbeechYou just throw the latest ISO in a folder and choose it from the grub menu... nice
21:33.59[g2]is that just bootloader chaining ?
21:35.06copperbeechbasically it uses the XP boot.ini to chain grub which you can then use to chain the ISO, or in my adapted case, chain my HD install.
21:35.19copperbeechit's nice because you never have to mess about with the MBR
21:36.25copperbeechit's also nice because you never have to burn a CD
21:36.46copperbeechand it's also nice again because to upgrade, all you have to do is copy over the ISO file
21:37.16copperbeechand it's also even nicer than a CD because all the cheat-codes are saved in the grub menu-lst
21:38.39[g2]copperbeech, I'll have to take a look at that and maybe try it out on the kids pc
21:39.13copperbeechwell it's pretty harmless so you should get away with it :-)
21:39.46[g2]no, it's 2.5 years since any of my PCs have Window 'em
21:40.11[g2]take the new laptop for example
21:40.25[g2]I tested it ran knoppix before I bought it at the store
21:40.38[g2]came home dd if=/dev/zero the whole drive
21:40.48[g2]then freshly installed a knoppix
21:40.58[g2]I've been to lazy to put gentoo on it
21:41.00copperbeechYou're lucky... The latest Compaq Laptop I got won't boot knoppix off a CD - the CD drive is USB and it falls over
21:41.13copperbeechboots Ubuntu live OK though
21:42.36copperbeechThe seem to change very often... I've had a Toshiba, a Dell, a Compaq and now they've gone to IBM
21:44.03copperbeechmake toolchain is one very very big build!
21:44.57[g2]I'd be interested to see how fast the hd's run
21:45.07[g2]the default setup is very slow
21:45.24[g2]3MB versus 24MB on my laptop
21:45.28copperbeechwhich hds - my laptop hds on which knoppix is running?
21:45.45copperbeechhow would you test it?
21:45.52[g2]hdparm -tT /dev/hda
21:45.56llagendijkjp30-work, are you there?
21:46.13[g2]use at your own risk of course
21:47.13caplink811-away~seen byront
21:47.14jbotbyront <~byron-por@ByronT.nslu2-linux> was last seen on IRC in channel #nslu2-linux, 6d 19h 24m ago, saying: 'np'.
21:47.24jbotthanks, caplink811-away
21:47.35*** join/#nslu2-linux [g2]-amd64 (
21:47.40caplink811-away~seen byront-away
21:47.42jbotbyront-away <byrontodd@ByronT.nslu2-linux> was last seen on IRC in channel #nslu2-linux, 10d 31m 52s ago, saying: 'is my serial on the way?'.
21:48.18copperbeechtiming buffered disk reads: 82 MB in 3.13 seconds = 26.16 MB/sec
21:48.39copperbeechalthough that was with make toolchain running full tilt
21:49.13jacquesyeah my desktops get 57MB/s
21:49.20jacqueswell the ones with 7200rpm drives
21:49.30jacquesI was just saying 38MB/s is damn good for a laptop
21:49.38[g2]it is!
21:49.53copperbeechyes it is...
21:50.13jacquesthat 7200rpm laptop drive is apparently good for something
21:50.24[g2]but when I'm on the laptop it'll either be when I'm far away  or I'll just ssh in to the desktops
21:50.39jacquesi dont think I can go down in speed when I upgrade - I need more than 60GB tho - 100GB 7200rpm drives are coming soon but will be $$$
21:51.31copperbeechyou are talking to someone who has been compiling on a slug for the last couple of months - moving to a laptop is rather an experience
21:51.59[g2]just like old-skool when things just took that long
21:52.08[g2]like batch jobs and card readers
21:53.26llagendijkg2, are you THAT old?
21:55.52copperbeechThe first computers I really remember as a kid were ZX Spectrums
21:56.25copperbeechI didn't even get my hands on a computer until I was about 13 (Acorn A3000 with an ARM2 processor)
21:56.26jp30-workhi, llagendijk
21:56.42llagendijkllagendijk, I did my first programming at school on a Sperry Remington from '58 I  believe....
21:57.00llagendijkhi jp30-work
21:57.15llagendijkjp30-work, any progress on apache?
21:57.47llagendijkcopperbeech, you must be much younger then I am
21:58.00jp30-workwell, i've fixed the libraries problem you noticed, but i doubt that it was responsible for the segfault
21:58.07copperbeechyeah - born in '75
21:58.17jp30-workhaven't had a chance to try it out on a wiley yet
21:59.13llagendijkjp30-work, I seem to have some problems with the package for apache: when I install it, it complains in the ipkg run about
21:59.16llagendijk[admin@pheasant1 info]$ ipkg -V3 configure apache
21:59.16llagendijkConfiguring unpacked packages
21:59.16llagendijkConfiguring apache
21:59.16llagendijkRunning script //opt/lib/ipkg/info/apache.postinst
21:59.17llagendijkmkdir: Cannot create directory `/usr/lib/ipkg/': Read-only file system
21:59.51llagendijkthe /usr/lib/ipkg part is funny, I can't find where that comes from
22:00.17jp30-workllagendijk, try building your own ipkg from unslung and using that. i have patched this bug but it isn't in the feed as jf-work isn't happy with putting ipkg in the feed yet
22:00.29jp30-workjf, rather
22:00.37llagendijkcopperbeech, I'm from 58
22:00.44llagendijkjp30-work, that is what I did
22:01.04jp30-workodd. i thought i'd patched that
22:01.17llagendijkI am actually trying to get gdb installed, so I compiled without stripping
22:01.32llagendijkjp30-work, what was the problem?
22:01.33jp30-workyou have ipkg 0.99.148-4?
22:01.57[g2]llagendijk, actually I *am* that old
22:02.09[g2]I used punch cards in college
22:02.13dyoungcopperbeech, that wasnt too bad.  I think it takes an hour fo rme.
22:02.34llagendijkg2, thought so, you must be at least as old as me to remember puch cards etc
22:02.45dyoungPunch Cards are cool.
22:02.45[g2]jacques, copperbeech ntfsresize is on the dvd and looks like it'll work
22:02.48jp30-workllagendijk, must think back to what i did with ipkg... hmm...
22:03.18dyoungUntil you fall down carrying your stack and lazily didnt band them together.
22:03.20[g2]dyoung, especially when you used the radix sorter to put the back in order using the numbers in column 73-80
22:03.21copperbeech[g2] are you going to just realease a DVD or a CD as well?
22:03.40dyoungyeah.  heh.
22:03.48[g2]copperbeech, that's under consideration
22:04.00llagendijkjp30-work, no, I have -2, will try to upgrade
22:04.07jacques[g2], which version is it?
22:04.14copperbeechI think it's an important factor for package developers rather than kernel developers
22:04.24[g2]My DVD has a copy of all the metadata for OE and all the tarballs to go with it
22:04.38[g2]jacques, version of the metadata or knoppix ?
22:04.46jp30-workgotta go...
22:04.50jp30-workbye for now
22:04.55[g2]see ya
22:04.59copperbeechyeah - thought so. That would be fantastic for kernel devs
22:05.09llagendijkjp30-work, cu later
22:05.36llagendijkllagendijk, doesn't feel old either
22:05.47copperbeechhowever - from my perspective (as a noob package dev) I would just want a basic package cross development env.
22:05.50jacques[g2], version of ntfsresize
22:06.18[g2]if that matters :)
22:07.03jacques[g2], ok great
22:07.25copperbeechwill play with some appweb compilations tomorrow...
22:07.27jacques"Choosing a distribution including ntfsresize version 1.9.0 or later is recommended, at least until NTFS gets resized, because it can also resize fragmented NTFS safely, there isn't even need for defragmentation in advance.
22:07.40copperbeechthats for all the help people...
22:08.21copperbeechthats = thanks
22:10.04[g2]jacques, Excellent
22:10.12[g2]thx for the info and advice!
22:10.32[g2]so now option b) is pretty easy
22:10.48[g2]cfdisk and see if there are not 4 partitions in use
22:11.09[g2]then ntfsresize the main partition if there's room
22:11.29[g2]the create a partition
22:11.58[g2]create a swap space on the partition or a second partition (depending on the number of original partitions)
22:12.07[g2]and create the persistent area
22:13.07[g2]and boot with that option on subsequent reboots use the persistent data
22:23.46llagendijkoff o bed
22:34.21*** join/#nslu2-linux jf-work (~none@jeanfabrice.nslu2-linux)
22:47.24*** join/#nslu2-linux jf-work (~none@jeanfabrice.nslu2-linux)
22:50.05*** part/#nslu2-linux kergoth (
22:51.15*** join/#nslu2-linux kergoth (
23:36.54*** join/#nslu2-linux odoc (

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