**** BEGIN LOGGING AT Sat Sep 12 02:59:56 2009
Sep 12 03:08:10 <morfic> well, unable to open an initial console i see as a full on success ;)
Sep 12 03:12:06 <morfic> from a booting the kernel from disk point of view
Sep 12 03:20:58 <tinker-f595> ?
Sep 12 03:28:24 <morfic> tinker-f595: with rsync i --exclude=/dev and haven't been running MAKEDEV since, kernel boots off usb disk, but can't open console w/o that run, so next boot it should be fine
Sep 12 03:28:46 <tinker-f595> huh?
Sep 12 03:30:24 <morfic> how'd i lose you there?
Sep 12 03:31:28 <morfic> i moved the running fs off the sheeva, but excluded /dev, i never ran MAKEDEV on /dev in a tired "let me just boot kernel off disk" frenxy
Sep 12 03:31:33 <morfic> frenzy too
Sep 12 03:34:03 <morfic> tinker-f595: most my time is consumed by trying to reset the sheeva in time to have both disks up for boot, i managed that almost perfectly, right now i sit on time outs again
Sep 12 03:34:41 <tinker-f595> ha
Sep 12 03:36:53 <morfic> chuckachucka!
Sep 12 03:37:03 <morfic> MAKEDEV magic
Sep 12 03:37:49 <morfic> tinker-f595: now where i can boot uImage.20090403 from disk, tell me why this owuld never work from 2.6.30.6 (mainlineLinux Yes) with the arc of 2097 and all that good stuff?
Sep 12 03:39:03 <tinker-f595> you should be able to boot from disk with 2.6.30.6
Sep 12 03:40:10 <tinker-f595> I am using 2.6.30.5
Sep 12 03:40:35 <tinker-f595> mainline stuff is orion
Sep 12 03:41:10 <morfic> well, i can restore all images now, can restore uboot to default environment, flip between booting from nand and disk
Sep 12 03:41:24 <morfic> i guess when i have my days off i will look into other kernel
Sep 12 03:41:38 <morfic> i will have enough to do setting up all services
Sep 12 03:42:03 <morfic> thanks for your factory defaults link, it helped me restore some sanity and helped get me back on track
Sep 12 03:43:04 <morfic> tinker-f595: i may just skip all newer kernels until all the scheduler and swap compressor work is done, unles i hit an actual problem :)
Sep 12 03:43:28 <tinker-f595> I think you are doing something incorrect in u-boot
Sep 12 04:21:36 <morfic> tinker-f595: what do you mean? it boots off usb disk.....
Sep 12 04:22:18 <tinker-f595> you said you could not get the later kernels to work
Sep 12 04:22:29 <morfic> just that the disks + passive hub + u-boot is not a good combo, once in linux, the drives act normal, something about u-boot that makes the drives go bye bye way quick
Sep 12 04:22:35 <tinker-f595> that is because they are orion kernels and you need that info in u-boot
Sep 12 04:22:47 <morfic> tinker-f595: oh, that, i will look into it later
Sep 12 04:23:21 <morfic> the only time i see orion referenced is in the nand setenv stuff, none of the external disk ones ever reference orion
Sep 12 04:24:03 <tinker-f595> I am about to build a hhd with a really recent kernel and fed 11 rfs on it for my openrd client
Sep 12 04:24:36 <tinker-f595> oh I thought you said you were booting the kernel from nand and using rfs on disk
Sep 12 04:24:48 <morfic> i am booting kernel from disk
Sep 12 04:30:03 <morfic> i will see how things go next time i try from disk
Sep 12 04:30:41 <morfic> setenv bootcmd 'usb start; ext2load usb 0:1 0x8000000 /boot/uImage; bootm 0x8000000; reset' i just change to /boot/uImage-2.6.30.6 and saveenv and reboot
Sep 12 04:31:06 <morfic> along the mainline and the arc 2097 thing
Sep 12 04:32:44 <morfic> you said arcNumber can be ignored since it's the device number, and this mainlineLinux yes/no should allow switcing between current kernel and mainline kernel w/o issues?
Sep 12 04:34:01 <tinker-f595> no I did not
Sep 12 04:34:22 <tinker-f595> arcNumber is the type of device you are running on.
Sep 12 04:34:42 <tinker-f595> If you do not have the arcNumber set correctly you will not boot
Sep 12 04:35:14 <tinker-f595> there are something like 3000 different arm devices registered
Sep 12 04:36:21 <tinker-f595> if you do not have the correct arcNumber set common-head.S does not know what to look for in the kernel space to jump to
Sep 12 04:37:05 <tinker-f595> to put it simply
Sep 12 09:10:43 <Super_Roach> Hey all
Sep 12 09:17:11 <Super_Roach> Could anyone take a look at the error message I'm getting when attempting to flash using the 1.0 release: http://pastie.org/614482
Sep 12 09:28:56 <fnordianslip> Super_Roach: apt-get install libftdi1
Sep 12 09:56:35 <Super_Roach> Thanks fnordianslip
Sep 12 09:56:51 <fnordianslip> np
Sep 12 09:58:24 <Super_Roach> What would be the next step? I'm trying to use console back into the openplug, but I don't think the serial is working... trying /dev/ttyS0 , 1, 2 gave a cannot communicate error. I'm trying at 9600 bps atm
Sep 12 10:02:13 <fnordianslip> o get the serial/usb console, i've been using "sudo screen /dev/ttyUSB0 115200"
Sep 12 10:02:24 <fnordianslip> to, even
Sep 12 10:05:53 <morfic> tinker-f595: when i said arcNumber can be ignored, i was refering to arcNumber 2097 as it is shown for mainline kernels, would be the same for regular kernels, didn't mean to make it sound like 2097 can be picked at random ;)
Sep 12 10:08:20 <morfic> <morfic> i change the arcNumber when reading the debian link, what does that need to be if i just copied ubuntu to sda1?<tinker-f595> same<tinker-f595> arcNumber is the type of device
Sep 12 10:08:26 <morfic> refering to that :)
Sep 12 10:17:39 <Super_Roach> Woot that opened a connection! i think its hung though, I don't get output and can't type though
Sep 12 10:18:39 <fnordianslip> Super_Roach:  what state is the plug in? freshly reflashed with installer1.0?
Sep 12 10:26:30 <Super_Roach> It's newly reflashed with installer 1.0
Sep 12 10:27:51 <Super_Roach> We just got the putty shell working (we're at a unconference),  ## Booting image at 008000000    Bad Magic Number
Sep 12 10:30:10 <fnordianslip> hmm.  disclaimer - only got mine yesterday, but i have been through this process once.  did installer1.0 run properly, without error?  i first tried with a USB stick that it didn't like, and i used another one instead, which was OK.  i let screen run while running runme.php in another shell, and when runme completed, the plug rebooted, and i started seeing output in screen
Sep 12 10:30:46 <Super_Roach> It did run successfully on the shell when i done the php5 runme.php
Sep 12 10:31:24 <Super_Roach> bad magic number seems to mean for me that it didn't boot on the serial though
Sep 12 10:34:12 <fnordianslip> hmm.  i did see that in an earlier attempt with my dodgy usbstick.  in dmesg, i see the following bootline:
Sep 12 10:34:13 <fnordianslip> console=ttyS0,115200 mtdparts=orion_nand:0x400000@0x100000(uImage),0x1fb00000@0x500000(rootfs) ubi.mtd=1 root=ubi0:rootfs rootfstype=ubifs
Sep 12 10:36:14 <fnordianslip> so it looks as thought he uImage should be at 0x100000
Sep 12 10:36:43 <fnordianslip> i reckon you should run runme.php again, and watch what goes on in the console too.
Sep 12 10:39:13 <Q_Contoxicated> So based on the doco, it looks like upgrading from alpha-installer-6 to installer-1.0 is worthwhile.
Sep 12 10:42:57 <fnordianslip> Q_Contoxicated: I couldn't say, i've only had mine since yesterday ...
Sep 12 10:44:31 <Q_Contoxicated> I've only updated the on-device Flash, looking to install to a 250GB HD.
Sep 12 10:44:39 <Q_Contoxicated> latest-greatest is probably my best bet.
Sep 12 10:44:58 <Q_Contoxicated> Need to modify installer schtuff to install to said HD.
Sep 12 10:45:06 <Q_Contoxicated> But, boozified, so now might not be best time...
Sep 12 10:45:07 <Q_Contoxicated> lawl
Sep 12 10:45:16 <Q_Contoxicated> I know I can unbrick it though, so not horribly concerned.
Sep 12 10:48:51 <fnordianslip> yep, that serial/jtag usb interface is awesomely convenient
Sep 12 10:50:37 <Q_Contoxicated> I'm still confused on how to install to USB drive though.  (HD in my case, not Flash or SD)
Sep 12 10:51:04 * Q_Contoxicated is currently reading through the 1.0 readme for the first time.
Sep 12 10:51:33 <Super_Roach> interesting
Sep 12 10:54:14 <Q_Contoxicated> So if anyone has more knowledge on how to install to HD, it'd be very appreciated.  (The documentation seems to be lacking on how to install to alternative media, leaving it up to the end-user to 'simply know' or 'figure it out on their own')
Sep 12 10:56:24 <fnordianslip> so am i and i'm not pissed
Sep 12 11:00:50 <Q_Contoxicated> I'm not either, it's just annoying. lawl
Sep 12 11:00:52 <fnordianslip> native compilation on the plug is so much faster than on my NSLU2.  code that would take over 10 hours to build on the slug is done in 30 mins on the plug. woohoo!
Sep 12 11:01:09 <Q_Contoxicated> Yeah, 1.2GHz versus what, 266Mhz?
Sep 12 11:01:22 <fnordianslip> underclocked to 133MHz by default
Sep 12 11:02:02 <Q_Contoxicated> Ouch.
Sep 12 11:02:04 <fnordianslip> the 32MB RAM was a big problem for the slug
Sep 12 11:02:25 <Q_Contoxicated> So, roughly 10% of the CPU power, and 1/16th the RAM.
Sep 12 11:02:31 <Q_Contoxicated> I can see the compilation time difference right there.
Sep 12 11:02:52 <Q_Contoxicated> Not counting architecture gains from whatever extra features the Plug has over your NSLU2.
Sep 12 11:03:23 <fnordianslip> indeedy.  still, might try and get the quadcore ubuntu box upstairs to run the crosscompiler with distcc anyway
Sep 12 11:06:18 <Q_Contoxicated> Mmmmmm.
Sep 12 11:11:59 <fnordianslip> are there any sensors for sensors-detect to find?  if so, i'll install lm-sensors and give it a whizz
Sep 12 11:13:51 <Q_Contoxicated> I'm not aware of any sensors, but I am far from a guru as to what's all there :-D
Sep 12 11:19:01 <fnordianslip> oops.  segfault running sensors-detect, and kernel oops. think i'll uninstall lm-sensors ...
Sep 12 11:20:41 <SuperRoach> back
Sep 12 11:22:18 <Q_Contoxicated> YAY
Sep 12 11:22:55 <fnordianslip> SuperRoach: success yet?
Sep 12 11:23:30 <SuperRoach> no luck yet
Sep 12 11:23:47 <SuperRoach> I'm just trying to connect back to it, and do a reinstall so I can try and see where the flash fails
Sep 12 11:24:28 <fnordianslip> my problem was cured by using a different usb stick.
Sep 12 11:24:52 <SuperRoach> What problem did you have? the bad magic error as well?
Sep 12 11:25:19 <armin76> fnordianslip: no sensors
Sep 12 11:25:21 <fnordianslip> yep, but there was evidence of problems before that too.
Sep 12 11:25:58 <fnordianslip> armin76: guessed as much, so no need to worry about the segfault/panic from lm-sensors then ;-)
Sep 12 11:26:02 <SuperRoach> neat, I've got a spare one I can try too
Sep 12 11:33:02 <SuperRoach> heard of unconferences by the way fnordianslip ?
Sep 12 11:36:18 <fnordianslip> i have heard of them, but never about them.  what is yours about?
Sep 12 11:37:55 <SuperRoach> Their great for meeting people... This one is about a lot of php, although it's not set in stone in any way
Sep 12 11:38:55 <fnordianslip> ah.  never ventured into php. in fact i think runme is the first php script i've ever run
Sep 12 11:46:09 <SuperRoach> woohoo, trying it again... done it by acident with no usb drive and had the same error... so might be the usb drive
Sep 12 11:47:24 <SuperRoach> doh, same error :/
Sep 12 11:47:41 <SuperRoach> would the log of it help at all? maybe there is a env value that is wrong
Sep 12 11:47:53 <fnordianslip> bad magic error, or installer error
Sep 12 11:48:04 <SuperRoach> The Bad magic error
Sep 12 11:48:52 <SuperRoach> Booting image at 008xx may be the problem if you said yours is at 01xxxx
Sep 12 11:49:06 <fnordianslip> does the installer look like it is flashing the new uboot, uboot-env, kernal and rootfs ok ?
Sep 12 11:52:36 <Q_Contoxicated> I'll have to look at the readme when I'm sober, and see if it includes enough info for a USB-HDD install...or else I'll just check the forums again.
Sep 12 11:54:40 <SuperRoach> yes it does... I left putty open to see it when it reboots. is there anything I should notice while its flashing?
Sep 12 11:55:13 <Q_Contoxicated> Yes, it'll take awhile between steps.  Other than that, I went back to other things.
Sep 12 11:55:26 <Q_Contoxicated> I left it for more than the prescribed amounts, with the alpha-6 installer.
Sep 12 11:55:33 <Q_Contoxicated> Said 5 minutes, I think I left it for 7 or 8
Sep 12 11:56:52 <SuperRoach> yeah, that time sounds about right. When it finished I noticed the putty console was streaming data as if it was rebooting
Sep 12 11:57:07 <fnordianslip> well, i've only done it once successfully, and haven't spent much time looking at the u-boot settings, but if it is trying to load the kernel from the wrong flash address, it would explain the problem you're having if the u-boot env is bad
Sep 12 11:57:37 <Q_Contoxicated> Best to leave the console (putty, whatever) open
Sep 12 11:57:45 <Q_Contoxicated> So you can possibly see any errors it runs into.
Sep 12 12:00:47 <SuperRoach> i can't see any besides the bad magic on booting the image
Sep 12 12:01:48 <SuperRoach> i do notice the image size is different to whats in the readme.txt .... readme is : 2837885 bytes read
Sep 12 12:01:49 <SuperRoach>    , the uboot/sheeva says 4194304
Sep 12 12:03:02 <Q_Contoxicated> I had a bad-magic number with alpha-6-installer
Sep 12 12:03:10 <Q_Contoxicated> I re-ran said installer, and it worked fine after that.
Sep 12 12:03:21 <Q_Contoxicated> YMMV, but it worked for me.
Sep 12 12:05:24 <SuperRoach> is there a way to verify the size of the image?
Sep 12 12:05:27 <fnordianslip> thats pretty much how it went for me, except i switched usb sticks also.  i'd try unplugging the usb cable, power-cycling the plug and trying runme.php again.  run screen or putty as soon as you've run runme
Sep 12 12:05:30 <SuperRoach> or check that it wrote
Sep 12 12:05:58 <SuperRoach> did you find your putty was able to stay open as you run reinstalls of it?
Sep 12 12:07:14 <fnordianslip> you could try manually running recover1 through recover4 from uboot too, i guess, but i am new with this
Sep 12 12:08:20 <fnordianslip> my screen session used to drop out a bit, and it turned out to be a dodgy usb extension cable.  after that, screen stayed connected via /dev/ttyUSB0 unless i cycled the plugs power
Sep 12 12:12:19 <fnordianslip> there are too many lines in uboot-env to read in the screen session - d'oh
Sep 12 12:12:45 <Q_Contoxicated> expand screen session's back-log, or enable logging?
Sep 12 12:15:00 <fnordianslip> Q_Contoxicated: can't remember the screen options for that, and the man page is huge ...
Sep 12 12:15:08 <Q_Contoxicated> heh
Sep 12 12:15:21 <Q_Contoxicated> and a quick google search can't help?
Sep 12 12:15:37 <fnordianslip> jfgi?
Sep 12 12:15:43 <Q_Contoxicated> yeah.
Sep 12 12:15:46 * Q_Contoxicated is too lazy
Sep 12 12:15:49 <Q_Contoxicated> and drunk
Sep 12 12:21:05 <fnordianslip> okay, -L logs into screenlog.0
Sep 12 12:26:00 <SuperRoach> lol
Sep 12 12:26:23 <SuperRoach> thanks fnordianslip ! We just done a recover 2 - 3
Sep 12 12:26:32 <SuperRoach> that has got us into it almost too!
Sep 12 12:26:43 <fnordianslip> almost?
Sep 12 12:27:09 <SuperRoach> im not sure from where to copy, the errors happen a fair bit, possibly not reading the rootfs
Sep 12 12:28:00 <fnordianslip> so it starts to load the kernel now?
Sep 12 12:28:16 <fnordianslip> load and run i guess
Sep 12 12:35:41 <fnordianslip> ah i get it.  the uImga egets copied form
Sep 12 12:35:53 <fnordianslip> oops.  unfinished crap above
Sep 12 12:40:17 <fnordianslip> don't run recover 3.  just recover1 and recover2.  recover2 invokes recover3 which then runs recover4, which reads the kernel and initrd into memory, then recover3 completes by erasing and writing the kernel into the MTD, then recover2 completes, so if you run anything after recover2, it may well end up not as desired, probably
Sep 12 12:57:14 <SuperRoach> OMFG IT WORKS
Sep 12 12:59:35 <fnordianslipeee> woot
Sep 12 13:01:59 <SuperRoach> besides myself lol
Sep 12 13:02:13 <SuperRoach> I haven't slept in a long time so i might hold off doing things till then
Sep 12 13:03:23 <fnordianslipeee> just spent a while reading the u-boot env and figured out how all the memory mapping and copying from usb stick is done
Sep 12 13:03:46 <fnordianslipeee> slleping, at a conference ?
Sep 12 13:03:51 <fnordianslipeee> sleeping, even
Sep 12 13:04:40 <SuperRoach> lol i know
Sep 12 13:04:43 <SuperRoach> its a two dayer
Sep 12 13:04:50 <SuperRoach> most of us havn't slept the day before getting ready
Sep 12 13:05:21 <fnordianslipeee> lol.
Sep 12 13:06:56 <fnordianslipeee> need to figure out how to crowbar 2.6.30.6 in there now
Sep 12 13:08:06 <fnordianslipeee> hmm.  why no lsusb?
Sep 12 13:12:00 <fnordianslipeee> ah. lsbutils package.  anyone know why my /etc/resolve.conf keeps changing?  is there a better way (interfaces?) to specifiy dns servers?
Sep 12 13:28:14 <fnordianslip> does anyone use compcache or swap on their plug?
Sep 12 13:31:55 <mwester> I always define swap space, but I hope it isn't actually used.  You need swap space defined if you're ever going to have one program use all the RAM, due to the behavior of the Linux OOM killer (which is screwed up IMO, but it is the way it is).
Sep 12 13:32:49 <fnordianslip> thats what i was thinking. have /home and a swap partition on sd card, so will use that swap.
Sep 12 13:32:51 <timtimred> that behavior is a last ditch resort though :)
Sep 12 13:33:01 <mwester> (actually it's more complicated than just "one program using all RAM", but that's a generalization that works reasonably close).
Sep 12 13:33:12 <fnordianslip> yep...OOM killer is pretty desperate
Sep 12 13:33:30 <fnordianslip> ulimit is your friend
Sep 12 13:34:02 <mwester> depends on your application.
Sep 12 13:34:24 <fnordianslip> ok.  ulimit is the kernels friend ...
Sep 12 13:35:18 <mwester> for most servers (and the plug is actually reasonably close to a small server), the default behavior of the OOM killer is appropriate, and if that's the case you should be fine without swap (i.e. all ram should be usable without swap).
Sep 12 13:36:29 <fnordianslip> i tend to agree.  my old slug is alot more precarious with only 32MB RAM
Sep 12 13:37:14 <mwester> The issue is if you have a single program that mallocs large amounts of memory, and tries to use "system()" or fork()...  then you either need swap which will never actually be used, or you need to change the OOM killer behavior (via sysctl setting IIRC).  And the slug has major issues with that, exactly because it has so little RAM.   opkg will trigger the OOM on that device.
Sep 12 13:38:13 <mwester> Just keep track of the stats, and avoid actually swapping on the SD card; it won't last as long as it should if you do that.
Sep 12 13:38:20 <fnordianslip> make -j4 does a pretty good job too, even with DISTCC helping out, as linker jobs get invoked in parallel too, and the poor thing gets quite depressed
Sep 12 13:38:28 <timtimred> SD is a crappy thing to swap to as well.
Sep 12 13:38:41 <mwester> (I prefer to swap on a swapfile, so that I don't have to dedicate a swap partition)
Sep 12 13:38:53 <fnordianslip> better than nothing though.  although, i am wondering about compcache
Sep 12 13:40:04 * ShadowJK has used swap on SD for over a year and the cards haven't died yet
Sep 12 13:41:26 <fnordianslip> i've only lost one SD card so far, in my N800, but I was running some code that did frequent writes onto it for a long time.
Sep 12 13:42:37 <fnordianslip> in my view, the job of the SD card is to be replaceable easily, not to perform well ;-) so if they wear out, so be it
Sep 12 13:42:46 <ShadowJK> I've lost a SD card, in my mp3 player. :P
Sep 12 13:43:02 <ShadowJK> I did 1 gigabyte of writes 5 times a week for 2 years
Sep 12 13:43:21 <ShadowJK> And then one day my mp3 files sounded funny :)
Sep 12 13:44:01 <fnordianslip> i've been pottering around with a redhat based embedded distro, and that has the readonly-root mechanism from fedora in it. it would be cool to get that into debian/ubuntu somehow.
Sep 12 13:51:23 <fnordianslip> is the ubifs rootfs compressed?  my guess is that it isn't
Sep 12 15:37:52 <tinker-f595> ShadowJK: but has the swap actually been used?
Sep 12 15:38:31 <ShadowJK> on my n810 it's in heavy use
Sep 12 15:38:56 <ShadowJK> 2gig in 5 days
Sep 12 15:39:44 <tinker-f595> well n810 is very constrained
Sep 12 15:40:22 <tinker-f595> my Sheevaplug never swaps to its swap partition
Sep 12 15:40:34 <ShadowJK> yeah
Sep 12 15:40:51 <tinker-f595> Q_Contoxicated: http://www.openplug.org/plugwiki/index.php/External_Hard_Drives
Sep 12 15:47:32 <fnordianslip> anyone know of a good guide to getting distcc to work for the plug on my quadcore ubuntu box.  i've just installed the CodeSourcery cross-compiler.
Sep 12 15:48:00 <rektide> sooo...
Sep 12 15:48:15 <rektide> there arent usb wifi adapters that support hostapd .  ;_;
Sep 12 15:48:20 <tinker-f595> rektide: hi
Sep 12 15:48:22 <rektide> i have no idea wtf i'm going to do
Sep 12 15:48:28 <rektide> hey there, good morning to you
Sep 12 15:48:48 <tinker-f595> why do you want a usb wifi adapter?
Sep 12 15:48:51 <rektide> besides cry myself back to sleep, eeee gaddss
Sep 12 15:49:44 <rektide> v. pci-express, its easier to position a usb wifi adapter in a sensible place
Sep 12 15:50:41 <fnordianslip> perhaps bluetooth would work instead, as a stopgap?
Sep 12 15:51:35 <tinker-f595> rektide: for your openrd?
Sep 12 15:52:27 <rektide> yes for openrd
Sep 12 15:52:44 <rektide> hrm i might be able to use bluetooth stopgap.  i have a bluetooth adapter, but i've never investigated PAN.
Sep 12 15:53:08 <rektide> my car computer is almost ready for install, but i dont have any way to talk to it!
Sep 12 15:53:35 <tinker-f595> rektide: and what are you going to do with your "car computer"
Sep 12 15:53:39 <fnordianslip> from what?
Sep 12 15:55:50 <tinker-f595> rektide: so you cannot find a wireless usb with the correct chipset for hostapd
Sep 12 15:55:53 <tinker-f595> ?
Sep 12 15:56:06 <rektide> tinker-f595: mostly it runs mpd and does file serving from a MS4UM
Sep 12 15:56:36 <rektide> http://www.openplug.org/plugwiki/index.php/User:Rektide
Sep 12 15:56:49 <rektide> ms4um = 4 bay esata enclosure
Sep 12 15:57:52 <tinker-f595> rektide according to some quick searches there are some usb wirelesss sets that should work
Sep 12 15:59:02 <rektide> i'm suspecting i was just BS'ed by someone in #wireless
Sep 12 15:59:16 <rektide> <@dragorn> rektide: I know of no usb device which has ap mode support
Sep 12 15:59:24 <tinker-f595> rektide: I looked at http://hostap.epitest.fi/hostapd/ which took me to http://wireless.kernel.org/en/developers/Documentation/mac80211 which then led me to http://wireless.kernel.org/en/users/Drivers So
Sep 12 15:59:38 <Nicolae> Has anyone tried to use an external enclosure with two hdds in JBOD mode with the sheevaplug? I've got one, some rosewill one with a jmicron chipset, and on the sheevaplug it only picks up one of the two drives. both show up fine in arch linux on my desktop. Running whatver the sheevaplug installer installs + the 2.6.30.6 kernel.
Sep 12 15:59:51 <tinker-f595> So maybe you need to build a kernel with the correct drivers
Sep 12 16:00:10 <rektide> I havent actually received my Alfa AWUS050NH yet
Sep 12 16:00:43 <rektide> which has the Ralink rt2870 chipset that I'm hoping has hostapd support
Sep 12 16:01:58 <rektide> i'll compile the rt2x00 driver's git tree as soon as the hardware shows up
Sep 12 16:02:07 <rektide> and keep my fingers crossed-- although that'll make typing interesting
Sep 12 16:02:25 <tinker-f595> rektide is that an Italian car?
Sep 12 16:02:40 <rektide> she's german
Sep 12 16:04:58 <rektide> hmm the User: page is pretty weak on "why" i'm doing this.  I'll fill in a better overview at some point
Sep 12 16:08:29 <tinker-f595> rektide: you should add the u-boot stuff form your page to the sheevaplug u-bbot page
Sep 12 16:09:17 <tinker-f595> rektide: http://www.openplug.org/plugwiki/index.php/U-Boot_Quick_Reference
Sep 12 16:12:32 <tinker-f595> rektide: who is german?
Sep 12 16:14:38 <rektide> my car is german.  why the Italian question ?
Sep 12 16:15:38 <tinker-f595> alfa
Sep 12 16:17:15 <rektide> oh, ha.  no, apparently a italian wifi adapter in a german car.
Sep 12 17:07:07 <eFfeM> tinker-f595: you have an openrd client board, haven't you ?
Sep 12 17:09:04 <eFfeM> if so do you also see excessive CPU usage when X is running ?
Sep 12 17:54:38 <rektide> X!  you heathens. ;)
Sep 12 17:54:46 <rektide> what a luxurious world you bask in
Sep 12 17:56:36 <fnordianslip> yep, i'm grateful that the plug has a serial port ;-)
Sep 12 17:59:34 <fnordianslip> so rektide, how are you powering your carputerplug ?
Sep 12 18:41:17 <Q_Continuum> There is no uboot-custom.txt in the installer folder in the 1.0 installation archive...
Sep 12 18:59:25 <rektide> fnordianslip: the openplug runs off the 12v bank i have in the back.  the esata array plugs into the inverter.
Sep 12 18:59:43 <rektide> which is silly, as i'm sure its power supply is just putting out 12v + some 5v
Sep 12 19:00:54 <fnordianslip> ah.  so the plug is running off a 12v to 5v dc-dc converter, and the disk array from an inverter?
Sep 12 19:01:53 <rektide> yup
Sep 12 19:02:19 <fnordianslip> whats the purpose of the disk array? just vast capacity, or what?
Sep 12 19:02:40 <rektide> capacity, speed & convenience
Sep 12 19:02:55 <rektide> its everything i'd want for any purpose in an easily tote-able form
Sep 12 19:03:04 <fnordianslip> is it sdd or hdd?
Sep 12 19:03:08 <fnordianslip> ssd, even
Sep 12 19:03:10 <rektide> hdd
Sep 12 19:03:34 <rektide> i'm only just beginning to really use it, and so far SMART hasnt freaked out
Sep 12 19:05:47 <fnordianslip> interesting.  i've been surprised how well HDD work in some pretty bizarre places, but i wouldn't be all that keen to expect them to last all that long
Sep 12 19:06:25 <rektide> its a boat of a car so it shouldnt be that "shocking"
Sep 12 19:06:40 <rektide> hopefully SMART tells me things are going poorly
Sep 12 19:07:04 <fnordianslip> fair enough. anti-vibration mounted?
Sep 12 19:07:19 <rektide> also, i'm working on an overlay FS that will have colossal read-ahead, such that i can shut down the drives 95% of the time
Sep 12 19:07:54 <rektide> not really.  i need to concoct some kind of platform for the array.
Sep 12 19:09:49 <fnordianslip> tie it to a balloon, and tow it behind
Sep 12 19:10:21 <rektide> it'd need one big ass balloon
Sep 12 19:10:45 <fnordianslip> how big is this array?
Sep 12 19:14:12 <rektide> 4 drive
Sep 12 19:14:33 <rektide> plus ~1.125x a drive size below the bottom drive with power supply in it
Sep 12 19:14:56 <rektide> probably weights 24~30lbs loaded up, oi!
Sep 12 19:15:22 <fnordianslip> sounds about the same size as my qnap TS-409 then
Sep 12 19:15:32 <rektide> yeah
Sep 12 19:15:58 <rektide> had i known, i may have gotten a TS-219p instead of an openrd
Sep 12 19:16:28 <rektide> little NAS unit with 2 drives and a separate JMB controller hosting a eSATA port, using the same kirkwood processor we're all coming to love
Sep 12 19:17:45 <fnordianslip> that must be quite new ... not come across it before
Sep 12 19:19:09 <rektide> i just found out about it via some debian-arm dudes website
Sep 12 19:19:20 <fnordianslip> still running the qnap distro on mine, never tried anything more open on it
Sep 12 19:19:22 <rektide> its got to be brand new though
Sep 12 19:20:34 <fnordianslip> still think its a bit overkill for a carputer though ;-)
Sep 12 19:20:51 <rektide> cum fileserve
Sep 12 19:21:16 <rektide> for a low power fileserver, theres not even anything to compare with
Sep 12 19:21:34 <fnordianslip> oh, it contains your pron ?
Sep 12 19:22:08 <rektide> would if i had any!
Sep 12 19:30:56 <fnordianslip> what kind of read/write speeds does the plug manage over USB2.0?  can it max out the disks or not?
Sep 12 20:31:53 <rektide> fnordianslip: i'm on an openrd platform
Sep 12 20:31:58 <rektide> so i have eSATA
Sep 12 20:32:07 <fnordianslip> oh.  flash.
Sep 12 20:32:31 <rektide> yeah its spiff.  almost worth the sacrifice of having a device thats 5x the size of an openplug. ;)
Sep 12 20:32:53 <fnordianslip> is the lead time as bad on them as the plug?
Sep 12 20:32:57 <rektide> havent run any numbers yet
Sep 12 20:33:01 <rektide> at least as bad; its insane
Sep 12 20:33:25 <rektide> took >1 month for my order to get to me
Sep 12 20:34:01 <fnordianslip> 5 weeks until dispatch for my plug
Sep 12 20:34:06 <fnordianslip> then 2 days to get to uk
Sep 12 20:34:57 <fnordianslip> now i'm just waiting for the n900
Sep 12 20:36:29 <rektide> dddammmnn
Sep 12 20:36:41 <rektide> thats so unacceptable. :/
Sep 12 20:36:45 <rektide> mine has a broken RTC
Sep 12 20:37:04 <fnordianslip> 1970 all over again?
Sep 12 20:37:06 <rektide> i need to tell them, but i cant imagine how bad its going to be
Sep 12 20:37:15 <rektide> seriously
Sep 12 20:37:41 <fnordianslip> is it  known issue?
Sep 12 20:38:04 <fnordianslip> does dmsg show errors trying to set it?
Sep 12 20:38:14 <rektide> dmesg shows errors, and the power on test in uboot FAILED
Sep 12 20:38:38 <fnordianslip> smeg
Sep 12 20:38:39 <rektide> "clock is at 09:59:59" \ "expected 10:00:00" or something akin ot that
Sep 12 20:38:48 <rektide> it doesnt move
Sep 12 21:53:43 <morfic> tinker-f595: how is your f11 install going?
Sep 13 00:30:38 <morfic> default 2.6.22 kernel does not support nfsd?
Sep 13 00:31:50 <ShadowJK> the default kernel doesn't support much anything
Sep 13 00:34:05 <morfic> minimalist is one way of making a default kernel i guess
Sep 13 00:35:18 <rektide> one way, not God's way
Sep 13 00:35:23 <rektide> the fridge is on my side
Sep 13 00:36:01 <morfic> rektide....hm...rektide, i came across links with your name in it, didn't i
Sep 13 00:36:34 <rektide> its possible.  there are links like that.
Sep 13 00:38:32 <morfic> ahh, fun, the mkimage in u-boot folder is a x86 binary :/, i was going to just compile in plug
Sep 13 00:39:30 <morfic> ok, so i build that too, no points for the lazy
Sep 13 00:41:21 * morfic wgets u-boot-latest <--- unless what's in nand needs to match what does mkimage ?
Sep 13 00:45:24 <morfic> maybe i should just grab 2.6.30.6 and see if i can get it to boot this time around
Sep 13 00:51:18 <morfic> oh well, here goes serial console, at least i have one on this thing
Sep 13 00:59:53 <morfic> finally, 2.6.30.6 actually giving me some love, reflashing every image there was and restoring to factory defaults shouldn't have done anything, but it let's me boot this kernel, set up the same as before, yay the wonder!
Sep 13 01:48:10 <tinker-f595> morfic: what distro you using?
Sep 13 01:48:47 <morfic> ubuntu
Sep 13 01:49:17 <morfic> tinker-f595: is this common, plug out of box won't boot a kernel, after you reflash every image, it boots them?
Sep 13 01:49:33 <tinker-f595> morfic apt-get install mkimage
Sep 13 01:49:42 <morfic> tinker-f595: i am using your external hdd link setenv stuff today the same i did with fresh plug
Sep 13 01:50:23 <tinker-f595> morfic: not seen boot problem you are talking about
Sep 13 01:50:30 <morfic> tinker-f595: that works, i was gonna follow the kernel compile  link i have open, does it generically with pulling source and building
Sep 13 01:51:18 <tinker-f595> mkimage just has to be in the path
Sep 13 01:52:00 <morfic> well, it's all i did, restore, boot from nand, copy kernel image to /boot, add mainlineLinux Yes and arcNumber 2097 (again), change /boot/uImage to /boot/uImage.30.6 and it came up as if it never had a problem
Sep 13 01:53:21 <tinker-f595> morfic: probably had a control character hidden in one of the arguments
Sep 13 01:53:44 <morfic> tinker-f595: c'mon as if i would change my serial console pasting ways :)
Sep 13 01:53:57 <tinker-f595> ha
Sep 13 01:54:52 <morfic> although i spent 75% of my time waiting for disks to come up in uboot in time, or stay up long enough is more like it, those 25% i spent to going over settings and i did the factory restore before the flashing of images, so at that point any previous error should have been gone
Sep 13 01:55:24 <tinker-f595> yep
Sep 13 01:57:17 <morfic> well, it's good it "just works" now
Sep 13 01:58:17 <morfic> u-boot, wonder what it does not do that linux does to have such trouble with usb disks, i mean, they may be external and set to sleep, but they never start "i have a booboo" blinking in linux (one or the other in uboot)
Sep 13 01:59:07 <tinker-f595> huh?
Sep 13 01:59:09 <morfic> i can hear the heads rest and hear it spin up on access in linux, the lag on first access after pause i accept, basically just wondering if u-boot will ever like my disks
Sep 13 01:59:31 <tinker-f595> I have no porblems
Sep 13 01:59:46 <tinker-f595> morfic what drive you using?
Sep 13 02:00:00 <morfic> tinker-f595: my disks don't just time out, they enter some error state even with your reset at end of the line
Sep 13 02:00:06 <morfic> WD10000H1U-00
Sep 13 02:00:28 <morfic> tinker-f595: next week's paycheck i look for a active usb hub, maybe it cures that
Sep 13 02:00:38 <tinker-f595> the reset is for if it times out, then the sheevaplug will restart
Sep 13 02:01:34 <morfic> well, the reset caught the disks one time, like time out, reset, perfect boot, you know.....once
Sep 13 02:03:33 <tinker-f595> hmm Western Digital Mybook...supposed to be self powered...
Sep 13 02:04:06 <morfic> it is plugged in, yes, all it does is use usb port to sense computer on/off as its own on/off "Switch"
Sep 13 02:04:08 <tinker-f595> comes with ac adapter so should not be a power problem
Sep 13 02:05:07 <morfic> they ran for a few months now on the slug, not like i reboot the slug a lot, but no issues with them, aside from the obvious lag for wake up, nfs never bitched because of it, it was what i would call perfect
Sep 13 02:05:21 <morfic> and nfs access now is prefect, just dir listing goes a little quicker
Sep 13 02:07:00 <tinker-f595> wonder if you have a bad usb cable?
Sep 13 02:09:41 <morfic> they never fail in linux, no data issues with them, they work fine on laptop too, they only do not get any love from u-boot
Sep 13 02:27:40 <SuperRoach> Hello there - what is the default ip of reflashed 1.0 openplug?
Sep 13 02:35:04 <tinker-f595> SuperRoach how did you reflash?
Sep 13 02:35:36 <tinker-f595> SuperRoach:  also did you set you mac address
Sep 13 02:38:00 <SuperRoach> tinker-f595, The Mac address I couldn't find from the docs, I had it as the default still - would that prevent it from booting?  I reflashed using the 1.0 just out
Sep 13 02:38:43 <tinker-f595> SuperRoach: did you think about looking at the label on your SheevaPlug?
Sep 13 02:39:01 <morfic> it's tiny
Sep 13 02:39:04 <tinker-f595> If you had read the docs you would have noticed that
Sep 13 02:39:32 <SuperRoach> tinker-f595, No, I meant about setting it in the actual scripts ( as in there is no uboot-custom.ini as said in the docs - forums have mentioned it being somewhere else)
Sep 13 02:42:30 <tinker-f595> it is uboot-custom.txt
Sep 13 02:42:54 <tinker-f595> never seen uboot-custom.ini mentioned anywhere
Sep 13 02:43:14 <tinker-f595> it is in the installer directory
**** ENDING LOGGING AT Sun Sep 13 02:59:56 2009