**** BEGIN LOGGING AT Sat Nov 25 02:59:57 2006
Nov 25 03:06:35 <Flipo>  Hello, I was wondering if anyone was successful accessing the nslu2 with unslug 6.8 using a Windows Mobile device ?
Nov 25 05:14:41 <NSLU2-IRC> hello
Nov 25 05:15:38 <NSLU2-IRC> is there anyone in here thats active?
Nov 25 05:16:29 <NSLU2-IRC> i have an unslug box.... and everytime i restart it there are 2 of every share. its like it thinks the same folder is on the other drive
Nov 25 05:19:48 <NSLU2-IRC> so see ya all later
Nov 25 05:19:53 <NSLU2-IRC> thx
Nov 25 12:16:22 <DieselDK> Hey
Nov 25 12:43:33 <rexsdog> hi
Nov 25 12:43:53 <rexsdog> can anyone tell me how the web interface's shares work? i cant find some of those shares in my smb.conf
Nov 25 12:49:29 <DieselDK> hey
Nov 25 13:16:52 <mwester> rexsdog: AFAIK nobody has ever bothered to completely decipher the Linksys' web interface share stuff.   It's partially broken, and fairly complicated.  Most folks that find it too unweildy install their own version of Samba on Unslung, and use a diversion script to kill the Linksys one and use their own.
Nov 25 21:55:54 <tuv> i'm running debianslug. few minutes ago, i started seeing strange behavior. whenever i run almost any command, i get the following: '-bash: fork: Resource temporarily unavailable'. few commands do work though like pwd and kill, but not ps :(
Nov 25 21:56:35 <tuv> what could cause this?
Nov 25 21:58:48 <mwester> Out of memory.
Nov 25 21:59:18 <tuv> mwester, how can i tell which app. is using up my memory? no ps
Nov 25 21:59:50 <mwester> Might be a clue printed on the serial console.
Nov 25 22:00:23 <mwester> But without one of those, it's a guessing game.  Reboot, run ps in a loop, and watch for the app that starts grabbing stuff.  You do have a swap partition, right?
Nov 25 22:00:48 <tuv> i have a huge swap
Nov 25 22:01:14 <tuv> is there any bash built-in command that shows running processes?
Nov 25 22:01:25 <tuv> seems like bash built-ins still work
Nov 25 22:02:03 <tuv> even 'halt' doesn't work. might need to use the button
Nov 25 22:02:04 <mwester> poke about in /proc
Nov 25 22:02:18 <mwester> use things like "echo" instead of ls
Nov 25 22:02:33 <mwester> dunno if the "cat" command will work for you, though.
Nov 25 22:03:28 <tuv> how can i use echo instead of ls?
Nov 25 22:10:06 <tuv> cat doesn't work
Nov 25 22:11:45 <mwester> shell tricks:
Nov 25 22:11:53 <mwester> replace "ls" by: echo *
Nov 25 22:12:55 <tuv> is there a way to see running processes?
Nov 25 22:13:27 <Texture> (doesn't work on my Debian/NSLU2... it just echoes a *)
Nov 25 22:13:41 <tuv> Texture, are you running bash?
Nov 25 22:14:02 <tuv> it works here, but i still need to see the contents of some files
Nov 25 22:14:09 <Texture> root:x:0:0:root:/root:/bin/bash
Nov 25 22:14:12 <mwester> Will echo the wildcard chars if there's no match, so if echos "*" the dir is empty
Nov 25 22:14:35 <Texture> yupp! 
Nov 25 22:14:38 <Texture> <-- dumb
Nov 25 22:15:02 <Texture> it works.
Nov 25 22:15:10 <mwester> I don't know enough about the debian build; don't know what's builtin in bash...  does cat not work?
Nov 25 22:15:32 <tuv> mwester, no.. all i need is a little cute cat right now :/
Nov 25 22:16:09 <mwester> Only have a little dog here, but it doesn't know debian either ;)
Nov 25 22:16:54 <tuv> i don't think this situation has anything to do with debian. what resolves it on any other linux would resolve it on this debian
Nov 25 22:17:30 <tuv> so i need to find out the offending app. and kill it, all by using bash only.. how romantic
Nov 25 22:18:19 <mwester> Well sort of - the diff is what tools are available in the shell.  For example, with openslug and unslung, if you were running /bin/sh, it would be busybox and many more commands might be available.
Nov 25 22:18:45 <mwester> You can find the process #s of the running processed by "echo /proc" I think
Nov 25 22:18:53 <tuv> but if i were using bash on unslung i would be in the same situation
Nov 25 22:19:18 <tuv> mwester, yes, but i can't tell which process is each, since there is no 'cat'
Nov 25 22:19:53 <mwester> Yep.  You can start killing processes starting with the higher numbers and see if it gets better.
Nov 25 22:20:09 <tuv> mwester, that's what i just started doing :)
Nov 25 22:21:15 <tuv> yay.. problem solved. it was apcupsd
Nov 25 22:25:24 <mwester> ipkg remove cups
Nov 25 22:33:03 <tuv> mwester, that wasn't cups. it's apcupsd, an APC UPS monitor daemon
Nov 25 22:33:37 <tuv> killed it then started it again and everything is fine now. at lease i know the suspect next time this happens, if it does
Nov 25 22:34:58 <tuv> now i notice another thing. gkrellmd takes more cpu than it used to. about 30-40% cpu. way more than usual. i suspect a library upgrade to have caused this, but have no clue which library would that be. is this possible or should i look for something else?
Nov 25 22:49:06 <tuv> hmm.. i think it's libc6
**** ENDING LOGGING AT Sun Nov 26 02:59:57 2006