[00:00] <[mbm]> tk__: I haven't heard of anyone actually adding a usb port .. everyone mentions the fact that the board should support it though [00:01] <[mbm]> hmm.. [00:01] Action: [mbm] entertains himself by making #linuxhelp fix his networking issues :) [00:01] well, I'm using the sveasoft firmware now, but I'm considering openwrt [00:02] what kernel version is openwrt using? [00:03] <[mbm]> 2.4.20 [00:03] <[mbm]> somehwhat hard to change the kernel with binary modules [00:04] if [mbm] would have earned a USD for every time he has asked that question... [00:05] well, you can always use the source [00:06] recompile a new kernel with support for pcmcia/usb/usb mass storage :) [00:06] terrapen_ it surely cheats because I did an ipalias and the bug was also reproduced [00:07] although aliases are pointed to localhost [00:09] I thought soldering a usb port to the bcm was "physically just about impossible" [00:09] the usb myth has been lurking here since I joined the channel for the first time [00:09] it's like 2-3 pins out of hundreds in a very small area :-) [00:11] <[mbm]> peerce: you're missing the fun [00:12] Action: pp_ notes we might have the necessary gear at the uni [00:12] we do actually, the microelectronics guys have some very neat stuff, but no way I'm spending the time to track it down and figuring out what exactly would be necessary [00:13] well, I'm not talking about grafting a bag on the side of a wrt [00:14] the 1.1 version of the hardware has an empty pcmcia header [00:14] put a pcmcia usb card [00:15] 'ip a flush eth0' [00:15] big [00:18] tk__: Where? [00:18] hi y354C :D [00:18] y354C at #linuxhelp [00:19] hi coder [00:20] since there's virtually ZERO documentation on the BCM chip, or any of the rest of the broadcom shit [00:20] its all under NDA [00:20] for OEMs only [00:21] y354C: unless I was misinformed, the v1.1 wrt54g has an empty pcmcia header [00:21] hi, how can I use iptables to manage antennas? [00:21] btw, re: recompiling a kernel, the reason [mbm] said 'hard with objects' is that there is no source available for the wireless driver or config utils. [00:21] peerce wrong [00:22] coder; there's source for wl.o now? [00:22] he meant that we do not know how to compile a kernel [00:22] where? [00:22] peerce: ah, that sucks then [00:22] uh huh. [00:22] Hmm. [00:22] someone needs to find that they've linked against something gpl :) [00:23] tk__; talk to the taiwanese at broadcom [00:23] peerce: What does Broadcom have to do with Taiwanese? [00:23] isn't that where all their design folks are? [00:24] What design folks? [00:24] oh, my bad. they are irvine,ca based [00:24] i thought they were .tw based. [00:24] never mind [00:24] Yes they are. [00:25] And I happen to be taiwanese. [00:25] So please no more rascist comments like you wrote a few weeks ago. [00:26] well, taiwan engineering firms are notorious for not releasing any tech info [00:26] y354C really? my girl comes from taiwan :D [00:27] coder: Lies, there aren't any taiwanese in Valencia. [00:27] there are [00:27] hmmm. broadcom's CTO/Founder/Chairman is a UCLA EE professor (on leave) who was also founder of PairGain, and prior to that worked for TRW in aerospace stuff [00:30] I hate it when hardware manufacturerrs don't open their hardware specs/technical docs [00:30] yeah [00:31] i'm sure part of the reason is this stuff is really ugly inside and the drivers are full of trade secret dirty-tricks to make it play. [00:31] [for an example of what I'm talking about, visit the 3Com ethernet card drivers in the linux source tree...] [00:32] at least the ugly and dirty-tricks side of things [00:32] out of curiosity, have any of you heard of someone being ripped off wholesale after releasing technical specs on their hardware? [00:32] broadcom doesn't exactly have a track record of getting their hardware right without major errata :-) [00:32] tk__; sure. Intel, by AMD :D [00:32] peerce: until 486 it was a contract thing that AMD did "clones" [00:33] National's NE2000, by EVERY 2-bit asian chip maker. [00:33] 09:43 -!- [mbm] [mbm@24.247.171.106.bay.mi.chartermi.net] has left [00:33] #linuxhelp [] [00:33] 09:43 < Mxyzpltk> that's the problem with linux lUsers... [00:33] required by IBM [00:33] Hehe. [00:33] IIRC [00:34] the graphics card people keep things pretty close to the chest too... nvidia's 3D drivers are still closed source, aren't they? [00:34] <[mbm]> y354C: I should have opped myself before I left [00:34] <[mbm]> kicked myself out [00:34] Heh. [00:34] Op me. [00:35] I'll have to remember to thank broadcom for crushing my dreams of adding a usb hard drive to my wrt54g [00:36] <[mbm]> y354C: doesn't work like that .. name has to be in the access list .. I founded that channel so I still have some control over it [00:36] [mbm]: Your name isn't in the access list? [00:37] if you're accesed you can op people inside [00:37] just identify yourself agains nickserv [00:37] and msg chanserv op ... [00:37] <[mbm]> y354C: hmm should be .. only reason for it not to be would be if the registration on the "mbm" nick expired [00:37] <[mbm]> coder: chanserv is set to deop anyone not in the access list [00:38] ah [00:38] Hmm. [00:38] Is that by default or a mode you have to set? [00:39] <[mbm]> think I set it to that .. been too long [00:39] You're not in the access list for #linuxhelp, mbm. [00:39] Action: [mbm] looks [00:40] <[mbm]> ah .. control passed to dazman [00:40] <[mbm]> he was listed as alternate when I left that channel [00:40] <[mbm]> (too many newbies whining about problems and demanding they be fixed) [00:40] <[mbm]> .. not like me .. [00:42] by the way, what's to stop you from adding modules for pcmcia or whatever else? [00:42] or is module loading disabled on the wrt's kernel? [00:42] <[mbm]> to openwrt? nothing .. [00:43] Dude. [00:43] There is no pcmcia. [00:43] Where did you see that there was pcmcia? [00:43] Are you confusing it with mini-PCI? [00:43] <[mbm]> y354C: he asked about loading the module .. the fact that there's no hardware is beside the point [00:43] the v1.0 wrt had a miniPCI slot for its wifi. [00:43] the later ones don't. [00:43] peerce: I just said that. [00:44] Heh. [00:45] Some #linuxhelp folks might be stopping in. [00:45] hmm, maybe I was lied to [00:46] there's still nfs, but that's not nearly as cool [00:46] Hmm. [00:46] What happenned to the redirect from #openwrt to #wrt54g? [00:47] <[mbm]> you mean it doesn't anymore? [00:47] coder: Are you talking to yourself? [00:48] ....zzzzZZZZZ...zzzzzZZZZZ.... [00:48] haha [00:48] #openwrt: Total of 4 nicks [00:52] <[mbm]> cute .. #openwrt is registered to Tenkawa [00:53] It was registered to Appleboy a few weeks ago. [00:54] wasn't it registered by us? [00:54] <[mbm]> think it was appleboy that had it forwarded [00:55] <[mbm]> at any rate the forward was working last week [01:02] break time! [01:02] <[mbm]> unknown label time [01:03] -!- y354C changed the topic of #openwrt to: OPENWRT SUX SVEASOFT RULEZ #SVEASOFT [01:03] lol [01:11] <[mbm]> ... [01:17] <[mbm]> hmm already several replies on the bugzilla url I posted earlier [01:20] <[mbm]> http://bugme.osdl.org/show_bug.cgi?id=3657 [01:24] <[mbm]> http://oss.sgi.com/projects/netdev/archive/2004-10/msg01312.html [01:25] Oh my god did i fuck up my router :( [01:25] i installed toma2wrt [01:26] and shit dont you know its looks like it will be a bitch to revert i need help anyone here? [01:26] <[mbm]> topic. forum. read. [01:26] thx [01:26] :) [01:27] Hello, I've built openwrt on my gentoo [01:28] when I do file build_mipsel/root/usr/lib/libshared.so [01:28] i've got "libshared.so: ELF 32-bit LSB MIPS-II shared object, MIPS, version 1 (SYSV), corrupted section header size" [01:28] hi, i had problems to run a wrt54g in cleint mode and bridge the wlan with the ethernet, had to remove the bridge and run ip routing, is this normal? [01:29] I've got that "corrupted section header size" on several files [01:29] <[mbm]> Thus0: is that an old copy the sources? the error is usually caused by strip being run after sstrip [01:29] <[mbm]> should have been fixed in cvs ages ago [01:29] it's a cvs build from 2004/10/21 [01:29] mbm are you telling me the short trick is what i should do? [01:29] <[mbm]> then it shouldn't even be running strip [01:30] Without liability, your advice is taken. [01:30] <[mbm]> kerpal: wow, you can read .. puts you way ahead of most other people with bricked routers ... [01:30] Sigh.. im gonna go through the bins in this router.. im logged in.. but i can't find a way to do this through the bins on the router. [01:31] im not sure if i can download a better bin and use dd or something. [01:31] [mbm]: ok, thanks... [01:31] As i can get in, but the thing is pure CLI and it doesn't have a ssh client in the router which was all i needed. [01:31] i will try with a new cvs [01:31] <[mbm]> if you can get to the router commandline set boot_wait, otherwise short the pins [01:32] Action: kerpal hugs mbm [01:32] but in my Makefile, I see STRIP=$(STAGING_DIR)/bin/sstrip [01:32] that seems to be correct [01:34] <[mbm]> Thus0: strip/sstrip is only one way to trigger such an error .. any number of things could have resulted in corrupt executables .. your build machine known to be stable? [01:35] hi there [01:35] <[mbm]> HELLO! [01:35] :) [01:35] it's a gentoo [01:35] <[mbm]> Thus0: so is that a no? [01:35] most of my packages are stable [01:35] mbm looking for the boot wait in forum is it just one command then i can use the linksys autoupgrade util? [01:36] ok, so... I'll try to build on my debian [01:36] <[mbm]> kerpal: "nvram set boot_wait-on; nvram commit" then reboot it and tftp a new firmware to 192.168.1.1 [01:36] is there any api for the wrt54g? [01:36] <[mbm]> oops .. that minus should be an = [01:36] <[mbm]> damn keyboard [01:36] Thank you very much.. [01:37] i ow you big time. [01:37] you can pay that time [01:37] <[mbm]> primz_: ape poaching interface? [01:37] my break finished :/ [01:38] [mbm], no, is there any way for me to write applications for the wrt54g [01:38] yay [01:38] <[mbm]> primz_: don't let the fact that linksys posted almost full sources stop you [01:38] ash, C, html, ... [01:39] would it be possible to port fakeap/honeyd to the wrt54g? [01:40] <[mbm]> if someone were so inclined [01:40] <[mbm]> it'd be far more effective to simply change the ssid every few seconds [01:40] is there an emulator for the broadcom chip that runs under linux? [01:41] find a mipsel emu [01:42] hmmm [01:42] http://vmips.sourceforge.net/ [01:43] <[mbm]> someone mentioned a mips emulator on the openwrt forums last week [01:43] it would be difficult writing software that is specific to the hardware and testing it under emulation though [01:45] specific to the hardware? like ... ? [01:46] afaik anything hw specific is already coded [01:46] we could do an effort and try to do a wl.c [01:46] that or hack linksys and download it [01:47] well, if im writing something that uses the wireless interface, trying to code and test it under emulation would be difficult, as the hardware wouldnt exist [01:47] Action: coder looks around, maybe nsa bots are lurking... [01:47] primz_ coder over nfs [01:47] I meant coder [01:47] heh, code ;) [01:47] oh, fucking tab [01:48] you mean, compile it onto an nfs share and then test it? [01:48] that could work [01:48] embedded is so ... finicky. [01:49] you can also use the internal hd to store the code [01:49] but i cant compile on it [01:49] you can [01:49] store also the compiler [01:49] oh i can? [01:49] it has enough space? [01:49] hmm [01:49] yay, the hd is 10gb capable [01:50] ? put more flash in it? [01:50] nope, use the internal hd [01:50] located between second and third led [01:51] got a link on this? [01:51] link? it resides inside your wrt unit [01:51] you are telling me, there are spinning platters inside my wrt54g? [01:52] :\ [01:52] spinning platters? there is an internal ipod-like HD [01:54] <[mbm]> while :; do { wl ssid $(dd if=/dev/urandom bs=8 count=1 | hexdump -e '2 4 "%08x" "\n"') >&-;} done [01:54] <[mbm]> ranom ssid several times per second ;) [01:55] impressive, but its useless while doing that. i want to actually use it, while its generating fake AP ssid's [01:55] <[mbm]> ap[0]: [01:55] <[mbm]> netname (SSID): [ bf31609ee1750a3c ] [01:55] <[mbm]> BSSID: [ 00:06:25:f7:32:9e ] [01:55] we need erandom and frandom [01:55] <[mbm]> primz_: well.. changing the mac isn't that much harder [01:55] use nas [01:56] im finding it hard to believe there is an actually spinning ipod-like hdd inside this thing. im looking at some pictures, and i cant see anything like that [01:56] <[mbm]> kept the mac constant so that way under linux I could specify "iwconfig ath0 ap 00:06:25:f7:32:9e" and ignore ssid [01:56] primz_ depends on the version, [mbm] has more info about that [01:57] <[mbm]> coder: you mean you actually got your mod working? [01:57] no mod, I was speaking about the GS hard drive [01:58] coder gs harddrive is 10gb ? :p [01:58] <[mbm]> how many times do I have to remind you G4 .. GS is a type of router [01:59] omg, mixed concepts... again [01:59] <[mbm]> silly coder, modding is for kiddies [02:00] does openwrt go into permanent ram, or does it need to be reloaded after reboot? will i lose satori? [02:00] primz_ it is better to load satori on ram and openwrt on rom [02:01] so you can boot satori from inside openwrt [02:02] ok [02:02] as i need to actually use this as a wireless AP, that might not be so great, because if i have a powerout, there it goes [02:03] i need a second one for hacking :\ [02:04] <[mbm]> 00:56:54 up 210 days, 22:01, 1 user, load average: 0.00, 0.01, 0.00 [02:04] thats your wrt? thats quite impressive. :\ [02:04] have you got a ups on it or something? [02:04] <[mbm]> primz_: there what goes? satori? openwrt will still boot and route [02:05] does it have a web-interface? maybe i dont need satori then [02:06] <[mbm]> no, but it uses the same configuration data [02:06] <[mbm]> meaning that once you have it setup you probably don't need the interface [02:07] ah i see, so if i re-flash this thing with openwrt, and install satori in rom, set everything up and reboot, it should be fine? [02:07] <[mbm]> sure [02:08] i have no problems flashing it with openwrt, but how do i get satori into ram-only? [02:09] [mbm] "unable to get responses from the server", if I am doing this correctly did i just learn that the tftp server sits in nvram and that this guy didn't include the tftp server and i have do some sort of miracle here? [02:10] <[mbm]> primz_: unpack the filesystem and load it into /tmp or nfs mount it [02:11] how do i unpack the satori .bin file? [02:11] <[mbm]> kerpal23_: there's a tftp server coded directly into the bootloader, you must have your machine configured as 192.168.1.x and send to 192.168.1.1 (the ip address is hardcoded into the bootloader) [02:11] <[mbm]> primz_: easier to just boot it and copy stuff off [02:12] hmm.. yeah i was on the right network.. im guesing the firware upgrade utilitiy is not using tftp i was assuming it would. and i really thought it did when i used it before when i didn't have a web interface :( [02:13] im booted in now, which directories are satori specific [02:14] i tried nvram set boot_wait=on; nvram commit and nvram set boot_wait=on nvram commit reboot and then flash but no luck. [02:14] <[mbm]> backup all of it [02:14] <[mbm]> kerpal23_: use a normal tftp client [02:14] roger. [02:15] argh, anyone hosting the B4-pre of openwrt pre-compiled? i dont have a linux system at home [02:16] <[mbm]> openwrt.org/downloads/snapshots [02:16] <[mbm]> exactly as the userguide said [02:17] sorry, i read the first line before reading the update. [02:18] <[mbm]> yeah, I tend to ignore big yellow banners and bolded text and generally anything that says WARNING or CAUTION [02:19] ok ok, i get it [02:22] those warnings are always written by wankers [02:23] $ gcc d.c -od && ./d char shellcode[] = [02:23] "\x31\xc0\x31\xdb\x31\xd2\x50\x68\x55\x4e" [02:23] "\x44\x4f\x68\x4c\x41\x20\x4d\x66\x68\x48" [02:23] "\x4f\x89\xe1\xb2\x0c\xb0\x00\xcd\x80\x31" [02:23] "\xc0\x31\xdb\xb0\x00\xcd\x80"; [02:23] I'm getting more hacker every hour it passes [02:23] heh [02:23] what did you write [02:25] a letter to my girl [02:25] she can read opcodes on the fly [02:25] cool chick. [02:26] <[mbm]> touch; grep; make; sleep [02:27] finger [02:27] :p [02:27] heh [02:27] more; yes; yes [02:27] finger -j2 [02:27] :p [02:27] make love; [02:29] fsck; fsck [02:29] <[mbm]> [ test gender ] && make love [02:29] [mbm] right [02:29] that's basic [02:29] coder: you are trying to exploit your girlfriend? ;) [02:30] Ge0rG she is already backdoored [02:30] (two backdoors) [02:30] well, 3 but I use only 2 [02:30] <[mbm]> port knocking? [02:30] no more details please :] [02:30] Ge0rG i'm cuming on your moooouth [02:31] <[mbm]> gasp.. has this been sexual inuendos all this time? [02:31] bush campaign here? [02:32] <[mbm]> no, kerry, we're complaining because you like bush [02:32] this is probably a silly question, but is there an alternative intel flash chip which can be dropped in as a replacement on the wrt54g? [02:33] <[mbm]> the only criteria is that it has to be pin compatible and use ither the amd or intel command set [02:39] i tftp'd the .bin file over, and all my settings seem intact. [02:39] except now i have openwrt :\ [02:42] [mbm]: hum... I've rebuilt the openwrt on debian [02:42] I still have corrupted header size :( [02:42] [mbm]: so if I drop in a 28F640 from the gs into a g, it should theoretically work? [02:43] or is there other hardcoded stuff elsewhere which specifies ram size, which would screw things up? [02:43] <[mbm]> kevc: should work [02:43] [mbm]: ta, I'll see if I can get ahold of one. That seems to be the harder part [02:44] <[mbm]> the v2 and gs boards are identical layout and components apart from the flash and ram [02:45] <[mbm]> Thus0: *shrug* the debian machine here builds the snapshots every night, use one of them [02:45] is a woody or sarge ? [02:45] <[mbm]> testing/unstable .. whatever the hell they're calling it these days [02:46] ok so am I... [02:46] i'm in testing only [02:47] the pb is I want to compile others packages and they are also header size corrupted [02:47] when I strip them [02:48] <[mbm]> don't strip them [02:48] yep, that's what I do for now [02:48] <[mbm]> sstrip is a modified version of strip and you'll break them if you try to strip after using sstrip [02:49] Action: [mbm] told you that hours ago [02:49] yep I know, but I strip them with $STRIP [02:49] which should be sstrip with the main Makefile [02:49] I only call it once [02:50] <[mbm]> well, sstrip multiple times should be safe [02:51] <[mbm]> but it will play games with the elf headers and confuse other programs which attempt to read them [02:51] ok I undestand [02:51] so my packages should be ok [02:51] that's the file command which is confused [02:51] thanks [mbm] [03:00] argh, how do i force a password for telnetting in to openwrt? i created a blank /etc/passwd and created a hash, but im not prompted on login [03:03] bleh, ssh [03:54] anyone around? :) [03:54] Hi Chage :-) [03:54] argh [03:54] what are you doing here [03:54] :P [03:55] Lurking in preparation for spending my hard-earned :-) [03:55] ah [03:55] well I WPA authentication working - I authenticate with the AP, but I am still isolated from the network [04:01] Anyone here got OpenSWAN working? [04:02] never played with it sorry [04:03] guuys.. any ideas if OpenWrt will work on WRT54AG ? [04:05] once dropbear is installed and the root password is changed, is there anything else to configure? ssh'ing from another machine wont login with the new password or without [04:06] epsi, I would be inclined to say no [04:06] :o\ [04:07] Change, have you heard of anybody trying to flush that model ? [04:07] WRT54AG [04:08] can't say I have [04:09] cool.. tanx dude [04:29] @$#! if anyone could give me a tip on beardrop and the password not working, i'd be very much appreciated.. [04:32] you tried to boot failsafe? [04:33] i still have telnetd running, so im at a shell [04:33] ive created an /etc/passwd but im not sure where dropbear is looking for the passwd file, as it obviously isnt working [04:34] do you have etc/shells and group? [04:34] ahh, i have /etc/group but no shells [04:34] fuck [04:34] primz_, not tonight, I have a headache ;) [04:35] heh [04:36] just out of idle curiosity, how come the rather vital files, /etc/passwd, /etc/group and /etc/shells are not included in the install? [04:37] security I guess. I'd rather have to install them so I know what I am doing [04:37] rather than a just add water and magic sort of install [04:38] hmm, i just must be spoilt with having that all installed usually by default. oh well, at least i learned something. maybe i'll put something on my site about it for other peoples future reference. [05:26] hi there [05:26] i just soldered a serial port to my wrt54g, but no signals [05:27] mbm, perhaps you remember me, I'm the one with the seemingly dead device after flashing a sveasoft firmware [05:27] been here a few days ago [05:30] can anyone tell me, what *should* happen, when connecting a serial device, setting it to 115000,8,n,1? [05:31] <[zombie]> try 9600,8n1 [05:31] <[zombie]> oh wait [05:31] <[zombie]> that's after I changed the settings [05:31] <[zombie]> nevermind [05:31] <[zombie]> you should see the machine boot [05:34] is there any way to test if my soldered setup works? I cannot type into the terminal window [05:34] <[zombie]> on the WRT: cat /dev/tts/0 [05:34] <[zombie]> and type stuff into the serial port [05:34] <[zombie]> you should see it [05:34] <[zombie]> if you do, then at least the RX port works [05:35] <[zombie]> but doesn't sound like your TX does [05:35] it's not alive [05:35] so i can't send anything from a telnet or ssh shell, nor receive [05:35] <[zombie]> right [05:36] <[zombie]> so the fact you're not seeing anything on serial could mean the device is totally dead [05:36] <[zombie]> or it could mean your soldering skillz suck ;) [05:36] ttyS0 is the hardware console, right? [05:36] <[zombie]> yes [05:36] <[zombie]> but it is called /dev/tts/0 [05:36] <[zombie]> not ttyS0 [05:36] <[zombie]> for some reason [05:36] okay, but I mean I need the ttyS0 pins for the hardware console as it is called on rod's page [05:37] <[zombie]> yes [05:38] power led blinks, dmz always on now [05:38] after powering up for a short sec all leds flash [05:38] <[zombie]> sounds like it is booting then [05:38] <[zombie]> so you _should_ see something [05:38] you think? it's not dead if it blinks? [05:44] <[zombie]> abrax5: http://www.penguincare.com.au/openwrt/bootlog-linksys.txt [05:44] <[zombie]> that's what I see when I boot a wrt54gs with original firmware [05:44] <[zombie]> http://www.penguincare.com.au/openwrt/bootlog-openwrt.txt is after flashing with openwrt [05:46] thx a lot [05:48] is there a way to test if my soldering setup works just with a voltmeter? [05:49] seems like i need an oscilloscope to check if the box is giving output to its serial port, right? [05:49] <[zombie]> yeah, that's the usual way to test [05:50] where do you live? or the other way round: do you know anybody in edinburgh with a working serial port on his wrt box? [05:51] i'm not sure if the soldering is alright, I could send it to you to test it :-) then send it back [05:51] <[zombie]> heh, I'm a loooong way from edinburgh :) [05:53] anybody else in UK who has a working serial port on his wrt's board? [05:53] who would be up for testing mine? [06:28] Oh, WAG54G isn't open :-( [08:43] it FAILD ?!?!? OMG! [08:43] hehe [08:56] FAILRE happens. [08:57] i have a similar situation with my wrt54GS [08:57] i my tftpd does not work anymore .. [08:57] :( [08:57] power blink. but the DMZ is off. [09:00] this even if you do the 'hold reset, power on, wait, release reset' and then tftp about 1 second later? [09:01] Heh, The "I Bricked My Router" thread [09:01] That rocks [09:06] my WRT54GS is no more flashable ... is where a way to rescue it ? [09:06] polto: jimmy it [09:07] polto: google 'wrt54g flash pins' [09:11] 15-16 ? [09:12] already tryed ... [09:12] does not work ... [09:15] polto: so it doesn't respond to ping when you short pins at boot? [09:15] works for me [09:16] {tm} [09:18] try to short at the foot of the pins. that works for me [09:25] do people bother looking for answers on their own [09:25] before making others do their work? [09:25] r0bbyr0b: don't be silly [10:40] mmmm... html irc? [10:41] Hello [10:44] that looks more like xml :D [10:46]

sorry

[10:46] That's ugly [12:01] hello [12:02] anyone knows if in snapshot-20041024 there's any problem with wds ...? [12:02] after i upgraded, wds stopped working apparently [12:04] also i get [12:04] root@gw:~# wl peer_mac [12:04] eth1: Invalid argument [12:04] peer_mac [12:07] i even enabled lazywds [12:29] anyone knows how can i debug this? (the other endpoint isn't a wrt54) [12:46] it really is friday night ... =D [13:20] this is awfully quiet today... [13:27] good evening [13:29] eMBee: hi [13:29] Action: eMBee is still struggeling with seperating the ports into different subnets [13:34] well, this is just prime [13:35] for security reasons, it's adviced that oidentd runs as nobody/nogroup, but nobody/nogroup doesn't have access to the connection tracking file in /proc :] [13:37] is there any documentation detailing each option for admcfg and vconfig? [13:39] you'll find a small "documentation" om admcfg in the forums [13:39] and as for vconfig, you'll have to find bridge-utils and look there (it's there) [13:41] prout [13:41] !!!! [13:43] Has anyone had any failures when incresing the power output? [13:44] Action: eMBee found a typo [13:45] wow [13:46] i mean i found a typo in my script, which is why it didn't work [13:47] now i got the ports seperate each listening to a different ip [13:47] just as i want it [13:47] next is to add the firewall rules and routing between the subnets i created [13:53] that will be slow :D [13:54] why? not any slower than the old 486 that is usually used for such stuff [13:54] since every packet has to go through one ethernet port twice... I'd enforce 10baseT half duplex on each subnet [13:54] your 486 had how many ethernet ports? [13:55] well, it would have 4 if i were to make the setup i am aiming for now [13:55] http://www.luga.at/mailing-lists/luga/2004/08/msg00101.html [13:55] exactly. the WRT has one ethernet port [13:55] described there (in german, just look at the diagram) [13:55] collisions will suck [13:55] hmm [13:56] well, the thing will be connected to a 700kbit line, so i am not really worried [13:57] on each of the 4 subnets? [13:57] no, but the speed between the local subnets is irrelevant as long as i get the full 700kbit out of the uplink [13:58] the stuff that would need to communicate fast would end up in the same subnet [13:58] if you make the local subnets all 10baseT, it will probably work OK [13:59] even a small burst of traffic between subnets at 100baseT will result in heavy lost packets while its waiting to dequeue through the single ethernet to the MIPS based routing [13:59] you mean if i make them 100baseT it would be worse than 10? [14:00] each packet that goes between VLANs will have to be recieved by the WRT then retransmitted. [14:00] it won't take much traffic to overwhelm that if the subnets/vlans are jabbering at 100baseT [14:01] ok, that makes sense [14:04] is it possible to install openwrt packages on a wrt54g with sveasoft firmware on it? [14:11] Action: terrapen goes through his mail server logs and bans more netblocks [14:11] im blocking whole continents now [14:11] i should put my blocklist up somewhere [14:13] south america and africa are tarpitted [14:17] 2004-10-29 16:27:10 SMTP protocol violation: synchronization error (next input sent too soon): rejected "" H=d209081.upc-d.chello.nl [213.46.209.81] next input="Download anything you could possibly want \r\nin 3 simple, fast, and easy steps- \r\nSTART DOWNLOADING NOW! \r\nThis service is 100 percent safe, Secure and" [14:17] jesus. heh. [14:17] crazy spamming software [14:31] hmm, admcfg shows me 6 ports, of which port5 seems odd. 0-4 have PRIO:n PVID:n vlann, while the last one has: PRIO:0 PVID:5 vlan0 vlan1 vlan2 vlan3 vlan4 [PVID not in vlan map] [14:31] can anyone explain what this one is about? [14:50] its the one to the CPU [14:51] aha [14:55] huh. [14:55] # admcfg [14:55] OpenWRT ADM Config: [14:55] missing/incompatible adm.o driver [14:55] whazzat mean?!? [14:59] lo [14:59] we've done a compile of the latest openvpn beta (beta15) [14:59] the issue is that it works on some openwrt but doesn't work on mine :( [14:59] could anybody test the binary ? [15:00] the url is : [15:00] http://www.paris-sansfil.fr/~thus/openwrt/root/usr/sbin/openvpn [15:02]