--- Day changed --- Log opened Wed Jun 11 00:00:01 2003 01:44 < green> Run: this is libc. If you are using gcc3, there are problems with backtraces. If you'd use gcc 2.95.x, you will gt longer one (not sure why ou need it in this case) 02:34 < BB> morn 02:36 -!- BB1 [~chris@217.79.122.210] has quit [ozone.oftc.net quark.oftc.net] 02:36 -!- green [green@linuxhacker.ru] has quit [ozone.oftc.net quark.oftc.net] 02:38 -!- ElectricElf [david@elf.noc.oftc.net] has quit [ozone.oftc.net ionosphere.oftc.net] 02:38 -!- ElectricElf [david@CPE0040f6b429d8-CM400049226055.cpe.net.cable.rogers.com] has joined #uml 02:39 -!- ElectricElf is now known as Guest72 02:40 -!- BB1 [~chris@217.79.122.210] has joined #uml 02:40 -!- green [green@linuxhacker.ru] has joined #uml 02:49 -!- caker [~null@68.52.196.167] has quit [Ping timeout: 488 seconds] 02:50 -!- Guest72 is now known as ElectricElf 03:48 < BB> Hmm telewest complained as i "swore" shock horror! 04:21 -!- ElectricElf [david@elf.noc.oftc.net] has quit [uranium.oftc.net ozone.oftc.net] 04:41 -!- caker [~null@pcp507591pcs.nash01.tn.comcast.net] has joined #uml 05:10 < Run> green: You here? 05:31 < green> sure 05:46 -!- trom [dankbar@pD9E823DE.dip.t-dialin.net] has joined #uml 05:47 < trom> hi 05:47 < trom> I can not get into the debug mode: 05:47 < trom> (none):/usr/src/uml2.4.20-6/linux-2.4.20# ./linux ubd0=/root/Debian-3.0r0.ext2 debug con=null con0=fd:0,fd:1 05:47 < trom> tracing thread pid = 23100 05:48 < trom> and the xterm shortly blinks and then disapears. 05:49 < trom> (none):/usr/src/uml2.4.20-6/linux-2.4.20# make dep ARCH=um && make linux ARCH=um 05:49 < trom> can anybody help me? 05:55 < green> do you have gdb in path? 06:11 < trom> It is functioning with kernel 2.4.18 and 2.4.18-1, but with all others no. I dont understand your question, but i can invoke gdb by symply typing gdb. 06:11 < trom> simply 06:22 -!- avoozzzl is now known as avoozl 06:22 < green> have you compiled with ptproxy support? 06:25 < trom> yes, i haven't touched anything in the .config. also tried mode=tt to turn of skas 06:28 < green> hm, interesting. And you have port-helper too, I suppose? 06:31 < trom> before i had a errormessage with porthelper and then somebody adviced me to install uml-utilities, with what it went. 06:31 < trom> an 06:49 -!- Run [carlo@node-d-5886.a2000.nl] has quit [Remote host closed the connection] 07:07 -!- Run [carlo@node-d-5886.a2000.nl] has joined #uml 07:07 < Run> That was not good :/ 07:07 < Run> I had to reboot to host... 07:10 -!- trom [dankbar@pD9E823DE.dip.t-dialin.net] has quit [Ping timeout: 492 seconds] 07:20 -!- trom [dankbar@pD9E819EC.dip.t-dialin.net] has joined #uml 07:21 < trom> which version of uml-utilities do you have. Does debug-mode of newer kernels not run with debian-latest-version 20020415-3? 07:24 < green> it should, I think. but 20020415 looks pretty old 07:24 < green> btw does it run if you do not supply "devug" option? 07:24 < trom> yes , but.. 07:25 < trom> the two other expected xterm do not appear. 07:26 < trom> the two other xterms 07:28 < green> sure, you disabled those with "con=null" 07:29 < Run> Hi green. 07:30 < green> hi Run 07:30 < Run> green: Could it be that 2.95 results in a longer bt because it uses stabs, while gcc 3.x uses dwarf by default? 07:32 < Run> Question: I have mounted my host fs (read-only). When the uml kernel locked up completely (and I couldn't do anything anymore, not even with uml_mconsole; hell, not even 'kill -9' worked) it turned out that the kernel process was hanging on a disk-io block. 07:32 < green> Run: I do not know. Try to experiment 07:33 < Run> Therefore, the root_fs was locked - and I couldn't reboot the uml either. 07:33 < Run> I had to reboot the host. 07:33 < green> well, you need to find out why your host kernel blocked on io 07:33 < Run> However - then the host couldn't umount its file systems because they were still 'busy'! 07:34 < green> sure. files still were opened 07:34 < Run> In other words: using hostfs is dangerous :/ 07:35 < Run> I think that if the uml kernel locks up - there should be a way to get rid of any file system locks or whatever - so at least the host isn't affected. 07:37 < green> Run: well, it as dangerous as using cat. If your kernel blocks on reading files, what do you expect to happen after this? 07:37 < trom> sure, you disabled those with "con=null" 07:37 < trom> Hi green. 07:37 < trom> hi Run 07:37 < trom> green: Could it be that 2.95 results in a lon 07:38 < green> uml is totally userspace application and therefore if you made the host to get in some inconsistent state with uml. the host is to blame 07:38 < trom> how do i enable them then? 07:39 < green> trom: pass con=term 07:39 < green> con=xterm even 07:40 < Run> green: Then how does hostfs work? It is totally safe to read/write to a hostfs mounted fs from uml while also read/writing to it from the host itself? 07:40 < trom> (none):/usr/src/uml2.4.20-6/linux-2.4.20# ./linux ubd0=/root/Debian-3.0r0.ext2 debug con=xterm 07:40 < trom> tracing thread pid = 23198 07:41 < Run> Safe for uml I mean - it will never lock up because of that? 07:41 < trom> blinking and nothing 07:42 < green> Run: you can get some data mismatches, but should be fine otherwise, I think I do not use hostfs myself 07:42 < green> trom: so what if you will not pass this 'debug' option? 07:43 < trom> ... 07:46 < trom> ./linux ubd0=/root/Debian-3.0r0.ext2 con=xterm con0=fd:0,fd:1 07:46 < trom> blinking and nothing 07:47 < trom> halt 07:47 < trom> INIT: Pid 122 [id 1] seems to hang 07:47 < trom> INIT: Pid 123 [id 2] seems to hang 07:47 < green> hm 07:47 < trom> (: 07:47 < green> are these messages from real init on the host? 07:49 < trom> once again I do not understand, host?, to shutdown uml i used halt, then that message came. 07:50 < Run> trom: look in /var/log/messages on the host. Is it there too? 07:50 < green> hm. So if it is "blinking and nothing", where do you type "halt" ? ;) 07:51 < trom> in uml 07:51 < trom> ah 07:52 < trom> there are popping up to windows, sure? But the window one is comming from offers also a login. 07:52 < trom> two 07:53 < Run> in english? 07:57 < trom> at least 2.4.18-1, which I know starts with 3 console windows, 2 of them pop up, and the other one is the cosole window from which one invoked umllinux.? 07:57 < Run> yes 07:58 < trom> k? 07:59 < trom> I can not see this 2 windows, I can not see the debugwindow in debugmode. 08:00 * Run uses skas3 ... sounded easier to me. 08:00 < Run> That works. 08:01 < trom> ? 08:02 < Run> This f*ing code I write has more #if's then normal code it seems - horrible.... 08:03 < Run> ~/c++/libcwd>cat `s` | egrep '^#if|^#el|^#endif' | wc --lines 08:03 < Run> 2412 08:03 < Run> So there. 08:03 < Run> Well, just 10% actually... 08:03 < Run> ~/c++/libcwd>cat `s` | wc --lines 08:03 < Run> 24221 08:03 < Run> :) 08:23 < trom> Run: nothing seen in var/log/messages 08:24 < Run> trom: ... I think you should recompile the uml kernel then *with* the hang check timer. 08:24 < Run> errr 08:24 < Run> *without* 08:26 < trom> what flag - option is it? 08:27 < Run> CONFIG_HANGCHECK_TIMER 08:33 < trom> It does not exist in my .config anyway. 08:39 < Run> Well, I don't use 2.4 kernels anymore. 10:41 -!- smcavoy [~smcavoy@207.61.160.163] has joined #uml 10:48 -!- smcavoy [~smcavoy@207.61.160.163] has quit [Remote host closed the connection] 11:15 -!- radical [~kernet@202.88.184.39] has joined #uml 12:50 -!- war- [war@oper.us] has joined #uml 13:23 < trom> hi 13:27 < trom> have i got on some peoples' nerves this afternoon? I applied the patch ones again, and now patching with the new uml-utilities installed, it runs * thx 13:44 -!- trom [dankbar@pD9E819EC.dip.t-dialin.net] has quit [Ping timeout: 492 seconds] 13:52 -!- albino [~albino@neuf.dorms.usu.edu] has quit [Ping timeout: 496 seconds] 13:52 -!- trom [dankbar@pD9E83E33.dip.t-dialin.net] has joined #uml 14:24 < Getty> more and more user on #debian.de on IRCnet are using UML ;) 14:24 < Getty> (and always this running gag with uml_net) 14:25 < Getty> is someone of you "Matt Zimmerman"? 14:29 -!- radical [~kernet@202.88.184.39] has quit [Remote host closed the connection] 15:46 < Getty> uml-utilities maintainer has answered 17:11 -!- smcavoy [~smcavoy@grimlock.drive-megawheels.net] has joined #uml 17:13 < avoozl> matt zimmerman is a crypto expert isnt he 17:16 < smcavoy> I'm looking at setting up several debian 3.0 umls using bridging for network access, rootstrap seems to want to do all that arp proxying stuff... anyway around it? 17:18 < ljlane> edit the rootstrap configuration file 17:18 < ljlane> or just don't use rootstrap 18:04 -!- smcavoy [~smcavoy@grimlock.drive-megawheels.net] has quit [Quit: ACK!] 18:17 -!- trom [dankbar@pD9E83E33.dip.t-dialin.net] has quit [Remote host closed the connection] 18:20 -!- avoozl is now known as avoozzzl 18:55 -!- ElectricElf [david@elf.noc.oftc.net] has joined #uml --- Log closed Wed Jun 11 19:40:23 2003 --- Log opened Wed Jun 11 23:23:51 2003 --- Log closed Wed Jun 11 23:25:12 2003 --- Log opened Wed Jun 11 23:25:35 2003 --- Log closed Thu Jun 12 00:00:00 2003