--- Day changed --- Log opened Tue Jul 08 00:00:03 2003 00:05 < Pahan> caker: What is the URL of your hosting thingie again? 00:05 < caker> http://www.linode.com/ 00:11 < Pahan> Is there a local Debian repository? 00:11 -!- ElectricElf [~david@elf.noc.oftc.net] has joined #uml 00:11 < caker> no locals yet - you're only the 3rd person to ask 00:23 -!- Dave\\\ [dave@ool-4352e135.dyn.optonline.net] has left #uml [] 00:23 -!- Dave\\\ [dave@ool-4352e135.dyn.optonline.net] has joined #uml 00:24 < Dave\\\> is there a local Debian repository? 00:24 < Lathiat> perosn number 4..? 00:24 < Lathiat> caker: setup apt-cache 00:24 < caker> heh - all together now 00:25 < Lathiat> IS THERE A LOCAL DEBIAN REPOSITORY!? 00:25 < caker> what's a debian? 00:25 < caker> j/k 00:26 < Lathiat> I heard they plan dropping the b from the next stable release 00:27 < Lathiat> caker: know whyat apt-cache is? 00:28 < caker> Lathiat: why don't you tell me what it's useful for, since you're going some where with this :-) 00:29 < Lathiat> wel basically 00:30 < Lathiat> u add debian repositories to it, and then u add your apt sources.list lines to the apt-cache 00:30 < Lathiat> then when you equest a package it downloads it, sends it ot the client, adn then caches it for when the next perosna sks for it 00:30 -!- idcmp [~idcmp@CPE0050dafe87e7-CM013359904800.cpe.net.cable.rogers.com] has joined #uml 00:30 < caker> excellent! 00:30 < Lathiat> that way once uve done a few installs and installed most usefull programs, u have most debian packages onsite, without mirroring every bludy package ull never use 00:30 < idcmp> has anyone ever had a problem where the user that uml is run as leaks into the instance? 00:31 < idcmp> getuid32() = 0 00:31 < idcmp> geteuid32() = 500 00:31 < idcmp> write(2, "mount: only root can do that", 28mount: only root can do that) = 28 00:31 < Lathiat> what do u mean? 00:32 < idcmp> init-2.05a# mount -av 00:32 < idcmp> mount: only root can do that 00:35 < idcmp> on the host, uml is running as uid 500 00:35 < idcmp> and somehow in the instance geteuid32 knows this :-/ 00:37 < caker> using tmpfs? passing real devices to uml? 00:37 < caker> tmpfs = hostfs 00:38 < idcmp> [root@parsan uml]# su - geist -c 'TMP=/uml/tmp ; /uml/linux-2.4.20-4um mem=256M ubd0=root_fs_cow,root_fs init=/bin/bash ubd1=swapfs fakehd con0=fd:0,fd:1 con=pty eth0=tuntap,umgeist umid=geist' 00:38 < idcmp> Nope. 00:38 < caker> what's in fstab? 00:39 < idcmp> /dev/hd/disc0/disc / ext3 noatime 0 1 00:39 < idcmp> /dev/hd/disc1/disc swap swap defaults 0 0 00:39 < idcmp> proc /proc proc defaults 0 0 00:39 < caker> ok i'm not a devfs expert but -- disc0 and disc1? 00:40 < idcmp> init-2.05a# df 00:40 < idcmp> Filesystem 1k-blocks Used Available Use% Mounted on 00:40 < idcmp> proc 2045026 1231066 729585 63% /proc 00:40 < idcmp> /dev/hd/disc0/disc 2045026 1231066 729585 63% / 00:40 < caker> what does "ls -ld /dev/hd/disc0/disc" say? 00:40 < idcmp> brw------- 1 root root 98, 0 Dec 31 1969 /dev/hd/disc0/disc 00:40 < caker> how about /dev/hd/disc1/disc ? 00:41 < idcmp> brw------- 1 root root 98, 16 Dec 31 1969 /dev/hd/disc1/disc 00:41 < caker> hmm 00:41 < idcmp> this was working a second ago 00:41 < idcmp> and the other three instances seem okay.. :-/ 00:41 < caker> is the fs file open already on the host? poweroff and check w/ lsof? 00:42 < caker> (poweroff the UML) 00:45 < Lathiat> caker: thats the coolest name, wish i hadve thought of that 00:45 < caker> Lathiat: which? 00:47 < idcmp> hmm no.. i'll tear down all the instances and see what's left.. gimme a sec 00:48 -!- kara [~kara@varstation.lanscape.net] has quit [Remote host closed the connection] 00:49 < Lathiat> linode... 00:49 < caker> ahh - yeah, my alternatives were pretty sad ... It took weeks to finally come up with that -- duh ;-) 00:50 < idcmp> nope.. somehow geteuid32() is returning the uid from the host 00:50 < caker> where are you stracing from? 00:50 < idcmp> i pass init=/bin/bash then do strace -ff /bin/mount -av in the instance 00:52 -!- silug [~steve@osiris.silug.org] has quit [Quit: ircII EPIC4-1.0.1 -- Are we there yet?] 00:57 < idcmp> hmm no 00:57 < idcmp> it's always 500 01:01 < caker> what are the permissions on the fs files? :-) 01:01 < idcmp> 500. 01:01 < idcmp> I just did an ls -l 01:01 < idcmp> everything is 500.500 01:01 < Lathiat> ah out-smarted 01:01 < Lathiat> is the UM lactualy running as 500 01:01 < Lathiat> u didnt start it as root or something? 01:02 -!- mpirun [~mpirun@65.64.191.142] has joined #uml 01:02 < idcmp> i started it as root and the problem remained, then i looked and saw every file on my fs is 500.500 01:02 < idcmp> anyone know if debian's dpkg has a --fixperms type thing like rpm does? 01:02 < idcmp> :) 01:03 < mpirun> would i, as a newbie programmer, have any chance of getting x86-64 guest working on x86-64 host if i followed the port-howto? 01:06 < Pahan> The UML is not x86-64. 01:06 < Pahan> It is of ARCH="um" 01:07 < mpirun> so it will never be able to execute anything other than elf32? 01:08 * Pahan cries in pain. 01:08 < Pahan> I think you are in for a wild ride. 01:08 < Dave\\\> wild ride? 01:08 < Dave\\\> woohoo 01:08 * Dave\\\ hops in 01:08 -!- Dave\\\ is now known as Dave\\ 01:08 * Pahan drops Dave\\ off a cliff. 01:09 < Pahan> mpirun: The UML is not SUPPOSED to be arch-dependant, but I guess you found a way in which it is. 01:09 < Pahan> Or maybe it's smarter than that. 01:10 < mpirun> does the ppc port use ppc guest, or i386 guest? 01:12 < Pahan> You'll know better if you just look at the code. 01:20 -!- desaster [~desaster@dsl-jklgw1o22.dial.inet.fi] has quit [Ping timeout: 490 seconds] 01:35 < idcmp> okay.. 01:35 * idcmp almost has that huge mess cleaned up 01:42 < idcmp> phfew.. man what a mess 01:42 < idcmp> caker: thanks for you patience :) 01:42 -!- idcmp [~idcmp@CPE0050dafe87e7-CM013359904800.cpe.net.cable.rogers.com] has quit [Quit: bed time] 01:42 < caker> idcmp: no prob 01:42 -!- Floob [~david@CPE00a0cc23b80f-CM400049226055.cpe.net.cable.rogers.com] has joined #uml 01:45 -!- Floob [~david@CPE00a0cc23b80f-CM400049226055.cpe.net.cable.rogers.com] has quit [] 01:46 -!- ElectricElf [~david@elf.noc.oftc.net] has quit [] 01:48 -!- ElectricElf [~david@elf.noc.oftc.net] has joined #uml 02:15 -!- QSECOFR [~rocky@lab.atipa.com] has quit [Ping timeout: 490 seconds] 02:32 -!- revenger [revenger@p508B252D.dip.t-dialin.net] has quit [Ping timeout: 488 seconds] 03:45 -!- _slh__ [~chatzilla@68.19.239.102] has quit [Ping timeout: 488 seconds] 04:13 < stratsi> anyone succeed to apply the skas patch on 2.4.20-8? 04:50 -!- desaster [~desaster@dsl-jklgw1o22.dial.inet.fi] has joined #uml 07:31 -!- mistral [mistral@jstevenson.plus.com] has quit [Ping timeout: 490 seconds] 07:41 -!- ElectricElf [~david@elf.noc.oftc.net] has quit [] 07:47 -!- Pahan [~pahan@adsl-64-123-27-105.dsl.austtx.swbell.net] has quit [Ping timeout: 490 seconds] 08:22 -!- ElectricElf [~david@elf.noc.oftc.net] has joined #uml 08:34 -!- ElectricElf [~david@elf.noc.oftc.net] has quit [Remote host closed the connection] 08:58 -!- ido [~Ido@212.150.75.108] has quit [] 09:18 -!- david [~david@207.166.203.138] has left #uml [Client exiting] 09:18 -!- david [~david@207.166.203.138] has joined #uml 09:18 -!- QSECOFR [~rocky@lab.atipa.com] has joined #uml 09:18 < david> hello 09:18 < david> caker: around? 09:28 -!- green [green@linuxhacker.ru] has quit [Read error: Connection reset by peer] 09:34 -!- kara [~kara@varstation.lanscape.net] has joined #uml 10:12 < stratsi> i can't resolve hosts in my uml, any suggestions? ping's working, tcp connections too 10:13 < stratsi> bash-2.05# host www.google.com 192.168.2.15 10:13 < stratsi> ;; connection timed out; no servers could be reached 10:23 < david> /etc/resolv.conf? 10:24 < david> do udp connections work, since that's what DNS uses 10:28 -!- green [green@217.76.32.60] has joined #uml 10:45 < stratsi> /etc/resolv.conf was ok...damn redhat, the default installation has firewall for UDP :) 10:45 < stratsi> it's ok now 10:52 < david> :-) 10:55 < green> hehe. That was you who agreed to "default" firewall thing during install ;) 11:23 -!- smcavoy [~smcavoy@grimlock.drive-megawheels.net] has joined #uml 11:38 < stratsi> my boss installed it...:) 11:38 < green> hehe 11:38 < caker> morning 11:41 < stratsi> evening! 11:41 < stratsi> even night 11:52 -!- mistral [mistral@jstevenson.plus.com] has joined #uml 12:00 -!- silug [~steve@osiris.silug.org] has joined #uml 12:05 -!- ticallion [~ticallion@213.175.160.87] has joined #uml 12:16 -!- smcavoy [~smcavoy@grimlock.drive-megawheels.net] has quit [Read error: Connection reset by peer] 12:59 -!- ticallion [~ticallion@213.175.160.87] has quit [Quit: bb in an hour or two] 13:40 -!- mistral [mistral@jstevenson.plus.com] has quit [Read error: Connection reset by peer] 13:40 -!- mistral [mistral@212.159.71.212] has joined #uml 14:00 -!- Ryback_ [~ra993482@malazarte.lsd.ic.unicamp.br] has joined #uml 14:59 -!- BB1 [~chris@62.3.208.26] has quit [Ping timeout: 490 seconds] 15:07 -!- mistral [mistral@212.159.71.212] has quit [Read error: No route to host] 15:07 -!- mistral [mistral@jstevenson.plus.com] has joined #uml 15:23 -!- _slh__ [~chatzilla@adsl-19-226-244.jax.bellsouth.net] has joined #uml 15:31 -!- stratsi [~stratsi@80.72.68.187] has quit [Ping timeout: 488 seconds] 16:59 < caker> I know this has been asked 100 times, but does skas apply to 2.5? 17:07 < david> caker: no 17:07 < caker> david: thanks 17:08 -!- Pahan [~pahan@64.123.27.105] has joined #uml 17:28 -!- Dave\\ [dave@ool-4352e135.dyn.optonline.net] has quit [] 17:28 -!- Dave\\ [dave@ool-4352e135.dyn.optonline.net] has joined #uml 17:31 -!- Ryback_ [~ra993482@malazarte.lsd.ic.unicamp.br] has quit [Quit: Client Exiting] 17:31 -!- stratsi [~stratsi@80.72.68.187] has joined #uml 18:00 -!- rob [rob@213.230.203.221] has joined #uml 18:00 < rob> hrm 18:00 < rob> has anyone used cbq before for bandwidth policing? 18:13 < desaster> yes, well, now i use htb 18:13 < desaster> it's tricky, don't expect me to be able to answer your possible question ;) 18:13 * BB is lazy and uses shaperd ;) 18:14 * rob wants real bandwith policing such as the rate-limit on Cisco switches and can't get it to work 18:16 < desaster> iirc www.lartc.org had some info about it 18:16 < desaster> keep in mind that you can only shape the traffic that you send! 18:17 < desaster> (which isn't a problem if your box is a gateway for example) 18:17 < desaster> here's a page about htb (which is preferred over cbq) 18:17 < desaster> [01:14] (which isn't a problem if your box is a gateway for 18:17 < desaster> example) 18:17 < desaster> oops 18:17 < desaster> let me try again: here's a page about htb (which is preferred over cbq) http://luxik.cdi.cz/~devik/qos/htb/ 18:19 < rob> :) 18:19 < rob> I can only shape the traffic that I send? I can't say limit an interface to a maximum download speed of 0.5mbit? 18:21 < desaster> "How to set single HTB up for more interfaces or for incoming packets 18:21 < desaster> You need IMQ for this because all qdisc can handle only outgoing traffic on single interface. See http://trash.net/~kaber/imq/. " 18:21 < desaster> (imq i have no idea about) 18:22 < rob> ok, guess its time for me to start reading 18:27 < desaster> perhaps start by limiting outgoing traffic as an exercise?-) 18:27 < desaster> before going into wacky imq stuff 18:27 < rob> I've already done that :), I'm looking into sorting it once and for all now 18:28 < rob> I can either look at "wacky imq" stuff, or try and figure out Zebra with OSPF 18:29 < BB> i tried to figure out zebra with bgp4 and gave in 18:29 < rob> I looked at Zebra once, then swore to avoid it 18:30 < BB> heh, i wanted to use two vpn links to send traffic down but the kernel is limited to one connection going one way it seems 18:30 < rob> :( 20:16 -!- ticallion [~ticallion@213.175.160.106] has joined #uml 20:34 -!- desaster [~desaster@dsl-jklgw1o22.dial.inet.fi] has quit [Quit: ircII EPIC4-1.1.12 -- Are we there yet?] 20:49 -!- jdike [~jdike@h69-21-248-234.69-21.unk.tds.net] has joined #uml 20:51 < jdike> hi guys 20:56 < ticallion> hey dude 21:02 < david> jdike: heya 21:03 < ticallion> sup david?! 21:03 < rob> hrm 21:03 * rob stabs the serial link between his two machines 21:03 < ticallion> heh 21:03 < rob> right, it works 21:03 < rob> woohoo! 21:03 < rob> but.. 21:04 < rob> I want to see the kernel boot messages 21:04 < rob> and the init messages 21:04 < rob> so I read around 21:04 < rob> and it seemed to me that I needed to do append="console=ttyS0,9600n8" 21:04 < rob> so I do that 21:04 < rob> and they dissappear 21:05 < david> ticallion: not much - dealing with results of bad weather 21:05 < ticallion> david: and those would be?! 21:06 < david> ticallion: trees falling down, power crapping out, cable crapping out, phone crapping out 21:06 < david> power has been off ~4 times today so far 21:06 < ticallion> lol 21:07 < david> sometimes for a couple of seconds, sometime for longer 21:07 < ticallion> i mean 21:07 * ticallion comforts david 21:08 < david> ticallion: it was actually funny 21:08 < david> there was a severe weather warning on our NBC affiliate eariler 21:08 < david> then the cable went out halfway through 21:08 < ticallion> david: dude, that's a daily occurence in this here utopia of a country, power outages 21:08 < ticallion> ugh 21:09 < david> ticallion: I used to deal with them back in England when I lived in the country - Not really a major problem when I have UPSes 21:09 < david> of course, my UPS here is mainly for protection rather than backup, so it lasted about 15minutes 21:11 < ticallion> same here... I've had my share of battery fumes from those truck batteries on larger upses, I think I damamged my vision because of those 21:11 < ticallion> david: where do u live now?! 21:12 < david> ticallion: near Cleveland, OH 21:12 < david> bbiab 21:13 < ticallion> sit true what they say bout ohio being hick country?! 21:14 -!- Dave\\ [dave@ool-4352e135.dyn.optonline.net] has quit [] 21:19 < david> ticallion: not north ohio 21:19 < david> ticallion: the middle is 21:19 < ticallion> david: how's the snowboarding season over there?! 21:19 < ticallion> I spotted some job offers from ohio but chose not to apply 21:25 -!- iDish [id@cuodan.net] has joined #uml 21:25 -!- iDish [id@cuodan.net] has quit [Client Quit] 21:26 < david> ticallion: no idea - No hills around here :-) 21:27 < ticallion> :( ohio still gon be in my deny list 21:29 -!- stratsi [~stratsi@80.72.68.187] has quit [Ping timeout: 490 seconds] 21:42 < caker> hello 21:44 < ticallion> hey dude 22:17 -!- jdike [~jdike@h69-21-248-234.69-21.unk.tds.net] has quit [Quit: Leaving] 22:19 -!- ticallion [~ticallion@213.175.160.106] has quit [Quit: food time] 22:43 -!- mistral [mistral@jstevenson.plus.com] has quit [Ping timeout: 490 seconds] 22:43 < david> caker: comparing vmstat in 2.4.21-rc2 against 2.4.22-pre3 is pretty interesting 22:44 < david> -rc5 looks good too 22:44 < david> -rc1 looks broke 22:45 < david> strange 22:46 < caker> It was pre-6 i think that introduced the major changes to rw_block (or whatever it was called. .) 22:48 < david> ll_rw_blk.c? 22:49 < caker> ya 22:50 -!- mistral [mistral@jstevenson.plus.com] has joined #uml 22:50 < david> caker: hrm,ok 22:50 < david> caker: I'm going to stick that q_full.diff on my UML kernel and see what it does 22:50 < caker> and rc6 where they thought it was fixed 22:50 < caker> david: sounds fun 23:23 -!- _slh__ [~chatzilla@adsl-19-226-244.jax.bellsouth.net] has quit [Read error: Connection reset by peer] 23:30 -!- stratsi [~stratsi@80.72.68.187] has joined #uml --- Log closed Wed Jul 09 00:00:00 2003