Back to Home / #uml / 2007 / 03 / Prev Day | Next Day
#uml IRC Logs for 2007-03-01

---Logopened Thu Mar 01 00:00:42 2007
01:06|-|mjf [] has joined #uml
01:21|-|rbarooah_ [~chatzilla@] has joined #uml
01:21|-|rbarooah_ changed nick to rbarooah
01:35|-|motp [~motp@] has joined #uml
01:45|-|da-x [] has joined #uml
02:00|-|da-x [] has quit [Ping timeout: 480 seconds]
02:03|-|N3o [~Neo@] has quit [Ping timeout: 480 seconds]
02:05|-|Ancalagon [] has joined #uml
02:10|-|dlezcano [] has quit [Ping timeout: 480 seconds]
02:11|-|rbarooah_ [~chatzilla@] has joined #uml
02:13|-|rasix [~jeruk@] has quit [Remote host closed the connection]
02:14|-|rbarooah [~chatzilla@] has quit [Ping timeout: 480 seconds]
02:14|-|rbarooah_ changed nick to rbarooah
02:25|-|rbarooah [~chatzilla@] has quit [Ping timeout: 480 seconds]
02:33|-|delsid [] has joined #uml
02:35<delsid>runnning under sidux
02:35<delsid>i got about 10 kernel panics in about 2 days
02:36<delsid>in the UML. is SKAS0 known to be unstable?
03:05|-|Serg [] has joined #uml
03:28|-|da-x [] has joined #uml
03:32|-|karrde_ [] has joined #uml
03:38|-|dlezcano [] has joined #uml
03:46|-|delsid [] has quit [Ping timeout: 480 seconds]
04:11|-|motp [~motp@] has quit [Quit: Leaving]
04:18|-|delsid [] has joined #uml
04:31|-|tyler [~tyler@] has joined #uml
05:00|-|richardw [] has joined #uml
05:38[~]delsid wonders why ppl hang out here if they never talk
05:55<peterz>wait till the east coast wakes up
05:58<delsid>hi peterz
05:59<delsid>i mainly want uml for security and good speed for running a browsing system on this p3 box
05:59<delsid>so if there's another libxine exploit or something i don't end up reinstalling again
06:01<delsid>anyway i've seen this uml instance crash about 10 times now
06:01<delsid>and i'm wondering if that should be concerning me
06:02<peterz>yeah, they generaly don't
06:03<peterz>but then again, I usually don't keep em around that long
06:05<delsid>you use them for testing/debugging then?
06:07<peterz>but there are others here who do use it like you do
06:07<peterz>I'm sure Jeff will want details on how it comes down after two days
06:07<delsid>do you run X on uml?
06:08<delsid>i chose freenx and am pretty happy with the performance
06:10<peterz>no, I'm happy it I reach userspace and can run some testcase
06:15<delsid>what kind of stuff do you write?
06:17<peterz>I mostly hack on the vm
06:26|-|delsid [] has quit [Ping timeout: 480 seconds]
06:29|-|delsid [] has joined #uml
06:30<delsid>Kernel panic - not syncing: Kernel mode signal 7
06:30<delsid>happened when writing to disk onto ext3
06:36<delsid>The filesystem where UML is keeping its physical memory files is full. This is normally a tmpfs mount, either /dev/shm or /tmp.
06:37<delsid>tmpfs 192952 61252 131700 32% /dev/shm
06:42<delsid>vi /etc/default/tmpfs
06:43<delsid>mount -t tmpfs tmpfs /dev/shm -o remount,size=536870912
06:47<delsid>tmpfs 524288 194012 330276 38% /dev/shm
06:47<delsid>disk writes chew up shm
07:16|-|mode2_ [] has joined #uml
07:21|-|mode2 [] has quit [Ping timeout: 480 seconds]
07:25|-|richardw_ [] has joined #uml
07:32|-|richardw [] has quit [Ping timeout: 480 seconds]
07:47|-|writerz_ [] has joined #uml
07:48|-|tyler [~tyler@] has quit [Ping timeout: 480 seconds]
07:52|-|mode2_ [] has quit [Ping timeout: 480 seconds]
07:55|-|tyler [~tyler@] has joined #uml
07:58|-|Spanska [~bleh@] has joined #uml
07:58<Spanska>Can I bridge a virtual ethernet interface with my host's one?
08:17<dlezcano>Spanska: yes
08:18<dlezcano>Spanska: you should have a console or a direct access to your host
08:21<Spanska>it's all written how to do that on the wiki?
08:22<dlezcano>Spanska: I don't know :)
08:22<Spanska>okay, well, I'll come back later, I have a course :\
08:22<dlezcano>Spanska: Do you want more informations about bridging the guest ?
08:22<dlezcano>Spanska: ok
08:23|-|Spanska [~bleh@] has quit [Quit: Leaving]
08:34|-|tyler [~tyler@] has quit [Remote host closed the connection]
08:40|-|baroni [] has joined #uml
08:44|-|tyler [~tyler@] has joined #uml
08:58|-|Ancalagon [] has quit [Quit: Chatzilla 0.9.75 [Firefox 1.0.4/20070116]]
09:53|-|baroni [] has quit [Remote host closed the connection]
10:22|-|hfb [] has joined #uml
10:45<delsid>it'd be nice to be able to mount hostfs directories in the uml invocation, without allowing the mount within the uml instance itself
10:45<delsid>like obdc=/mnt/hda5/shared
10:52|-|rbarooah_ [~chatzilla@] has joined #uml
10:52|-|rbarooah_ changed nick to rbarooah
11:11|-|tchan [] has quit [Quit: WeeChat 0.2.4-cvs]
11:11|-|tchan [] has joined #uml
11:21|-|the_hydra [~mulyadi@] has joined #uml
11:27|-|baroni [] has joined #uml
11:48<kokoko1>hi the_hydra
12:10|-|dlezcano [] has quit [Read error: Connection reset by peer]
12:14|-|baroni [] has quit [Remote host closed the connection]
12:15|-|hfb [] has left #uml [Leaving]
12:15|-|baroni [] has joined #uml
12:25|-|mode2_ [] has joined #uml
12:30|-|Spanska [] has joined #uml
12:31<Spanska>Is there a way to bridge an Ethernet interface of the guest with the host?
12:32|-|writerz_ [] has quit [Ping timeout: 480 seconds]
12:33<the_hydra>IIRC yes, use bridging
12:34|-|flatronf700B [~flatronf7@] has quit [Ping timeout: 480 seconds]
12:36<Spanska>How can I do so?
12:37|-|kos_tom [] has joined #uml
12:37<Spanska>After I boot the system I downloaded off the web site, I have no NIC at all. The only interface I have is "lo"
12:38<the_hydra>hm no idea, IMHO you need to compile a UML kernel by your own
12:42<the_hydra>have you really checked the Wiki?
12:43<Spanska>nope, I will right after I'm done with making available my files in the guest
12:44<the_hydra>i remember UML kernel has a parameter to show the configuration used when it was compiled
12:45<the_hydra>I forgot the name, so if you find that out quickly and paste the result somewhere e.g pastebin maybe I can give a hint
12:46<Spanska>Okay, I'll brb and check that right after I'm back.
12:46<the_hydra>i am going to offline soon, so I can't wait too long
12:48<Spanska>Ok, checking right now sorry.. had to go to bathroom
12:51<Spanska>I found URLs for pre-compiled kernels so far. What am I looking for exactly?
12:52<the_hydra>run your UML kernel with "--help" parameter, and see which parameter will yield the complete text of kernel configuration?
12:52<Spanska>I found stuff about tun as wel.
12:53<the_hydra>ok maybe that's it
12:53<the_hydra>now run your UML kernel with that param only
12:53<the_hydra>and paste the result on pastebin or somewhere else
12:54<the_hydra>ok wait
12:57<the_hydra>sems like everything is OK
12:59<the_hydra>is that you who asked to this channel yesterday about the same problem?
13:03<the_hydra>please follow
13:04<the_hydra>sorry gtg
13:04<the_hydra>maybe other peoplecan help you better
13:04|-|the_hydra [~mulyadi@] has quit [Quit: using sirc version 2.211+KSIRC/1.2.4]
13:37|-|writerz_ [] has joined #uml
13:42|-|mode2_ [] has quit [Ping timeout: 480 seconds]
13:47|-|mode2_ [] has joined #uml
13:52|-|writerz_ [] has quit [Ping timeout: 480 seconds]
13:55|-|rbarooah [~chatzilla@] has left #uml []
14:07|-|Serg [] has quit [Ping timeout: 480 seconds]
14:15|-|Spanska [] has quit [Quit: Leaving]
14:53|-|dlezcano [] has joined #uml
15:26|-|rw__ [] has joined #uml
15:33|-|richardw_ [] has quit [Ping timeout: 480 seconds]
15:42|-|writerz_ [] has joined #uml
15:45|-|mode2_ [] has quit [Ping timeout: 480 seconds]
16:35|-|tyler [~tyler@] has quit [Remote host closed the connection]
17:42|-|ousado__ [] has joined #uml
17:47|-|mjf [] has quit [Quit: leaving]
17:49|-|ousado_ [] has quit [Ping timeout: 480 seconds]
18:05|-|rw__ [] has quit [Remote host closed the connection]
18:14|-|flatronf700B [~flatronf7@] has joined #uml
19:49|-|Aaron_Mason [~chatzilla@] has joined #uml
19:50<Aaron_Mason>hey, i want some opinions here... is it a better idea to run all UMLs under individual users, or will running them all under the one user be fine?
20:05<Aaron_Mason>whoa, one at a time guys :P
20:37<delsid>hi Aaron_Mason
20:38<delsid>in respect to what, security?
20:38<Aaron_Mason>security, resource use, etc
20:39<delsid>don't know. i am getting started with UML
20:39<delsid>are you using a precompiled UML or rolling your own?
20:40<caker>Aaron_Mason: I run all mine under different users
20:41<Aaron_Mason>delsid: i roll my own
20:41<Aaron_Mason>caker: does that have any real benefits against running them under the one user?
20:42<caker>Aaron_Mason: none that I can think of
20:45<Aaron_Mason>caker: I suppose you could enforce process limits if you ran each under individual users if you wished to
20:45<Aaron_Mason>i mean, like if you were running a vds host and had plans with differing limits
20:45<caker>I run under a SKAS3 patched host, in which case that wouldn't have any effect
20:45<caker><-- that's me
20:45<Aaron_Mason>good point
20:46<delsid>can the skas3 patch be applied to 2.6.20 kernel?
20:46<Aaron_Mason>delsid: theres a skas patch in the blaisorblade repository
20:47<Aaron_Mason>for 2.6.20
20:47<delsid>oh last i saw was 2.6.18 thanks. is skas3 faster than skas0
20:47<caker>it is
20:50<delsid>do you use the v8.2 or pre9
20:50<caker>I've been using v8.2
20:52<delsid>thanks.. what do you think of my idea for a command line invocation for a non-modifyable hostfs mount? something like ubdc=/mnt/hda2/shared ?
20:53<Aaron_Mason>hey, if i want to use just the daemon network support, i dont need any others to be enabled, right?
20:53<delsid>i'd want to disable global hostfs mounts
20:53<delsid>you mean /usr/bin/uml_switch Aaron_Mason ?
20:54<Aaron_Mason>delsid: yeah
20:59[~]delsid discovers that caker is a UML guru
20:59<delsid>i bet it's tiring to deal with n00bs :)
21:00<caker>man, that's ancient and probably wrong :)
21:01<delsid>also cool
21:04<caker>that's me :)
21:05<delsid>hope your business is doing well
21:05<delsid>not many hosting companies seem to contribute to OSS
21:05<delsid>thoguh i know a couple of examples
21:07<caker>delsid: thank you
21:10<Aaron_Mason>hey, should I be worried about this: WARNING: vmlinux - Section mismatch: reference to .init.text:malloc from .got after '_GLOBAL_OFFSET_TABLE_' (at offset 0x8205260)
21:27<Aaron_Mason>heh... if i do a tracert from another computer on the network to a UML machine.... it shows it going through the linux box to the uml machine... I'm such a geek
21:27<Aaron_Mason>(or traceroute, as the case may be on another linux box)
21:33[~]delsid is using tap0
21:34[~]delsid listens to the caker interview
21:35<Aaron_Mason>yeah, my uml_switch implementation uses tap0 to connect to the network... saves opening up a tap for each vm
21:39|-|mode2_ [] has joined #uml
21:40<delsid>i almost convinced my employer to let me set up our 'datacenter' using UML guests for each server, with redundant hardware back in 2003
21:41<Aaron_Mason>what happened? funding dry up?
21:41<Aaron_Mason>hey mode2_
21:41<delsid>boss liked having one physical box per service, and i couldn't make a solid business case for it
21:42<delsid>i'd have done it on my own time too :)
21:42<Aaron_Mason>i can see the guys reason for it... put everything on the one box, you have a single point of failure
21:42<delsid>two boxes with drdb/heartbeat failover replacing a whole wall of PCs
21:43|-|ousado_ [] has joined #uml
21:44<delsid>i guess so. i couldn't figure out how to prevent the host OS from seeing the network at all, while allowing the guest UMLs access to the net
21:45<delsid>wanted to do the heartbeat/switchover using seperate NICs
21:46|-|writerz_ [] has quit [Ping timeout: 480 seconds]
21:51|-|ousado__ [] has quit [Ping timeout: 480 seconds]
22:03|-|delsid [] has quit [Quit: leaving]
22:18|-|delsid [] has joined #uml
22:21|-|mistik1 [] has quit [Read error: Connection reset by peer]
22:35|-|Aaron_Mason [~chatzilla@] has quit [Quit: Chatzilla 0.9.77 [Firefox]]
23:48|-|apic [] has quit [Ping timeout: 480 seconds]
23:52|-|apic [] has joined #uml
---Logclosed Fri Mar 02 00:00:29 2007