Back to Home / #xen / 2005 / 04 / Prev Day | Next Day
#xen IRC Logs for 2005-04-03

---Logopened Sun Apr 03 00:00:42 2005
00:53--- <<-- Sir_Ahzz [] has quit (Ping timeout: 480 seconds)
00:54--- ---> Sir_Ahzz [] has joined #xen
00:54--- Channel: mode/#xen [+o Sir_Ahzz] by ChanServ
01:28--- <<-- matta-lt [~matta@] has quit (Quit: Hey! Where'd my controlling terminal go?)
03:26--- <<-- Sir_Ahzz [] has quit (Ping timeout: 480 seconds)
03:26--- ---> Sir_Ahzz [] has joined #xen
03:26--- Channel: mode/#xen [+o Sir_Ahzz] by ChanServ
04:44--- <<-- akw [] has quit (Quit: Leaving)
04:58--- ---> CryptWizard [] has joined #xen
05:50--- <<-- muli [] has quit (Ping timeout: 480 seconds)
07:24--- User: *** CryptWizard is now known as CryptWizard-Sleep-
07:39--- ---> hebutterworth [] has joined #xen
08:00--- <<-- DEac- [] has quit (Ping timeout: 480 seconds)
08:12--- ---> DEac- [] has joined #xen
08:55--- <<-- DEac- [] has quit (Quit: Leaving)
09:02--- ---> MarkW [] has joined #xen
09:07--- ---> DEac- [] has joined #xen
09:26--- <--- MarkW [] has left #xen (Kopete 0.10 :
09:48--- ---> andyw [] has joined #xen
10:04--- <<-- andyw [] has quit (Quit: Leaving)
10:36--- <<-- JViz [] has quit (Ping timeout: 480 seconds)
10:39--- ---> monrad [] has joined #xen
12:20@riel ahhh, gotta love the subtleties of context switching code
12:20* riel tries Keir's smp bugfix in the Fedora kernel
12:22@Sir_Ahzz mmm smp
12:23@riel looks like I might be able to get xen in a very nice shape for fedora core 4, test 2
12:59--- ---> MarkW [] has joined #xen
13:02Arnald is there any way to control the vif numbers?
13:03Arnald my gkrellm net window is about six feet wide ;)
13:04Arnald if I could just stop them incrementing each time I do an xm create . . .
13:04@Sir_Ahzz heh
13:04mikegrb EEE
13:05@Sir_Ahzz use the agregate gkrelm net monitor then.
13:05Arnald so far up to vif76.0 ;)
13:06Arnald Sir_Ahzz: I have a little tidying up to do first!
13:07Arnald Sir_Ahzz: BTW, never heard of it, is it part of gkrellm, or an add-on?
13:07* Arnald wonders if the counter eventually wraps around, or just blows up! ;)
13:07@riel only one way to find out
13:08Arnald riel: or I could ask you ;)
13:09@riel I might test it after the Xen summit
13:09--- ---> tompuppy [] has joined #xen
13:10Arnald riel: I will keep going as long as the machine stays up - at least Xen appears to be fairly stable ATM
13:10@riel yes, it would appear so
13:10--- <<-- hebutterworth [] has quit (Remote host closed the connection)
13:10@riel the only thing that keeps driving me nuts is the incompatible changes in the kernel/hypervisor interface ;)
13:12Arnald er, right - good luck with that ;)
13:13tompuppy !weather 60007
13:14--- <<-- tompuppy [] has quit (Quit: )
13:14* riel wonders when the RHN people will notice that 3 CPU RHEL4 system in their statistics
13:26MarkW Arnald: I think the theory is that Xen domain IDs don't wrap
13:27MarkW so it's only going to get worse for your gkrellm :-)
13:27--- ---> MadMethod [] has joined #xen
13:27MarkW I don't actually know how large an integer Linux can cope with as an ethernet device number. presumably quite large...
13:33--- <<-- Method [] has quit (Ping timeout: 480 seconds)
13:34cw it's really a string, you can even change it on the fly
13:35cw i think the limit is 16 chars but i would have to check
13:38Arnald MarkW: OK thanks, I'll bear that in mind ;)
13:39MarkW i think domids are 16 bits these days
13:40MarkW so i guess it might fit in there
13:40MarkW for your own sanity with gkrellm, you could maybe hack the network scripts to rename the devices (i think you can do that...)
13:43Arnald MarkW: yes, just checked, gkrellm seems to create virtual interfaces on-the-fly ie. no config files I can edit
13:44MarkW i meant the Xen network scripts
13:44MarkW there's a script that's called every time a vif is created, maybe there's something you can do with that
13:44Arnald MarkW: yes, I know - just pointing out that the alternative is not an option
13:45Arnald thanks
13:45MarkW ah ok
13:45MarkW cheers
13:56knewt Arnald: gkrellm doesn't get rid of the old vifs when the old domains are shut down?
15:17MarkW knewt: i think his problem is the length of the Xen domain IDs over time
15:17MarkW when the domain IDs get big, so do the vif numbers
15:20--- ---> muli [] has joined #xen
15:59--- <<-- Sir_Ahzz [] has quit (Ping timeout: 480 seconds)
16:00--- ---> Sir_Ahzz [] has joined #xen
16:00--- Channel: mode/#xen [+o Sir_Ahzz] by ChanServ
16:49--- User: *** CryptWizard-Sleep- is now known as CryptWizard
17:02--- ---> matta [~matta@] has joined #xen
17:15--- User: *** CryptWizard is now known as CryptWizard{School
17:17--- <<-- Icy [] has quit (Ping timeout: 480 seconds)
17:34--- ---> Bluefox [] has joined #xen
18:00--- ---> Robot101 [] has joined #xen
18:01* mikegrb cowers in fear
18:02mikegrb please don't eat me mr robot!
18:02* Robot101 bleeps ominously
18:02MarkW mikegrb: it's ok, he's with me - i think i'm near the kill switch
18:03mikegrb thanks mark
18:04--- ---> AJ_ [] has joined #xen
18:04* AJ_ waves at Robot101 and MarkW
18:30--- <<-- monrad [] has quit (Ping timeout: 480 seconds)
18:52--- ---> Lampy [] has joined #xen
19:25Arnald knewt: sorry for the delay, but no ;)
19:25Arnald anyway, bedtime here - g'night all
19:34MarkW Arnald: good night
19:36* riel is looking forward to the xen summit
19:39knewt gah. have a good time.
19:41@riel doubtful ;)
19:41@riel we're flying out overnight from tuesday to wednesday
19:42@riel I'll probably feel pretty bad most of the time
19:43knewt hmm. thinking about it, yeah. i always feel worse when going east compared to going west. well, on long haul anyway
19:44* riel is counting on 2-3 hours resting time between arrival in cambridge and the start of the xen summit
19:44@Sir_Ahzz owww.
19:44@riel after skipping a night
19:44@Sir_Ahzz finally got the stump chopped out, but my back hurts now. 8-P
19:46--- <--- Lampy [] has left #xen ()
19:47knewt well, if not a good time then have a productive time. wish i was going to be there, but can't be helped.
19:47MarkW Sir_Ahzz: well done, chainsaw fixed?
19:47@Sir_Ahzz yeah, too much sandy dirt to use it on the lower roots though. 8-P
19:47@Sir_Ahzz it'll be usefull to cut up the stump though. :)
19:47@Sir_Ahzz 2.5 ft diameter.
19:48@Sir_Ahzz the chansaw has an 18" blade. barely long enough to get 1/2 way through it.
19:49MarkW problematic!
19:50@Sir_Ahzz nah. slice down one side, turn it 90 deg, slice, repeat till cut.
19:52@Sir_Ahzz the sucky part is the sunburn I got yesterday.
19:52@Sir_Ahzz but, no more rotten smelly stump right next to the front door. :)
19:52@Sir_Ahzz it was starting to really smell.
19:52MarkW ewww!
19:53@Sir_Ahzz gonna play some lbreakout2 before I have to go fix dinner.
19:53* knewt ponders going to sleep, it being almost 2am
19:59MarkW knewt: the night is still young!
20:20--- <<-- DEac- [] has quit (Ping timeout: 480 seconds)
20:23--- <--- AJ_ [] has left #xen ()
20:34--- ---> DEac- [] has joined #xen
20:55--- ---> JViz [] has joined #xen
21:04--- ---> rusty [] has joined #xen
---Logclosed Sun Apr 03 21:08:44 2005
---Logopened Sun Apr 03 20:08:47 2005
20:08--- ---> mikegrb [] has joined #xen
20:09--- ---> monrad [] has joined #xen
20:10-!- Irssi: Channel: Join to #xen was synced in 99 secs
---Logclosed Sun Apr 03 20:13:17 2005
---Logopened Sun Apr 03 20:13:18 2005
20:13--- ---> mikegrb [] has joined #xen
20:14-!- Irssi: Channel: Join to #xen was synced in 94 secs
20:27--- ---> Andrew_ [] has joined #xen
20:34Andrew_ hi, has anyone here had problems enabling dma on an ide drive after booting with the xen kernel? after I use xen I get an operation not permitted when I try to enable dma
20:39rusty Andrew_: CONFIG_PCI_IDE=y?
20:54--- <<-- mael [] has quit (Ping timeout: 480 seconds)
21:57Andrew_ rusty: sure, does that mean I have to recompile xen? or is that an option somewhere else?
21:59@riel in the dom0 kernel
22:01--- <<-- MarkW [] has quit (Quit: Kopete 0.9.2 :
22:01Andrew_ ah, strange... thats not in the one thats in the binary dist?
22:02@riel dunno, I don't use the binary dist from xensource
22:02* riel builds the Fedora RPMs ;)
22:02Andrew_ ah haha
22:03Andrew_ ok, well Ill definitely check that... the only hint I had before was that I might need to specify the exact chipset or something
22:03@riel (and those seem to work)
22:03Andrew_ cool
22:03@riel if you find a problem with the Fedora RPMs, file a bugreport and I'll look into it
22:03Andrew_ I was trying to get it working quick for some tests and used knoppix and installed to HD
22:03rusty Andrew_: I had this same problem w/o Xen.
22:03* riel hopes the xensource people can focus on xen development, and not on packaging issues ;)
22:04Andrew_ I might try that, Ive never been font of redhat/fedora though hehe
22:04Andrew_ rusty: ah ok
22:04@riel Andrew_: you could use Debian's xen 1.2 RPMs ;)
22:04* riel runs
22:04@riel ummm s/RPMs/debs/
22:04Andrew_ wow, I think Ill try the debs
22:04@riel they're only 2 years out of date
22:04Andrew_ if I say gentoo next, are you going to tell me theres an ebuild? :-D
22:05Andrew_ oh haha, right 1.2
22:08Andrew_ I haven't actually played with Xen much... I was working with something called coLinux which works very similar to Xen (with less protection) but couldn't manage to get it to compile with linux as the host
22:08Andrew_ and then I recently found out about Xen and its binary distribution worked out of the box so I figured that would be worth a try
22:10@riel Xen seems to be working pretty well so far
22:10@riel I've put some reasonably heavy stress tests on the RPMs that will go into FC4 test 2
22:10Andrew_ nice
22:11Andrew_ Im amazed that its open source, it seems to be a very elegant solution, which OSS is not usually known for ;-)
22:11* riel starts another aim7 run ;)
22:11Andrew_ aim7?
22:12@riel you've never seen proprietary software from nearby, have you? ;)
22:12@riel OSS is surprisingly clean compared to some other software
22:12@riel you can see that every time some piece of formerly closed source software is released as open source
22:12Andrew_ haha you are correct, I guess the proprietary stuff is only elegant as far as its the standard
22:13@riel just look at the original Mozilla codebase
22:13@riel as one example
22:13@riel or some of the device drivers that have been opened up by vendors
22:14Andrew_ oh I meant elegant only as far as the user experience hehe
22:14Andrew_ but Im also not complaining, everything I use is open source except vmware
22:15Andrew_ and hopefully if I ever release any software it will be OSS
22:15@riel I've seen user experience improve across the board, both open and closed software
22:16Andrew_ agreed, like installing knoppix the other day, that was the easiest operating system install Ive ever worked with :)
22:31--- <<-- rusty [] has quit (Quit: Client exiting)
22:47--- User: *** riel is now known as surriel
22:59--- <--- VS_ChanLog [] has left #xen (Rotating Logs)
22:59--- ---> VS_ChanLog [] has joined #xen
23:00Andrew_ btw, Ive seen a little bit of info, but does anyone know how the freebsd support in xen is coming along?
23:18--- ---> sleon [] has joined #xen
23:18sleon hi
23:18sleon where to get aim7 from?
23:19Robot101 aim as in AOL Instant Messenger?
23:20sleon Robot101, nono it is a unix benchmark suite
23:20sleon it is from sco group
23:20sleon and the reomved it from their website
23:20sleon it was in contrib dir :D
23:21Robot101 hmm you're right
23:23sleon i was told it is very good
23:25Robot101 ?
23:25sleon cool
23:26--- <<-- cw [] has quit (Quit: satan)
23:26--- <<-- MadMethod [] has quit (Ping timeout: 480 seconds)
23:36--- ---> anticw [] has joined #xen
23:58--- ---> Method [] has joined #xen
---Logclosed Mon Apr 04 00:00:23 2005