--- Day changed

--- Log opened Thu Jun 16 23:59:02 2005
00:49 -!- Hunger [Hunger.hu@LevNor.Hu] has quit [uranium.oftc.net quasar.oftc.net]
00:52 -!- Hunger [Hunger.hu@LevNor.Hu] has joined #xen
01:19 < enum> yo.. anyone awake?
01:19 < chrish01> yea
01:19 < enum> yo chrish01
01:20 < chrish01> howdy
01:20 < enum> I was wondering if anyone is activly working on the web interface for xen
01:20 < enum> I saw it once.. was kinda not helpful ( no offense to anyone who did it )
01:21 < enum> but, I am workong on something similiar for my systems, and was wondering if anyone would be interested in something like that when it is complete
01:21 < chrish01> it got taken out of the 3.0 branch because it was stale and no longer worked
01:21 < chrish01> i would love to help someone make a new one though
01:22 < chrish01> no clue who did the original one (xensv) though
01:22 < enum> brb.. gonna go grab my lapptop
01:22  * chrish01 is really diggin' RubyOnRails
01:26 -!- [e]num [~Administr@ip-207-145-127-226.lax.megapath.net] has joined #xen
01:26 < [e]num> yo
01:26 < chrish01> sup
01:28 < [e]num> so anyhow.. so the source should be available in the current source right?
01:28 < chrish01> the xensv source is in the 2.0 release
01:28 < chrish01> but, i dont think it is compatable at all with 3.0
01:29 < [e]num> is 3.0 current, or is it cvs
01:29 < [e]num> err beta
01:29 < chrish01> 3.0 is "unstable"
01:29 < chrish01> 2.0 is both stable and testing
01:30 < [e]num> hmm.. I will need to install a copy of 3.0 thn
02:32 -!- chrish01 [~chrish01@72-254-24-162.client.stsn.net] has quit [Quit: Leaving]
02:48 -!- bunoc [~bunoc@lez.mag.keio.ac.jp] has quit [Remote host closed the connection]
02:49 -!- Tv [~Tv@GMMDCXLVIII.dsl.saunalahti.fi] has joined #xen
02:50 -!- chrish01 [~chrish01@72-254-24-162.client.stsn.net] has joined #xen
02:51 -!- thelsdj [~thelsdj@24-117-42-27.cpe.cableone.net] has left #xen []
03:49 -!- giuppi [~ciao@host191-163.pool8254.interbusiness.it] has quit [Remote host closed the connection]
04:00 -!- rusty [~rusty@bh02i525f01.au.ibm.com] has quit [Quit: Client exiting]
04:01 < mael> hi chrish01 
04:02 < mael> how is your work going?
04:02 < mael> I read that you were about to reverse engineer the protocol
04:02 < chrish01> alright.
04:02 < chrish01> it sucks that 3.0 is so freakin different than 2.0
04:02 < mael> shouldn't that be easier to ask on the ML?
04:02 < mael> it is not a proprietary software and the dev is known :)
04:03 < chrish01> probably. i havent been productive at all this last week. did a considerable amount of refactoring on local working copy
04:03 < chrish01> async work and such
04:06 < mael> async work?
04:07 < chrish01> the repo does syncronous connections right now.
04:07 < chrish01> im making it async as to not freeze the gui
04:07 < mael> oh ok
04:08 < mael> yeah it seems a nice design
04:08 < mael> you store the configuration information in memory then?
04:08 < chrish01> yea, im pretty satisfied with where the gui is going (except the preferences mock up)
04:08 < chrish01> yea
04:08 < chrish01> the new design will poll once rather than multiple times also
04:09 < chrish01> and then read the PyObject from the servers TreeStore 
04:09 < mael> the "preferences" seems mozilla inspired to me, so far
04:09 < chrish01> its crap
04:09 < mael> well if you say so :)
04:09 < chrish01> heh
04:10 < chrish01> i think a new web app for 3.0 would be better spent time. but no matter what, we need a system that has some *real* authentication
04:10 < mael> yep
04:10 < mael> and I hate latency you have with webapp
04:11 < chrish01> rails is pretty damn fast. im considering it.
04:11 < chrish01> but again, it makes sine to use python :(
04:11 < mael> also, webapp are server memory and cpu waster
04:12 < chrish01> yea. i cant believe the reasoning behind basically dumping a python list to HTTP and calling it an api though
04:12 < chrish01> we need either SOAP or XML-RPC system IMO
04:12 < mael> soap is very expensive also
04:12 < mael> xml-rpc would probably be enough
04:12 < chrish01> soap is. xml rpc is pretty fast
04:13 < chrish01> brb. gotta kill xchat
04:13 -!- chrish01 [~chrish01@72-254-24-162.client.stsn.net] has quit [Quit: Leaving]
04:14 -!- chrish01 [~chrish01@72-254-24-162.client.stsn.net] has joined #xen
04:14 < Tv> I love xen.. 6 domUs running without a hitch.
04:14 < Tv> this is about the point where uml started to choke
04:14 < chrish01> Tv, same
04:14 < chrish01> i switched our routers from UML to xen and was about to get around 20 domains
04:14 < mael> chrish01:  xul could be interesting on the client side
04:14 < Tv> Yeah, my limit is memory more than anything else.
04:14 < chrish01> mael, hahah
04:15 < chrish01> omg, anything but XUL (or python) :D
04:15 < mael> argh
04:15 < mael> actually it make sense (even if it's too young) as it allow reactive and complete client side apps
04:16 < mael> and the load to produce a rdf content is not so heavy for the server side
04:17 < mael> did I say something wrong?
04:17  * mael *blush*
04:17 < chrish01> eheh
04:18 < chrish01> i think an rss feed from servers would be an interesting idea :)
04:21 < chrish01> brb
04:21 -!- chrish01 [~chrish01@72-254-24-162.client.stsn.net] has quit [Quit: Leaving]
04:21 -!- chrish01 [~chrish01@72-254-24-162.client.stsn.net] has joined #xen
04:26 -!- frealek [~nobo@4va54-1-81-56-99-20.fbx.proxad.net] has joined #xen
04:32 -!- joink [~joink@186.80-202-132.nextgentel.com] has joined #xen
04:34 -!- joink_ [~joink@186.80-202-132.nextgentel.com] has quit [Ping timeout: 480 seconds]
04:36 -!- Administrator_ [~Administr@ip-207-145-127-226.lax.megapath.net] has joined #xen
04:36 -!- [e]num [~Administr@ip-207-145-127-226.lax.megapath.net] has quit [Read error: Connection reset by peer]
04:39 < mael> chrish01:  what's the problem with your xchat man?
04:39 < chrish01> nothin now. i didnt have the python gnome extras package intalled
04:39 < chrish01> using Ross Burton's xchat plugin now
04:40 < mael> aaah you see, you love python :)
04:40 < chrish01> hahah
04:42 < mael> chrish01:  btw, do you have an url for medsphere?
04:42 < mael> the wiki is not complete
04:43 < mael> I'd like to add more info but I don't get them :)
04:43 < chrish01> www.medsphere.com
04:44 -!- masm [~masm@c-217-70-66-127.bragatel.pt] has joined #xen
04:45 < chrish01> man, i have too many projects
04:45  * chrish01 is hacking on a personal finance app for gnome right now too
04:45 < mael> are the info correct on the wiki?
04:45 < mael> chrish01:  what app?
04:46 < chrish01> from what i read yes mael, you did a great job with that
04:46 < mael> I know a dev from the grisbi project
04:46 < chrish01> mael, it will be called gnome-money. its a tiny personal finance app like a super lightweight MS Money
04:46 < chrish01> the other guis are just so damn bad
04:46 < mael> you know grisbi?
04:47 < mael> it's not that bad I think
04:47 < chrish01> sec
04:47 < chrish01> ive seen it before, but not lately
04:47 < chrish01> ewwww. gtk+ 1 :)
04:48 < chrish01> grisbi looks much more complex than it needs to be
04:48 < Tv> I've been reasonably happy with gnucash, barring a few display corruption and crash bugs.
04:49 < chrish01> gnucash is single handedly the worst ui design in linux
04:49 < chrish01> just like quicken is the worst in windows and mac
04:50 < mael> well we I still used that sort of stuff under windows, I prefered quicken to ms money
04:50 < mael> it was back in the 90's though
04:51 < chrish01> as much as i hate ms's business tacticts, i think ms money was one of their under-rated apps
04:53 < chrish01> UI design is definitely an art. not everyone has it
05:00 < mael> chrish01:  btw, "actions" are not greyed when selecting domain0, does it means you can stop dom0? ;)
05:01 < mael> mmh yes you can
05:01  * mael just tested
05:01 < Tv> that should be fun ;)
05:02 < Tv> although I guess one future design would be multiple dom0's, for HA..
05:02 < mael> well I just have to restart the machine :)
05:04 < chrish01> hehe
05:04 < chrish01> mael, that is a simple check to prevent that
05:10 < chrish01> ok its late. bedtime
05:11 < mael> g'night
05:11 -!- chrish01 [~chrish01@72-254-24-162.client.stsn.net] has left #xen [Leaving]
05:47 -!- kmacy [~kmacy@adsl-69-228-221-109.dsl.pltn13.pacbell.net] has joined #xen
05:53 -!- yarihm [~yarihm@vpn-global-dhcp1-232.ethz.ch] has joined #xen
05:58 -!- athomas [~athomas@hardpress.demon.co.uk] has joined #xen
06:14 -!- kmacy [~kmacy@adsl-69-228-221-109.dsl.pltn13.pacbell.net] has quit [Quit: Leaving]
06:24 -!- Manix [~mark@office.dl.lcn.com] has joined #xen
06:31 -!- sunny [mdkuser@lexington.opencurve.org] has quit [Quit: leaving]
06:32 -!- sunny [mdkuser@lexington.opencurve.org] has joined #xen
06:32 -!- frealek [~nobo@4va54-1-81-56-99-20.fbx.proxad.net] has quit [Quit: bbl]
06:49 -!- masm [~masm@c-217-70-66-127.bragatel.pt] has quit [Read error: No route to host]
07:07 -!- mejlholm [~mejlholm@port79.ds1-abc.adsl.cybercity.dk] has quit [Quit: Client exiting]
07:54 -!- masm [~masm@c-217-70-66-127.bragatel.pt] has joined #xen
09:47 -!- unriel is now known as riel
09:48 -!- sdague [~sdague@bi-02pt1.bluebird.ibm.com] has joined #xen
10:04 -!- jimix [~jimix@ip13.194.susc.suscom.net] has joined #xen
10:15 -!- DEac- [~deac@xdsl-81-173-160-152.netcologne.de] has quit [Ping timeout: 480 seconds]
10:16 -!- rharper [~rharper@pixpat.austin.ibm.com] has joined #xen
10:26 -!- DEac- [~deac@xdsl-213-196-198-145.netcologne.de] has joined #xen
10:32 -!- athomas [~athomas@hardpress.demon.co.uk] has quit [Quit: Leaving]
10:33 -!- hollis [~hollis@user-0vvde2g.cable.mindspring.com] has joined #xen
10:46 -!- aliguori [~anthony@cpe-70-112-81-91.austin.res.rr.com] has joined #xen
10:55 -!- masm [~masm@c-217-70-66-127.bragatel.pt] has quit [Quit: Leaving]
11:26 -!- yarihm [~yarihm@vpn-global-dhcp1-232.ethz.ch] has quit [Quit: Leaving]
11:30 -!- Ahma [Ahti@l43.ip1.netikka.fi] has joined #xen
11:32 -!- Ahma [Ahti@l43.ip1.netikka.fi] has quit [Quit: ]
11:42 -!- hollis [~hollis@user-0vvde2g.cable.mindspring.com] has quit [Quit: leaving]
11:56 -!- aliguori [~anthony@cpe-70-112-81-91.austin.res.rr.com] has quit [Quit: Leaving]
12:13 -!- hollis [~hollis@pixpat.austin.ibm.com] has joined #xen
12:14 -!- aliguori [~anthony@pixpat.austin.ibm.com] has joined #xen
12:42 -!- knewt [~jmb@p213.54.72.161.tisdip.tiscali.de] has quit [Quit: leaving]
12:53 -!- kmacy [~kmacy@adsl-69-228-221-109.dsl.pltn13.pacbell.net] has joined #xen
12:55 -!- Administrator_ [~Administr@ip-207-145-127-226.lax.megapath.net] has quit [Ping timeout: 480 seconds]
13:02 -!- niv [~nivedita@bi01p1.co.us.ibm.com] has joined #xen
13:10 -!- kmacy [~kmacy@adsl-69-228-221-109.dsl.pltn13.pacbell.net] has quit [Remote host closed the connection]
13:25 -!- Administrator_ [~Administr@ip-207-145-127-226.lax.megapath.net] has joined #xen
13:25 -!- enum [~Administr@ip-207-145-127-226.lax.megapath.net] has quit [Read error: Connection reset by peer]
13:25 -!- malahal [~malahal@bi01p1.co.us.ibm.com] has quit [Quit: Leaving]
13:26 -!- cdub [~chrisw@216-99-217-87.dsl.aracnet.com] has joined #xen
14:03 -!- DEac- [~deac@xdsl-213-196-198-145.netcologne.de] has quit [Quit: Verlassend]
14:10 -!- DEac- [~deac@xdsl-213-196-197-250.netcologne.de] has joined #xen
14:12 -!- tony [~tony@cerebus.immunix.com] has quit [Ping timeout: 480 seconds]
14:22 -!- tessier [~treed@146.82.146.22] has joined #xen
14:23 -!- woody_ [~woody@bi01p1.co.us.ibm.com] has quit [Quit: Leaving]
14:41 -!- chrish01 [~chrish01@wsip-70-183-17-66.oc.oc.cox.net] has joined #xen
15:01 < chrish01> so what are the plans for the tools in 3.0? will there be a switch to vm-tools being the main tools, or continue with xend?
15:03 < Administrator_> ooh.. good question
15:03 -!- Administrator_ is now known as enum
15:04 < chrish01> cause the boss said to go ahead and just try to target 3.0 with xenmanager first. so it would save me some time to know which toolset to wrap :)
15:04 < enum> yeah, I will be gearing the web manager towards xen 3.0
15:05 < chrish01> web manager?
15:05 < chrish01> you working on a new one?
15:05 < enum> I talked to you about it last ngiht 
15:05 < chrish01> oh ok
15:05 -!- sunny [mdkuser@lexington.opencurve.org] has quit [Quit: poof]
15:05 -!- sunny [mdkuser@lexington.opencurve.org] has joined #xen
15:05 < enum> is it a hopeless project?
15:05 < enum> I just started doing this cause I wanted a better one that the one out there
15:06 < chrish01> if you are thinking of updating the old xensv code i might say yes :)
15:06 < enum> no.. all new
15:06 < chrish01> ok, i can help you if you need it
15:07 < enum> 2.6.9-11 is the newest
15:07 < enum> oops
15:07 < enum> haha wrong window
15:07 < chrish01> np
15:09 < aliguori> chrish01: xend will be the main toolset for 3.0 barring some unforeseen event
15:09 < chrish01> ok
15:09 < chrish01> i still cant seem to get xend to listen on 8000 in 3.0
15:09 < chrish01> no clue why. everything seems copasetic in logs
15:10 < aliguori> enum: a web manager eh?
15:10 < aliguori> enum: do you have any code yet?
15:10 < enum> I have snippets I have played with.. but for 2.0, and nothing really web manageable yet
15:10 < aliguori> enum: any basic concepts?
15:10  * aliguori likes the idea of a web-based management system for xen
15:10 -!- JViz [Anomaly@cpe-065-190-033-248.triad.res.rr.com] has quit [Quit: YES THEY DESERVE TO DIE, AND I HOPE THEY BURN IN HELL!]
15:10 < enum> to be able to create, delete, and maintain all aspects of dom0 and guests via web interface :)
15:11 < enum> I just started thia a few days ago
15:11 < chrish01> enum, what language?
15:11  * chrish01 crosses fingers for not python
15:11 < enum> perl modules, executed by php web interface
15:12 < chrish01> oh wow :)
15:12 < enum> ?
15:12 < chrish01> why not a single language?
15:13 < enum> I am proficient in both of these.. and no reason to over complicate the matter :)
15:13 < aliguori> enum: hmmm...
15:14 < albeiro> can opensolaris run on xen ?
15:14 < albeiro> and what about solaris10 ?
15:14 < enum> aliguori, ?
15:14 < aliguori> that requires running a full apache daemon in dom0 doesn't it?
15:14 < aliguori> albeiro: not currently
15:14 < enum> not nesicarily
15:14 < chrish01> albeiro, soon :)
15:14 < enum> aliguori, what was xensv running off of?
15:14 < aliguori> i wouldn't hold your breath :-)
15:14 < chrish01> albeiro, there are some dudes at sun working on it, and as soon as VT/Pacifica comes out, xen 3.0 will too on those cpu's
15:14 < aliguori> i thought xensv used it's own tiny httpd server
15:15 < enum> k, I can work something out :)
15:15 < chrish01> xensv used twisted
15:15 < enum> still in process
15:15 < enum> because apaxche uses too much mem to run in dom0 unless you wanna give dom0 an assload of mem
15:15 < aliguori> i'm a big fan of minimalistic dom0s.
15:15 < enum> which I only give mine 128m
15:15 < aliguori> so if i were writing a web interface, i'd probably stick with c
15:16 < aliguori> even though it would be more painful
15:16 < chrish01> enum, id like to get dom0 down to like 32mb
15:16 < albeiro> chrish01: never heard about VT/Pacifica
15:16 < chrish01> albeiro, allows virtualization to be done in the cpu
15:16 < chrish01> and effectively not require a ported guest OK
15:16 < chrish01> OS
15:16 < aliguori> albeiro: hardware extensions to Intel/AMD to allow full virtualization
15:16 < aliguori> only for 64 bit processors
15:16 < aliguori> AFAIK
15:17 < chrish01> aliguori, are the vt/pacifica only going to be 64bit cpus?
15:17 < albeiro> ah, heard about it
15:17 < aliguori> chrish01: um, it's been a while since i read the vt spec.  but the pacifica spec is definitely 64-bit centric
15:17 < albeiro> any ETA ?
15:17 < aliguori> i'm pretty sure the vt one is too
15:17 < albeiro> leaks from intel ? ;p
15:18 < chrish01> albeiro, i think feature freeze in august/september and i would say 2 months for bug hunting
15:18 < aliguori> albeiro: i think it's publicly known that there is actual vt hardware (since status is being posted to the xen-devel list of nightly builds) so it can't be that far off
15:19 < albeiro> nice
15:19 < chrish01> so the new control tools seem to be in a fork of xen-unstable and only available via cvs?
15:19 < albeiro> i am buing it when possible ;)
15:19 < chrish01> same
15:19 < albeiro> and what does it ahve in common with solaris @ xen ?
15:19 < albeiro> sun is waiting for it ?
15:19 < chrish01> no
15:20 < chrish01> it will allow us to run solaris unmodified ontop of xen (assuming there will be a 64bit build for the new cpu)
15:20 < aliguori> chrish01: define "new control tools".  there's both a new xend and then there's vm-tools
15:20 < chrish01> aliguori, xend changes done by mike wray at hp
15:20 < albeiro> i see. keep up good work :)
15:21 < murble> enum: you probably want to give dom0 lots of memory anyway well atleast if you are running lvm or raid etc.
15:22 < aliguori> murble: why?
15:22 < murble> aliguori: well not lots, but plenty enough to have an apache as well if you wanted.
15:22 < chrish01> no way am i running apache in my dom0
15:22 < aliguori> murble: why do you need lots of memory for lvm?
15:22 < chrish01> personally, i think a web frontend is almost too much
15:22 < enum> I am looking into smaller more mem friednly http daemons.. since this interface will be an admin tool, and not serving actual "internet pages"
15:22 < murble> aliguori: well device-mapper tends to get starved when making snapshots etc
15:22 < murble> of memory and then oopses
15:23 < murble> which isn't much fun.
15:23 < murble> pvmove etc.
15:23 < chrish01> enum, if you use RubyOnRails, they have a self hosted web server you can use :-)
15:23 < chrish01> and then you can get some super nice AJAX while you are at it
15:24 < enum> hmm.. all good suggestions. I will look into all of them
15:24 < enum> I will be downloading a copy of unstable tonight to play with
15:24 < enum> chrish01, did anyone answer about vm-tools?
15:25 < aliguori> murble: ah, snapshots, okay :-)
15:25 < chrish01> enum, i believe xend will still be default
15:25 < aliguori> enum: answer what about vm-tools?
15:25 < aliguori> oh, yes, xend will still be default
15:25 < enum> rockin
15:25 < aliguori> really, xend should be a separate project IMHO
15:25 < chrish01> i want to target users with xenmanager that only have a default install of xen. otherwise, not as many people will use it
15:26 < enum> btw, this proceess will also be going into a c-panel type app I am working on.. since I use xen for web hosting :)
15:26 < chrish01> nice
15:28 < murble> enum: c-panel type applications shoudl be taken out and shot.
15:29 < enum> yeah. but the reailty is they get used a lot
15:29 < enum> and are overpriced
15:29 < murble> anyone who needs such an application shouldn't be a customer of mine.
15:29 < murble> because i know i'll have lots of work when they get their instance/machine compromised etc.
15:29 < murble> or they forget to make backups and the disk dies etc.
15:29 < murble> here sir is your new image.
15:31 < chrish01> i think a system like windows MMC for linux would be nice
15:31 < chrish01> but impossible with distro differences
15:33 < enum> so LFS it.. if it only needs to serve one purpose.. make the baselayout and modify it
15:33 < enum> then put it on cd :)
15:33 < chrish01> hmm, seems Tools NG has plans for a web service API
15:42 < chrish01> man, if only vm-tools was in gobject
15:42 < chrish01> then i could automate the wrappers completely
15:42 < aliguori> chrish01: =-D
15:42 < aliguori> chrish01: I was just having this exact conversation
15:42 < chrish01> oh?
15:43 < chrish01> the gapi generaters for python and mono are great
15:43 < aliguori> I'm working on an internal proposal to create a gobject-like API for Xen.  Still trying to convince folks like sdague that it's the right approach :-)
15:43 < hollis> what is gobject?
15:43 < aliguori> although you suggesting that independently may help my argument :-)
15:43 < chrish01> gobject is a great object hierarchy system for C
15:43 < chrish01> brings OO to C basically
15:44 < aliguori> hollis: it's an object mechanism for C used by glib and therefore gtk and most gtk apps.  has lots of nice features like delayed binding, sig/slots, etc.
15:44 < hollis> now I *have* to see this :)
15:44 < chrish01> gobject is the shit
15:45 < aliguori> hollis: poke through gaim's source code if you want a feel for it..
15:45 < aliguori> but yeah, connecting signals to a domain object for xen events would be fantastic
15:45 < chrish01> or even better ... gnumeric
15:45 < chrish01> gnumeric is prolly some of the best code ever written for desktops
15:46 < chrish01> aliguori, no doubt
15:46  * aliguori wonders if chrish01 is a gnumeric developer :-)
15:46 < chrish01> aliguori, nope
15:46 < chrish01> just used it to learn gobject myself
15:46 < aliguori> plus, gobjects create very nice bindings for high level languages...
15:54 < chrish01> brb
15:54 -!- chrish01 [~chrish01@wsip-70-183-17-66.oc.oc.cox.net] has quit [Quit: Leaving]
15:59 -!- chrish01 [~chrish01@wsip-70-183-17-66.oc.oc.cox.net] has joined #xen
15:59 -!- chrish01 is now known as chris|lunch
16:14 < sdague> aliguori: I'm being more convinced after reading more.  I just want to make sure it isn't *different enough* from standard C that it would confuse people, as we really want to have as many users as possible :)
16:15 < aliguori> sdague: :-)
16:28 -!- tessier [~treed@146.82.146.22] has quit [Quit: Leaving]
16:43 < cilkay> enum, I just noticed your comments about a Cpanel type app and the http server above. Are you somewhat familiar with Twisted?
16:43 -!- joink_ [~joink@186.80-202-132.nextgentel.com] has joined #xen
16:45 -!- joink [~joink@186.80-202-132.nextgentel.com] has quit [Ping timeout: 480 seconds]
16:48 -!- chris|lunch is now known as chrish01
16:51 < chrish01> aliguori, so what do you think the chances are of a gobject vm-tools?
16:55 -!- hollis [~hollis@pixpat.austin.ibm.com] has quit [Quit: later]
16:55 -!- rharper [~rharper@pixpat.austin.ibm.com] has quit [Quit: Leaving]
17:22 -!- JViz [Anomaly@cpe-065-190-033-248.triad.res.rr.com] has joined #xen
17:25 -!- chrish01 [~chrish01@wsip-70-183-17-66.oc.oc.cox.net] has quit [Quit: Leaving]
17:32 -!- chrish01 [~chrish01@wsip-70-183-17-66.oc.oc.cox.net] has joined #xen
17:32 -!- riel is now known as unriel
17:34 -!- sdague [~sdague@bi-02pt1.bluebird.ibm.com] has quit [Ping timeout: 480 seconds]
18:05 < aliguori> chrish01: dunno, we're still thinking about it
18:05 < chrish01> aliguori, well, i hope you get the chance to do it :)
18:10 -!- mikegrb [~michael@mikegrb.netop.oftc.net] has quit [Ping timeout: 480 seconds]
18:12 < aliguori> chrish01: right now i'm contemplating whether to use glib or just try to write an object system from scratch.  glib as a dependency kind of sucks for a system toool
18:12 < newsham> do you need a full blown object system like glib?
18:13 < newsham> or can you get away with just a few method pointers in a struct?
18:13 < aliguori> newsham: that's my exact thought
18:14 < newsham> lots of small object-like systems in unix kernels.  thankfully they havent pulled in glib yet ;-)
18:14 < newsham> (not that I have anything against glib)
18:16 -!- mikegrb [~michael@mikegrb.netop.oftc.net] has joined #xen
18:19 -!- chrish01 [~chrish01@wsip-70-183-17-66.oc.oc.cox.net] has quit [Ping timeout: 480 seconds]
18:25 < enum> cilkay, not particularly.. my first experience with it is installing xen :)
18:25 -!- chrish01 [~chrish01@wsip-70-183-17-66.oc.oc.cox.net] has joined #xen
18:25 < enum> but I am going to look into it for sure
18:25 < chrish01> what'd i miss? :-)
18:25 < cilkay> enum, you should. Twisted already has a means of doing distributed objects, Spread.
18:25 < enum> cilkay, can you exlaborate a bit more?
18:26 < cilkay> Using Spread, I believe it would be perfectly feasible to write a rich client application that can control the VMs.
18:27 < cilkay> Python has bindings for a bunch of GUI toolkits.
18:27 -!- surriel is now known as riel
18:27 < cilkay> You can use a rich GUI client that instantiates remote objects via Spread.
18:27 < chrish01> i was using the xend interface to talk to xend 2.0 server (for xenmanager). it worked pretty well and xend uses twisted
18:28 < chrish01> unfortunately, does seem to work well (or at all) for 3.0
18:28 < cilkay> yes, but spread gives you other possibilities
18:28 < chrish01> whats spread?
18:30 < cilkay> http://twistedmatrix.com/projects/core/spread
18:30 < chrish01> ty
18:31 < cilkay> If you control both ends of the wire, which in this case you would, I think this would be a good approach.
18:32 < cilkay> There is a lot of depth to the Twisted project so before implementing things in other languages/technologies, looking at what Twisted has to offer would be a good idea.
18:33 < cilkay> I can see having a need for a pure browser interface for controlling VMs but a rich GUI client would also be quite useful.
18:33 < chrish01> thats exactly what xend did
18:33 < cilkay> yes
18:33 < cilkay> Well, not really.
18:33 < chrish01> which worked very well, but still sucks
18:33 < cilkay> How?
18:33 < chrish01> cause its a pain in the ass to do anything but python
18:34 < chrish01> not all of us *want* to use python
18:34 < chrish01> i would rather see something written in gobject
18:34 < cilkay> ugh
18:34 < chrish01> that can easily be wrapped in a multitude of languages
18:34 < cilkay> I don't get it.
18:34 < cilkay> Python is written in C. You can embed it in whatever you want.
18:35 < chrish01> so
18:35 < cilkay> Is this is a religious issue with gobject or something?
18:35 < chrish01> i dont want to embed it
18:35 < chrish01> no, its just that gobject can be automagically wrapped to like 5 different languages
18:35 < chrish01> mono/python/ruby/etc ...
18:35 < cilkay> mono is not a language but I understand what you're saying
18:36 < chrish01> nod
18:36 < chrish01> most people dont recognize that, so i just advoid it
18:36 < chrish01> but in this case it is, since we only need to target the cil
18:36 < cilkay> Well, that's a different beast altogether.
18:36 < chrish01> naw, thats simple
18:36 < cilkay> Are you suggesting that dom0 should be running a Mono env?
18:37 < chrish01> any sort of *real* rpc system would be great
18:37 < chrish01> fuck no
18:37 < chrish01> but a xml-rpc system where i could interfact from a mono client is important
18:37 < cilkay> You do realize that implementing an xmlrpc server in Twisted is possible in very few lines of code, right?
18:37 < chrish01> this dumping of python lists like it is now is not good enough
18:38 < cilkay> You don't have to.
18:38 < chrish01> cilkay, i have never used twisted and plan to keep it that way
18:38 < cilkay> If xmlrpc is fine for you, you already have it.
18:38 < chrish01> but, that is good
18:38 < cilkay> You clearly have some sort of issues with Python, which is unfortunate, but Twisted already provides what you are looking for.
18:39 < chrish01> what im looking for? no. because i have no intension of writing that part
18:39 < chrish01> i am only concerned in writing a gui for people to actually use
18:39 < cilkay> You don't have to, xmlrpc server is there.
18:40 < cilkay> chrish01, I've been down this road. It is a continuum really. I started out doing xmlrpc on a project I'm working on and eventually factored it out as there is no need for it if you control both ends of the wire.
18:40 < chrish01> thats a very closed minded approach IMO
18:41 < cilkay> Security is an issue as xmlrpc really has no auth mechanism.
18:41 < chrish01> and xend currently does?
18:41 < chrish01> thats the problem to start out with
18:41 < cilkay> Twisted does.
18:41 < chrish01> there is no *right* way to do it unfortunately
18:41 < aliguori> chrish01: xend uses nothing
18:41 < chrish01> exactly
18:41 < cilkay> Did you know, e.g. that Twisted has an sshd implementation?
18:41 < aliguori> cilkay: are you a twisted developer?
18:41 < cilkay> no
18:42 < aliguori> ok, just checking :-)
18:42 < chrish01> cilkay, again. i have never used it. so ... that would be a big freakin' NO
18:43 < cilkay> I am not even a guru at it. But I have found it quite approachable and the folks on #twisted and #twisted.web have been very helpful.
18:44 < chrish01> i personally dont care what runs in domain 0 as long as it meets a few requirements. 1) lightweight and fast. 2) provides more than a printed python list
18:44 < chrish01> something that can reasonably be attached to from a thin-client
18:44 < cilkay> What's a "printed python list"?
18:44 < chrish01> have you seen the output of the xend?
18:44 < cilkay> no
18:44 < chrish01> telnet to your send server on 8000
18:44 < chrish01> GET /xend/node/ HTTP/1.0
18:44 < cilkay> Is it any different than a Python list?
18:45 < chrish01> s/send/xend
18:45 < chrish01> its pretty printed (indented and such)
18:46 < cilkay> I get a connection refused.
18:47 < cilkay> I can try it on localhost.
18:47 < chrish01> this will be on xen 2.0 of course
18:47 < cilkay> Bah, no telnet installed. Not going to bother.
18:47 < chrish01> 3.0 is a whole nother beast
18:47 < cilkay> Mine is 2x
18:48 < cilkay> chrish01, so what you prefer as an output?
18:48 < cilkay> An xmlrpc response?
18:49 < chrish01> that would be great (i realize its a slightly more overhead) but that could be fixed with a GET var
18:49 < cilkay> I don't think the overhead would be an issue at all.
18:50 < cilkay> Though it is more verbose than a python list, we're not talking about anything particularly resource intensive.
18:51 < chrish01> nod
18:59 < enum> cilkay, do you know if twisted.web is inastalled with twisted by default?
19:03 < enum> :w
19:05 -!- aliguori [~anthony@pixpat.austin.ibm.com] has quit [Quit: Leaving]
19:11 -!- hollis [~hollis@user-0vvde2g.cable.mindspring.com] has joined #xen
19:39 -!- niv [~nivedita@bi01p1.co.us.ibm.com] has left #xen [Leaving]
19:52 -!- chrish01 [~chrish01@wsip-70-183-17-66.oc.oc.cox.net] has quit [Quit: Leaving]
20:20 -!- chrish01 [~chrish01@wsip-70-183-17-66.oc.oc.cox.net] has joined #xen
20:32 < newsham> Is there any 3.0 specific documentation yet?  or does one simply read the 2.0 docs and then follow the src?
20:33 < chrish01> there is very little docs
20:33 < chrish01> and much things have changed since 2.0
20:33 < chrish01> especially developer docs (obviously)
20:34 < newsham> I'm supposed to be doing some work with 3.0 in the near future and I'm still somewhat green..  
20:34 < newsham> what would you recommend?
20:34 < chrish01> i would recommend getting familiar with 2.0 first
20:34 < chrish01> 3.0 wont be out until september at the earliest
20:35 < chrish01> that is more like the feature freeze date
20:35 < newsham> so the interface.pdf from the web page then...
20:35 < chrish01> i would start with user.pdf
20:35 < newsham> (the project entails taking some code that currently runs in 2.0 [i have it installed and running already] and updating it)
20:35 < chrish01> what type of work? developing/administration/etc?
20:35 < newsham> developement.
20:35 < newsham> I already have 2.0 installed and running, and 3.0 installed (in a vmware) for work
20:36 < chrish01> ah. well i would start by learning the 2.0 api while you can. unfortunately the 3.0 api is a moving timeline
20:36 < newsham> (or -unstable if you prefer)
20:36 < newsham> *nod* thats what I've been told.
20:36 < chrish01> im developing a gtk gui for it right now and its increasingly difficult
20:36 < newsham> ok, thanks for the advice.. i got some reading to do :)
20:37 < chrish01> yea. does your development involve working with the xend api?
20:37 < chrish01> ie: provisioning, etc ...
20:37 < newsham> no, targetting an OS to a user domain
20:38 < newsham> which already runs (somewhat stably) in 2.0
20:38 < chrish01> well, things should run pretty much the same in -unstable as -stable
20:38 < chrish01> its only the control interfaces that will be changing
20:38 < chrish01> ie: new event channels, etc ...
20:38 < newsham> ahh.. 
20:44 -!- hollis [~hollis@user-0vvde2g.cable.mindspring.com] has quit [Quit: leaving]
21:32 -!- chrish01 [~chrish01@wsip-70-183-17-66.oc.oc.cox.net] has quit [Ping timeout: 480 seconds]
21:34 -!- chrish01 [~chrish01@wsip-70-183-17-66.oc.oc.cox.net] has joined #xen
21:43 -!- jimix [~jimix@ip13.194.susc.suscom.net] has quit [Quit: jimix]
22:19 -!- chrish01 [~chrish01@wsip-70-183-17-66.oc.oc.cox.net] has quit [Remote host closed the connection]
22:59 -!- sunny [mdkuser@lexington.opencurve.org] has quit [Quit: leaving]
22:59 -!- sunny [mdkuser@lexington.opencurve.org] has joined #xen
23:25 -!- dwh [~sc@c-24-21-80-144.hsd1.or.comcast.net] has joined #xen
--- Log closed Fri Jun 17 23:59:00 2005