TheGrebs.com:IRC Logs:#xen:2007:02:06

---	Log	opened Tue Feb 06 00:00:04 2007
00:01	\|-\|	Netsplit over, joins: mejlholm\|uni
00:01	\|-\|	LostFrogRL [~qscript@wsip-68-225-90-115.dc.dc.cox.net] has joined #xen
00:01	\|-\|	Netsplit over, joins: jwb_, VS_ChanLog
00:11	\|-\|	LostFrogRL [~qscript@wsip-68-225-90-115.dc.dc.cox.net] has quit [Read error: Connection reset by peer]
00:18	\|-\|	Netsplit synthon.oftc.net <-> larich.oftc.net quits: VS_ChanLog, jwb_, mejlholm\|uni
00:30	\|-\|	Netsplit over, joins: VS_ChanLog
00:46	\|-\|	aw [~awilliam@c-24-9-84-32.hsd1.co.comcast.net] has joined #xen
00:46	\|-\|	jwb_ [~jwboyer@71-36-216-66.roch.qwest.net] has joined #xen
00:46	\|-\|	LostFrogRL [~qscript@wsip-68-225-90-115.dc.dc.cox.net] has joined #xen
00:46	\|-\|	mdday [~mdday@cpe-024-163-122-235.nc.res.rr.com] has joined #xen
00:46	\|-\|	aliguori [~anthony@cpe-70-112-17-156.austin.res.rr.com] has joined #xen
00:46	\|-\|	Jaff [~jk@bart.cs.aau.dk] has joined #xen
00:46	\|-\|	paelzer [~kvirc@blueice1n1.de.ibm.com] has joined #xen
00:46	\|-\|	brendan [~brendan@67.17.128.36] has joined #xen
00:46	\|-\|	MackAck [~Mack@80.178.86.105.adsl.012.net.il] has joined #xen
00:46	\|-\|	nix [~nix@spirit.infernix.net] has joined #xen
00:46	\|-\|	visik7 [~visi@host178-38-dynamic.3-87-r.retail.telecomitalia.it] has joined #xen
00:46	\|-\|	Maxou [~max@212.85.158.10] has joined #xen
00:46	\|-\|	segher [~segher@dslb-084-056-129-197.pools.arcor-ip.net] has joined #xen
00:46	\|-\|	Basic_py [~Basic@warden.real-time.com] has joined #xen
00:46	\|-\|	matti [~matti@83.64.128.18] has joined #xen
00:47	\|-\|	matti changed nick to Guest1632
01:25	\|-\|	visik7 [~visi@host178-38-dynamic.3-87-r.retail.telecomitalia.it] has quit []
02:15	\|-\|	mulix [~mulix@87.69.40.180.cable.012.net.il] has quit [Remote host closed the connection]
02:22	\|-\|	mastermind [~mastermin@mastermind.kaltenbrunner.cc] has quit [Quit: Client exiting]
02:53	\|-\|	rock_ [~rock@82-32-43-245.cable.ubr06.azte.blueyonder.co.uk] has joined #xen
02:55	<rock_>	Hi there. If I reboot the host only 3 out of 6 DomU's start up. Where can I look to fix this?
03:06	\|-\|	visik7 [~visi@host178-38-dynamic.3-87-r.retail.telecomitalia.it] has joined #xen
03:12	\|-\|	mejlholm\|uni [~mejlholm@bart.cs.aau.dk] has joined #xen
03:34	\|-\|	cableman [~cableman@3e6b7156.rev.stofanet.dk] has joined #xen
03:52	\|-\|	rock_ [~rock@82-32-43-245.cable.ubr06.azte.blueyonder.co.uk] has quit [Ping timeout: 480 seconds]
03:54	\|-\|	rock_ [~rock@82-32-43-245.cable.ubr06.azte.blueyonder.co.uk] has joined #xen
04:27	\|-\|	DoberMann [~james@AToulouse-156-1-118-89.w90-30.abo.wanadoo.fr] has joined #xen
04:31	\|-\|	athomas [~athomas@host86-136-133-229.range86-136.btcentralplus.com] has joined #xen
04:33	\|-\|	zez [~zez@217-133-59-42.b2b.tiscali.it] has joined #xen
05:23	\|-\|	bernarde [~bernarde@32.104.18.240] has joined #xen
05:28	\|-\|	nix changed nick to infernix
05:35	\|-\|	rock_ [~rock@82-32-43-245.cable.ubr06.azte.blueyonder.co.uk] has quit [Ping timeout: 480 seconds]
05:43	\|-\|	zez [~zez@217-133-59-42.b2b.tiscali.it] has quit [Remote host closed the connection]
05:46	\|-\|	_bernarde [~bernarde@32.104.18.240] has joined #xen
05:46	\|-\|	_bernarde [~bernarde@32.104.18.240] has quit [Remote host closed the connection]
05:47	\|-\|	_bernarde [~bernarde@32.104.18.240] has joined #xen
05:47	\|-\|	bernarde [~bernarde@32.104.18.240] has quit [Read error: Connection reset by peer]
06:24	\|-\|	cableman [~cableman@3e6b7156.rev.stofanet.dk] has quit [Remote host closed the connection]
06:27	\|-\|	LostFrog [~qscript@wsip-68-225-90-115.dc.dc.cox.net] has joined #xen
06:27	\|-\|	LostFrogRL [~qscript@wsip-68-225-90-115.dc.dc.cox.net] has quit [Read error: Connection reset by peer]
06:27	\|-\|	rock_ [~rock@82-32-43-245.cable.ubr06.azte.blueyonder.co.uk] has joined #xen
06:50	\|-\|	weasel [weasel@weasel.noc.oftc.net] has quit [Quit: reboot]
06:55	\|-\|	weasel [weasel@asteria.debian.or.at] has joined #xen
07:10	\|-\|	athomas [~athomas@host86-136-133-229.range86-136.btcentralplus.com] has quit [Quit: Leaving]
07:14	\|-\|	athomas [~athomas@host86-136-133-229.range86-136.btcentralplus.com] has joined #xen
07:17	\|-\|	cableman [~cableman@3e6b7156.rev.stofanet.dk] has joined #xen
07:19	\|-\|	cableman [~cableman@3e6b7156.rev.stofanet.dk] has quit [Remote host closed the connection]
07:21	\|-\|	bernarde_ [~bernarde@32.104.18.240] has joined #xen
07:22	\|-\|	bernarde_ [~bernarde@32.104.18.240] has quit [Remote host closed the connection]
07:22	\|-\|	bernarde [~bernarde@32.104.18.240] has joined #xen
07:25	\|-\|	zez [~zez@217-133-59-42.b2b.tiscali.it] has joined #xen
07:25	\|-\|	_bernarde [~bernarde@32.104.18.240] has quit [Ping timeout: 480 seconds]
07:31	\|-\|	msinhore [~msinhore@mail.samurai.com.br] has joined #xen
07:48	\|-\|	danp1 [~berrange@nat-pool-bos.redhat.com] has joined #xen
07:49	\|-\|	agrif_away changed nick to agriffis
07:53	\|-\|	agriffis changed nick to agrif_away
07:53	\|-\|	athomas_ [~athomas@host86-136-133-229.range86-136.btcentralplus.com] has joined #xen
07:57	\|-\|	athomas [~athomas@host86-136-133-229.range86-136.btcentralplus.com] has quit [Ping timeout: 480 seconds]
08:03	\|-\|	cdub [~chrisw@wlanconf-nat-pool-bos.redhat.com] has joined #xen
08:23	\|-\|	zez_ [~zez@217-133-59-42.b2b.tiscali.it] has joined #xen
08:23	\|-\|	zez [~zez@217-133-59-42.b2b.tiscali.it] has quit [Read error: Connection reset by peer]
08:26	\|-\|	rharper [~rharper@bi01p1.co.us.ibm.com] has joined #xen
08:31	\|-\|	visik7 [~visi@host178-38-dynamic.3-87-r.retail.telecomitalia.it] has quit [Remote host closed the connection]
08:32	\|-\|	agrif_away changed nick to agriffis
08:32	\|-\|	visik7 [~visi@host178-38-dynamic.3-87-r.retail.telecomitalia.it] has joined #xen
08:37	\|-\|	rharper [~rharper@bi01p1.co.us.ibm.com] has quit [Ping timeout: 480 seconds]
08:41	\|-\|	KriP [kp@yancey.unixworld.dk] has joined #xen
08:44	\|-\|	agriffis changed nick to agrif_away
08:53	\|-\|	DavidS [~david@pnsgw3-client211.demo.tuwien.ac.at] has joined #xen
08:54	\|-\|	rock_ [~rock@82-32-43-245.cable.ubr06.azte.blueyonder.co.uk] has quit []
08:54	\|-\|	rock_ [~rock@82-32-43-245.cable.ubr06.azte.blueyonder.co.uk] has joined #xen
08:59	\|-\|	hollisb [~hollisb@bi01p1.co.us.ibm.com] has joined #xen
09:03	\|-\|	bestorga [~bestorga@c-67-189-96-160.hsd1.or.comcast.net] has joined #xen
09:15	<rock_>	Hi, when I boot the host the guests now start automatically, however, on some DomU's I have to manually run 'route add -host 89.16.162.1 dev eth0' - where is the best place to put this so it happens on boot?
09:16	\|-\|	weaselTM [weasel@asteria.debian.or.at] has joined #xen
09:18	\|-\|	cdub_ [~chrisw@wlanconf-nat-pool-bos.redhat.com] has joined #xen
09:19	\|-\|	Netsplit synthon.oftc.net <-> larich.oftc.net quits: DavidS, VS_ChanLog, hollisb, cdub, bestorga, weasel
09:19	\|-\|	weaselTM changed nick to weasel
09:19	\|-\|	Netsplit over, joins: hollisb
09:27	\|-\|	bestorga [~bestorga@c-67-189-96-160.hsd1.or.comcast.net] has joined #xen
09:27	\|-\|	bernarde [~bernarde@32.104.18.240] has quit [Remote host closed the connection]
09:29	\|-\|	bernarde [~bernarde@32.104.18.240] has joined #xen
09:29	\|-\|	VS_ChanLog [~stats@ns.theshore.net] has joined #xen
09:30	\|-\|	cableman [~cableman@3e6b7156.rev.stofanet.dk] has joined #xen
09:32	\|-\|	tonyb [~tony@pythia.bakeyournoodle.com] has quit [Server closed connection]
09:32	\|-\|	tonyb [~tony@pythia.bakeyournoodle.com] has joined #xen
09:35	\|-\|	DoberMann_ [~james@AToulouse-156-1-118-89.w90-30.abo.wanadoo.fr] has joined #xen
09:36	\|-\|	Maxou_ [~max@edony.tuxfamily.net] has joined #xen
09:36	\|-\|	Netsplit synthon.oftc.net <-> xenon.oftc.net quits: jwb_, paelzer, brendan, aw, mdday, DoberMann, infernix, Guest1632, MackAck, Maxou, (+4 more, use /NETSPLIT to show all of them)
09:37	\|-\|	mejlholm [~mejlholm@port79.ds1-abc.adsl.cybercity.dk] has joined #xen
09:39	\|-\|	Netsplit over, joins: Basic_py
09:39	\|-\|	Netsplit over, joins: infernix
09:40	\|-\|	jwb_ [~jwboyer@71-36-216-66.roch.qwest.net] has joined #xen
09:40	\|-\|	aliguori [~anthony@cpe-70-112-17-156.austin.res.rr.com] has joined #xen
09:40	\|-\|	Jaff [~jk@bart.cs.aau.dk] has joined #xen
09:40	\|-\|	paelzer [~kvirc@blueice1n1.de.ibm.com] has joined #xen
09:40	\|-\|	brendan [~brendan@67.17.128.36] has joined #xen
09:40	\|-\|	MackAck [~Mack@80.178.86.105.adsl.012.net.il] has joined #xen
09:40	\|-\|	segher [~segher@dslb-084-056-129-197.pools.arcor-ip.net] has joined #xen
09:41	\|-\|	aw [~awilliam@c-24-9-84-32.hsd1.co.comcast.net] has joined #xen
09:46	\|-\|	weasel [weasel@weasel.noc.oftc.net] has quit [Remote host closed the connection]
09:46	\|-\|	matti_ [matti@acrux.romke.net] has joined #xen
09:46	\|-\|	rharper [~rharper@bi01p1.co.us.ibm.com] has joined #xen
09:46	\|-\|	hollisb_ [~hollisb@bi01p1.co.us.ibm.com] has joined #xen
09:47	\|-\|	Netsplit charon.oftc.net <-> neutron.oftc.net quits: Basic_py, hollisb, bestorga, mejlholm, infernix
09:47	\|-\|	paelzer [~kvirc@blueice1n1.de.ibm.com] has quit [Quit: KVIrc 3.2.0 'Realia']
09:48	\|-\|	mdday [~mdday@cpe-024-163-122-235.nc.res.rr.com] has joined #xen
09:50	\|-\|	visik7 [~visi@host178-38-dynamic.3-87-r.retail.telecomitalia.it] has quit [Read error: Connection reset by peer]
09:52	\|-\|	mejlholm [~mejlholm@port79.ds1-abc.adsl.cybercity.dk] has joined #xen
09:53	\|-\|	visik7 [~visi@host178-38-dynamic.3-87-r.retail.telecomitalia.it] has joined #xen
09:56	\|-\|	infernix [~nix@spirit.infernix.net] has joined #xen
09:57	\|-\|	sputhenp [~sputhenp@202.80.58.210] has joined #xen
09:57	\|-\|	bestorga [~bestorga@c-67-189-96-160.hsd1.or.comcast.net] has joined #xen
09:57	<DoberMann_>	does a Xen 3.04 live CD exists ? and where can i find it ?
09:59	<DoberMann_>	i can't find any information dealing with that
10:00	\|-\|	weasel [weasel@asteria.debian.or.at] has joined #xen
10:00	\|-\|	Basic_py [~Basic@warden.real-time.com] has joined #xen
10:01	\|-\|	cableman [~cableman@3e6b7156.rev.stofanet.dk] has quit [Remote host closed the connection]
10:02	\|-\|	rharper [~rharper@bi01p1.co.us.ibm.com] has quit [Ping timeout: 480 seconds]
10:02	\|-\|	VS_ChanLog [~stats@ns.theshore.net] has quit [Ping timeout: 480 seconds]
10:06	\|-\|	agrif_away changed nick to agriffis
10:08	\|-\|	Netsplit charon.oftc.net <-> neutron.oftc.net quits: Basic_py, weasel
10:09	\|-\|	DrX0 [~DrX0@pool-71-106-181-83.lsanca.dsl-w.verizon.net] has joined #xen
10:15	\|-\|	ahs3_away changed nick to ahs3
10:18	\|-\|	DrX0 [~DrX0@pool-71-106-181-83.lsanca.dsl-w.verizon.net] has quit [Ping timeout: 480 seconds]
10:19	\|-\|	DrX0 [~DrX0@pool-71-105-109-220.lsanca.dsl-w.verizon.net] has joined #xen
10:20	\|-\|	Netsplit over, joins: Basic_py
10:34	\|-\|	Netsplit synthon.oftc.net <-> xenon.oftc.net quits: segher, jwb_, MackAck, aliguori, brendan, Jaff
10:36	\|-\|	Netsplit over, joins: Jaff
10:41	\|-\|	cableman [~cableman@3e6b6dca.rev.stofanet.dk] has joined #xen
10:41	\|-\|	rharper [~rharper@bi01p1.co.us.ibm.com] has joined #xen
10:44	\|-\|	mulix [~mulix@87.69.40.180.cable.012.net.il] has joined #xen
10:48	\|-\|	DavidS [~david@pnsgw1-client028.demo.tuwien.ac.at] has joined #xen
10:50	\|-\|	jwb_ [~jwboyer@71-36-216-66.roch.qwest.net] has joined #xen
10:50	\|-\|	MackAck [~Mack@80.178.86.105.adsl.012.net.il] has joined #xen
10:50	\|-\|	segher [~segher@dslb-084-056-129-197.pools.arcor-ip.net] has joined #xen
10:50	\|-\|	brendan [~brendan@67.17.128.36] has joined #xen
10:53	\|-\|	hackjack82 [~jack@137.204.24.128] has joined #xen
10:53	\|-\|	rharper [~rharper@bi01p1.co.us.ibm.com] has quit [Read error: Operation timed out]
10:53	<hackjack82>	hi,
10:54	<hackjack82>	i'm tring to measure VM downtime during a live migration,i tried measure ping -f but there must be a better way. any suggest?
10:56	<brendan>	iperf perhaps?
10:56	<brendan>	if it has a udp mode
10:56	<hackjack82>	iperf? to masure what?
10:56	\|-\|	aliguori [~anthony@bi01p1.co.us.ibm.com] has joined #xen
10:56	<brendan>	similar to ping -f, but a bit more controlled
10:57	<hackjack82>	with iperf can i measure a ms downtime?
10:59	<brendan>	hum, I guess it's still a bit of a hack. mxtraf would produce a nice graph
11:00	<icblenke>	it looks like xen/sv/DomInfo.py calls server.xend_domain_migrate
11:01	<icblenke>	my guess is that xend_domain_migrate is a libxc thing.
11:02	<brendan>	it calls xc_save with a live flag.
11:02	<hackjack82>	so you advice to try mxtraf?
11:02	<hackjack82>	instead of iperf?
11:02	<brendan>	there's probably an easier way. mxtraf is nice looking, but not well packaged
11:02	<icblenke>	you could just doctor the python to spit out a timestamp before and after the xend_domain_migrate.
11:03	<brendan>	python doesn't know when the domain has actually been paused
11:03	<brendan>	you'd get a several-seconds gap
11:03	<icblenke>	granted, that wouldn't tell you the ms outage for the live migration, but rather the _entire_ migration interval.
11:03	<icblenke>	right.
11:03	<brendan>	you could doctor xc_save :)
11:04	<icblenke>	to do the ms gap, you'd need to modify libxc to spit out a timestamp before and after the last stage.
11:04	<brendan>	(really xc_linux_save)
11:04	<icblenke>	right.
11:05	<brendan>	I can't remember whether the xen live migration paper explains what it uses for measurement
11:05	<icblenke>	it doesn't appear to.
11:05	<hackjack82>	no they don't
11:05	<hackjack82>	i don't know python...
11:05	<icblenke>	I'd suspect some kind of libxc doctoring though.
11:05	<hackjack82>	i'm searching a simple solution
11:05	<icblenke>	hackjack82: do you know C?
11:05	\|-\|	DavidS [~david@pnsgw1-client028.demo.tuwien.ac.at] has quit [Quit: Leaving.]
11:06	<brendan>	it's probably best to have a process in the guest domain do the measuring
11:06	<brendan>	libxc may tell you it's done before the remote domain is really back up and running
11:06	<brendan>	though libxc for start + python for end might work
11:07	<hackjack82>	ok i know C even if I'm not a advanced....
11:07	<hackjack82>	inside the guest how can i measure the downtime?
11:08	<icblenke>	as long as the domU wasn't run with an independant_wallclock, you could just have the domU spit out millisecond timings and note the discrepancy "gap" in the output timings.
11:08	<brendan>	or have the guest send a packet to the host every millisecond, and let the host figure it out
11:09	<icblenke>	I like the idea of getting the guest's perspective.
11:09	<icblenke>	that, and you don't have to worry about networking delay.
11:09	<brendan>	in a LAN that should be less than a millisecond
11:09	<icblenke>	right. still...
11:09	<brendan>	and I trust the host's idea of time more :)
11:10	<icblenke>	the host's idea of time and the guest's idea of time are the same.
11:10	<hackjack82>	ok so it's the same ping -f way
11:10	<LostFrog>	Do you need a SAN for migration?
11:10	<icblenke>	unless you break them apart with independant wallclock.
11:10	<brendan>	icblenke: the remote host may have clock skew
11:10	<icblenke>	LostFrog: yes, or something like drbd.
11:10	<icblenke>	for live migration, anyway.
11:10	<brendan>	(or probably NFS root even)
11:10	<icblenke>	actually, for any migration. the block devices must be local.
11:11	\|-\|	weaselTM [weasel@2001:858:10f:6::1] has joined #xen
11:11	<hackjack82>	how can i measure the gap from the domU level? with gettimeofday i think i dont' have a ms precision
11:11	<icblenke>	brendan: if not timed from dom0, but from a "remote" box on the LAN, sure.
11:11	<icblenke>	er, wait, this is between servers.
11:12	<icblenke>	nevermind me.
11:13	<icblenke>	hackjack82: that's why you run ntp.
11:13	<LostFrog>	You should be able to get resolution up to 1/HZ
11:13	<LostFrog>	whatever HZ is for the kernel.
11:14	<hackjack82>	doing what?
11:14	\|-\|	weaselTM changed nick to weasel
11:14	<LostFrog>	oooh.. you want me to google for the actual function? :)
11:14	<icblenke>	the default HZ is usually 100 or 1000.
11:14	<hackjack82>	maybe... you will my angel :-) i'm quite desperate....
11:14	<hackjack82>	ntp?
11:15	<LostFrog>	I'm looking for a CD at the moment.. I don't know how manage to lose them.
11:15	<icblenke>	to sync the dom0 clocks.
11:15	\|-\|	jerone [~jerone@bi01p1.co.us.ibm.com] has joined #xen
11:16	<icblenke>	ideally, you just peer the two servers via ntp.
11:16	<hackjack82>	sorry I can't understand you...
11:17	<hackjack82>	please explain if you can....
11:17	<hackjack82>	I know i'm asking much...
11:17	<icblenke>	you can either monitor the availablity from a remote node and use that system's clock as the reference clock while you flood ping...
11:18	<icblenke>	or you can synchronize the system clocks between the two xen hosts and monitor the system time _inside_ the domU, recording the jiffy time and looking for discrepancies/jumps in the clock to know when you've been migrated.
11:19	<hackjack82>	using the ping method: i use the delta between each packet... but on the network i'm not sure the measured time is good....
11:20	<hackjack82>	with the second method....
11:20	<hackjack82>	how can I measure time gaps?
11:20	<hackjack82>	in C? gettimeofday?
11:20	<icblenke>	querying the system clock's jiffies.
11:21	<hackjack82>	ok but how?
11:21	<icblenke>	A timeval consists of two integers. Tv_sec specifies seconds since midnight, January 1st, 1970. On 32 bit systems, this value will wrap around sometime around 2038, if memory serves me correctly. Tv_usec specifies how many microseconds have elapsed of the current second.
11:21	<icblenke>	There is a system call available called gettimeofday which takes a pointer to a timeval as an argument, and fills it in with the current time.
11:22	<icblenke>	To save yourself the trouble of reimplementing the code to convert jiffies and cycle counts into a timeval, you may wish to call do_gettimeofday (defined in arch/i386/kernel/time.c), which is the architecture-specific function that does the coversion work for sys_gettimeofday (defined in kernel/time.c), the implementation of gettimeofday.
11:23	<hackjack82>	ok, I tried to make a simple C program with a gettimeofday -nanosleep-gettimeofday.... but it really doen's work...
11:24	<icblenke>	the larger problem for you will be getting the time of day more granularly than a jiffy
11:25	<hackjack82>	i have gaps of 3-4 ms....
11:25	<icblenke>	I doubt your jiffy will have a granularity of less than a millisecond
11:25	<icblenke>	and potentially as much as 10ms.
11:25	\|-\|	rpg [~rpg@bi01p1.co.us.ibm.com] has joined #xen
11:25	\|-\|	DoberMann_ changed nick to DoberMann[PullA]
11:25	<hackjack82>	doing a printf(tv_usec)
11:26	\|-\|	mastermind [~mastermin@mastermind.kaltenbrunner.cc] has joined #xen
11:26	<hackjack82>	so this is not a good method,is?
11:27	<icblenke>	gettimeofday() can return the calendar time with microseconds of precision, if your platform supports it.
11:27	<icblenke>	hackjack82: perhaps. perhaps not. it depends on your HZ value.
11:27	<hackjack82>	ok, it should do it.... but it doen's do....
11:27	<hackjack82>	if a do a nanosleep of 1 ms in a for cycle of gettimeofday
11:27	<icblenke>	http://tinyurl.com/29gxwd - read that article.
11:27	<hackjack82>	i lost many ms from one to other....
11:29	<hackjack82>	sorry i have to run away now....
11:29	<icblenke>	you're probably best timing from a remote host over the network with something like a flood ping.
11:29	<hackjack82>	i hope to find a solut
11:29	<hackjack82>	tnx with the help
11:29	<icblenke>	send 1000 packets a second, and you have millisecond timing.
11:29	<hackjack82>	very much
11:29	<hackjack82>	i will try
11:29	\|-\|	hackjack82 [~jack@137.204.24.128] has quit [Quit: Sto andando via]
11:31	\|-\|	athomas_ [~athomas@host86-136-133-229.range86-136.btcentralplus.com] has quit [Quit: Leaving]
11:41	\|-\|	cdub_ [~chrisw@wlanconf-nat-pool-bos.redhat.com] has quit [Read error: Connection reset by peer]
11:43	\|-\|	tonyb [~tony@pythia.bakeyournoodle.com] has quit [Read error: Operation timed out]
11:43	\|-\|	visik7 [~visi@host178-38-dynamic.3-87-r.retail.telecomitalia.it] has quit [Write error: connection closed]
11:43	\|-\|	sputhenp [~sputhenp@202.80.58.210] has quit [Read error: Connection reset by peer]
11:43	\|-\|	tonyb [~tony@pythia.bakeyournoodle.com] has joined #xen
11:43	\|-\|	visik7 [~visi@host178-38-dynamic.3-87-r.retail.telecomitalia.it] has joined #xen
11:44	\|-\|	rpg [~rpg@bi01p1.co.us.ibm.com] has quit [Read error: Operation timed out]
11:46	\|-\|	Netsplit synthon.oftc.net <-> scorpio.oftc.net quits: tonyb
11:48	\|-\|	plectrum changed nick to plectrum_away
11:50	\|-\|	Netsplit synthon.oftc.net <-> saturn.oftc.net quits: jwb_, mejlholm, brendan, mulix, hollisb_, bestorga, cableman, aw, DrX0, mastermind, (+10 more, use /NETSPLIT to show all of them)
12:01	\|-\|	sdague [~sdague@user-12lc6iq.cable.mindspring.com] has quit [Read error: Connection reset by peer]
12:03	\|-\|	VS_ChanLog [~stats@ns.theshore.net] has joined #xen
12:03	\|-\|	sdague [~sdague@user-12lc6iq.cable.mindspring.com] has joined #xen
12:03	\|-\|	Netsplit over, joins: Basic_py
12:04	\|-\|	Netsplit over, joins: visik7
12:04	\|-\|	Netsplit over, joins: mdday, mulix
12:04	\|-\|	Netsplit over, joins: cableman, MackAck
12:04	\|-\|	Netsplit over, joins: mastermind
12:04	\|-\|	jdmason [~jonmason@adsl-71-129-194-249.dsl.pltn13.pacbell.net] has joined #xen
12:04	\|-\|	Netsplit over, joins: segher
12:04	\|-\|	matti [matti@acrux.romke.net] has joined #xen
12:04	\|-\|	Netsplit over, joins: mejlholm
12:04	\|-\|	sputhenp [~sputhenp@202.80.58.210] has joined #xen
12:04	\|-\|	Netsplit over, joins: infernix, jwb_
12:04	\|-\|	Netsplit over, joins: aw
12:04	\|-\|	Netsplit over, joins: jerone, brendan
12:04	\|-\|	matti changed nick to Guest1673
12:04	\|-\|	Netsplit over, joins: DrX0
12:04	\|-\|	hollisb [~hollisb@bi01p1.co.us.ibm.com] has joined #xen
12:06	\|-\|	Netsplit over, joins: Jaff
12:09	\|-\|	rpg [~rpg@bi01p1.co.us.ibm.com] has joined #xen
12:11	\|-\|	pvanhoof [~pvanhoof@d54C0EE14.access.telenet.be] has joined #xen
12:12	\|-\|	aliguori [~anthony@bi01p1.co.us.ibm.com] has joined #xen
12:17	\|-\|	DoberMann[PullA] changed nick to DoberMann[Course2caddie]
12:18	\|-\|	zez_ changed nick to zez
12:20	\|-\|	Netsplit over, joins: tonyb
12:22	\|-\|	KriiP [kp@yancey.unixworld.dk] has joined #xen
12:23	\|-\|	LostFrog [~qscript@wsip-68-225-90-115.dc.dc.cox.net] has quit [Read error: Connection reset by peer]
12:26	\|-\|	zez_ [~zez@217-133-59-42.b2b.tiscali.it] has joined #xen
12:28	\|-\|	cdub [~chrisw@wlanconf-nat-pool-bos.redhat.com] has joined #xen
12:29	\|-\|	DoberMann[Course2caddie] changed nick to DoberMann
12:32	\|-\|	msinhore_ [~msinhore@mail.samurai.com.br] has joined #xen
12:34	\|-\|	Netsplit charon.oftc.net <-> kinetic.oftc.net quits: KriP, zez, msinhore
12:35	\|-\|	visik7 [~visi@host178-38-dynamic.3-87-r.retail.telecomitalia.it] has quit []
12:41	\|-\|	LostFrogRL [~qscript@wsip-68-225-90-115.dc.dc.cox.net] has joined #xen
12:46	\|-\|	Netsplit synthon.oftc.net <-> larich.oftc.net quits: jdmason, jwb_, cdub, sdague, msinhore_, zez_, mejlholm, brendan, mulix, VS_ChanLog, (+18 more, use /NETSPLIT to show all of them)
12:55	\|-\|	rock_ [~rock@82-32-43-245.cable.ubr06.azte.blueyonder.co.uk] has quit [Ping timeout: 480 seconds]
12:56	\|-\|	Netsplit over, joins: sdague
12:57	\|-\|	Netsplit over, joins: pvanhoof
12:57	\|-\|	Netsplit over, joins: aw, segher
12:57	\|-\|	Netsplit over, joins: Basic_py, mulix, mdday, infernix
12:57	\|-\|	matti [matti@acrux.romke.net] has joined #xen
12:57	\|-\|	dipwad [~nada@71-13-79-30.static.aldl.mi.charter.com] has joined #xen
12:57	\|-\|	matti changed nick to Guest1678
12:58	\|-\|	Netsplit over, joins: cdub
12:58	\|-\|	Netsplit over, joins: mejlholm
12:58	\|-\|	Netsplit over, joins: MackAck
12:58	\|-\|	Netsplit over, joins: cableman
12:59	\|-\|	Netsplit over, joins: jdmason
13:00	\|-\|	Netsplit over, joins: Jaff
13:00	\|-\|	dipwad [~nada@71-13-79-30.static.aldl.mi.charter.com] has quit [Remote host closed the connection]
13:02	\|-\|	rpg [~rpg@bi01p1.co.us.ibm.com] has joined #xen
13:03	\|-\|	mastermind [~mastermin@mastermind.kaltenbrunner.cc] has joined #xen
13:03	\|-\|	jwb_ [~jwboyer@71-36-216-66.roch.qwest.net] has joined #xen
13:03	\|-\|	sputhenp [~sputhenp@202.80.58.210] has joined #xen
13:06	\|-\|	brendan [~brendan@67.17.128.36] has joined #xen
13:11	\|-\|	LostFrogRL [~qscript@wsip-68-225-90-115.dc.dc.cox.net] has quit [Read error: Connection reset by peer]
13:13	\|-\|	aliguori [~anthony@bi01p1.co.us.ibm.com] has joined #xen
13:14	\|-\|	jerone [~jerone@bi01p1.co.us.ibm.com] has joined #xen
13:14	\|-\|	hbaum [~hbaum@bi01p1.co.us.ibm.com] has joined #xen
13:16	\|-\|	hollisb [~hollisb@bi01p1.co.us.ibm.com] has joined #xen
13:26	\|-\|	rharper [~rharper@bi01p1.co.us.ibm.com] has joined #xen
13:37	\|-\|	cableman [~cableman@3e6b6dca.rev.stofanet.dk] has quit [Quit: Leaving]
13:41	\|-\|	jimix [~jimix@67.189.187.152] has joined #xen
13:41	\|-\|	danp2 [~berrange@wlanconf-nat-pool-bos.redhat.com] has joined #xen
13:42	\|-\|	danp1 [~berrange@nat-pool-bos.redhat.com] has quit [Ping timeout: 480 seconds]
13:42	\|-\|	danp2 [~berrange@wlanconf-nat-pool-bos.redhat.com] has quit [Read error: Connection reset by peer]
13:55	\|-\|	jimix [~jimix@67.189.187.152] has quit [Ping timeout: 480 seconds]
13:55	\|-\|	visik7 [~visi@host178-38-dynamic.3-87-r.retail.telecomitalia.it] has joined #xen
13:58	\|-\|	rpg [~rpg@bi01p1.co.us.ibm.com] has quit [Read error: Operation timed out]
13:59	\|-\|	infernix [~nix@spirit.infernix.net] has quit [Quit: Leaving]
14:00	\|-\|	nix [~nix@spirit.infernix.net] has joined #xen
14:00	\|-\|	danp1 [~berrange@wlanconf-nat-pool-bos.redhat.com] has joined #xen
14:01	\|-\|	Netsplit synthon.oftc.net <-> xenon.oftc.net quits: nix
14:01	\|-\|	Netsplit over, joins: nix
14:05	\|-\|	Guest1678 changed nick to matti
14:05	\|-\|	cdub [~chrisw@wlanconf-nat-pool-bos.redhat.com] has quit [Read error: Connection reset by peer]
14:06	\|-\|	bernarde [~bernarde@32.104.18.240] has quit [Remote host closed the connection]
14:06	\|-\|	danp1 [~berrange@wlanconf-nat-pool-bos.redhat.com] has quit [Read error: Connection reset by peer]
14:07	\|-\|	cdub [~chrisw@wlanconf-nat-pool-bos.redhat.com] has joined #xen
14:07	\|-\|	danp1 [~berrange@wlanconf-nat-pool-bos.redhat.com] has joined #xen
14:08	<movement>	aliguori: there?
14:08	<aliguori>	movement: yup
14:09	<movement>	so how did your old xm/ssh stuff work?
14:09	\|-\|	danp1 [~berrange@wlanconf-nat-pool-bos.redhat.com] has quit [Read error: Connection reset by peer]
14:09	\|-\|	rpg [~rpg@bi01p1.co.us.ibm.com] has joined #xen
14:10	\|-\|	danp1 [~berrange@wlanconf-nat-pool-bos.redhat.com] has joined #xen
14:10	<aliguori>	movement: if you execute ssh xm serve, you can do xml-rpc over stdio of that connection
14:12	<icblenke>	now _that_ is cool.
14:12	<movement>	hrm
14:13	<movement>	is that approaching right?
14:14	<danp1>	movement: surely then you'd end up with two levels of auth - first PAM lookup done by SSH, and the 2nd done by XenD
14:14	<movement>	danp1: I don't see how to avoid it.
14:14	<danp1>	to be honest, i'm not entirely convinced XenD needs to / should be using PAM
14:15	<danp1>	at least not in any way which maps to the general system users
14:15	<icblenke>	can't you short circuit the pam config for the xenapi to always succeed?
14:15	<movement>	icblenke: then any user with local access can control xend - not good
14:15	<movement>	danp1: yes, see xen-api...
14:15	<danp1>	movement: yeah, its not conceptually the correct authentication domain
14:15	<movement>	it needs to use a PAM service other than 'login'
14:15	<icblenke>	right.
14:15	<danp1>	XenD users/ auth should not be related to general UNIX logins
14:16	<movement>	it's awkward. you have to have a secure transport AND pam stack stuff. I don't actually see the point in the password part of xend api at all
14:17	<movement>	there's no meaningful way to use it, since the transport must be secure anyway
14:17	<danp1>	movement: yeah if you've used SSH to access it ther'es little point
14:17	<danp1>	i've always wanted XenAPI to use SSL directly though
14:22	<aliguori>	movement: i absolute hate the fact that xend is maintaining it's own set of users
14:22	\|-\|	danp1 [~berrange@wlanconf-nat-pool-bos.redhat.com] has quit [Read error: Connection reset by peer]
14:23	<aliguori>	we should have just punted and let the management tools define the authenticated transport
14:23	\|-\|	danp1 [~berrange@wlanconf-nat-pool-bos.redhat.com] has joined #xen
14:26	<movement>	aliguori: yeah, that's what I just said to ewan :)
14:27	<aliguori>	:-)
14:27	<aliguori>	movement: Ewan was quite insistent about doing the authentication in the API..
14:27	\|-\|	lst [liquid@74-134-248-102.dhcp.insightbb.com] has joined #xen
14:28	<aliguori>	it's really very useless to us as we will have a CIM provider on every Xen machine anyway. the xen-api stuff will never see the network at all.
14:29	<lst>	this question may be better suited for here, if tweaking sysctl settings on a domU, does one need to also tweak the dom0? in this case, tcp settings primarily to scale a webserver that is running as the sole domU on the dom0.
14:29	\|-\|	danp1 [~berrange@wlanconf-nat-pool-bos.redhat.com] has quit [Read error: Connection reset by peer]
14:30	\|-\|	danp1 [~berrange@wlanconf-nat-pool-bos.redhat.com] has joined #xen
14:36	<icblenke>	bridged? no.
14:36	\|-\|	danp1 [~berrange@wlanconf-nat-pool-bos.redhat.com] has quit [Read error: Connection reset by peer]
14:36	\|-\|	zez [~zez@217-133-59-42.b2b.tiscali.it] has joined #xen
14:36	\|-\|	danp1 [~berrange@wlanconf-nat-pool-bos.redhat.com] has joined #xen
14:37	<icblenke>	even routed, I'd still say no. as the dom0's kernel isn't actually processing the TCP stream, just routing the IP datagrams along.
14:43	\|-\|	KriP [kp@yancey.unixworld.dk] has joined #xen
14:45	<hollisb>	bleh, this "compat" stuff is out of control
14:49	\|-\|	DrX0 [~DrX0@pool-71-105-109-220.lsanca.dsl-w.verizon.net] has joined #xen
14:49	\|-\|	danp1 [~berrange@wlanconf-nat-pool-bos.redhat.com] has quit [Read error: Connection reset by peer]
14:49	\|-\|	danp1 [~berrange@wlanconf-nat-pool-bos.redhat.com] has joined #xen
14:49	\|-\|	danp1 [~berrange@wlanconf-nat-pool-bos.redhat.com] has quit []
15:02	<lst>	ic: that's what i thought, changing the domU is seemingly having no effect though
15:03	<lst>	load is nominal, mem is nominal, but the webserver is still completely unresponsive and a ton of connections in time_wait
15:03	<lst>	or last_fin
15:04	\|-\|	Netsplit synthon.oftc.net <-> saturn.oftc.net quits: mastermind, hbaum, jerone, jwb_, rharper, hollisb, aliguori, sputhenp
15:04	\|-\|	jdmason [~jonmason@adsl-71-129-194-249.dsl.pltn13.pacbell.net] has quit [Read error: Connection reset by peer]
15:05	\|-\|	Netsplit over, joins: rharper, hollisb, hbaum, jerone, aliguori
15:05	\|-\|	Netsplit over, joins: sputhenp, jwb_, mastermind
15:10	\|-\|	VS_ChanLog [~stats@ns.theshore.net] has joined #xen
15:14	\|-\|	Netsplit synthon.oftc.net <-> saturn.oftc.net quits: mastermind, jwb_, jerone, hbaum, rharper, hollisb, aliguori, sputhenp
15:14	\|-\|	Netsplit over, joins: rharper, hollisb, hbaum, jerone, aliguori
15:15	\|-\|	Netsplit over, joins: sputhenp, jwb_, mastermind
15:26	\|-\|	VS_ChanLog [~stats@ns.theshore.net] has quit [Ping timeout: 480 seconds]
15:27	\|-\|	AlexanderMyodov [~Honeyman@89.109.0.50] has joined #xen
15:27	\|-\|	caker [~caker@caker.netrep.oftc.net] has quit [Ping timeout: 480 seconds]
15:33	\|-\|	AlexanderMyodov [~Honeyman@89.109.0.50] has quit [Quit: Óõîæó ÿ îò âàñ]
15:35	\|-\|	agriffis changed nick to agrif_away
15:38	\|-\|	andytxm [~andy@82-32-43-245.cable.ubr06.azte.blueyonder.co.uk] has joined #xen
15:40	<andytxm>	Hi. On a DomU it takes 20 seconds to get a response from 'route' - any ideas? date ; route ; date
15:46	<aw>	how about route -n to avoid the dns overhead?
15:46	\|-\|	mejlholm [~mejlholm@port79.ds1-abc.adsl.cybercity.dk] has quit [Quit: Leaving]
15:48	<andytxm>	aw: oh .. yes, that works, what is that all about? A DomU can't ping any other DomU.
15:49	<aw>	sounds like networking in your domU isn't setup very well
15:51	<andytxm>	hmmm is there anything Xen specific I could of messed in the DomU's? I'm not an expert, but neither am I new to Debian networking so I'm not sure what it would be. The DomU's have been up for 5+ hours and used to be accessible.
15:51	<andytxm>	The only other thing I can think of is some of the DomU's require me to run this 'route add -host X.X.X.1 dev eth0' from the console before they work which is a PITA
15:53	<aw>	I usually just let domUs dhcp. if they can do that, things are usually in good shape
15:53	<andytxm>	maybe the it is the way I have set up routing from the Dom0 as I can't use bridging.
15:54	<andytxm>	I'd prefer static as I just a single static for Dom0 and a small subnet for the DomU's - I'll read up on the routing on Dom0 - thanks!
16:00	\|-\|	VS_ChanLog [~stats@ns.theshore.net] has joined #xen
16:04	<andytxm>	hmmm not a lot to configure with routed...
16:08	<andytxm>	and maybe it is something to do with:- xen_net: Memory squeeze in netback driver.
16:19	<hensema>	ah
16:19	<hensema>	try setting maxmem to about 16 MB more than memory
16:20	<hensema>	memory = 512;
16:20	<hensema>	maxmem = 528;
16:24	\|-\|	tessier_ [~treed@gw.drjays.com] has quit [Ping timeout: 480 seconds]
16:28	\|-\|	Netsplit synthon.oftc.net <-> saturn.oftc.net quits: mastermind, jwb_, jerone, hbaum, rharper, hollisb, aliguori, sputhenp
16:28	\|-\|	tessier_ [~treed@gw.drjays.com] has joined #xen
16:29	\|-\|	Netsplit over, joins: rharper, hollisb, hbaum, jerone, aliguori
16:29	\|-\|	Netsplit over, joins: sputhenp, jwb_, mastermind
16:29	\|-\|	caker [~caker@caker.netrep.oftc.net] has joined #xen
16:34	<andytxm>	hensema: thanks, will give it a go
16:44	\|-\|	jwb_ changed nick to jwb_gone
16:53	\|-\|	KriP [kp@yancey.unixworld.dk] has quit [Quit: Leaving]
17:00	\|-\|	mdday changed nick to mday_away
17:05	\|-\|	DoberMann changed nick to DoberMann[ZZZzzz]
17:35	\|-\|	andytxm [~andy@82-32-43-245.cable.ubr06.azte.blueyonder.co.uk] has quit [Ping timeout: 480 seconds]
17:46	\|-\|	hollisb [~hollisb@bi01p1.co.us.ibm.com] has quit [Quit: leaving]
17:49	\|-\|	aliguori [~anthony@bi01p1.co.us.ibm.com] has quit [Quit: Leaving]
17:50	\|-\|	rharper [~rharper@bi01p1.co.us.ibm.com] has quit [Quit: Leaving]
17:59	\|-\|	rpg [~rpg@bi01p1.co.us.ibm.com] has quit [Ping timeout: 480 seconds]
17:59	\|-\|	jerone [~jerone@bi01p1.co.us.ibm.com] has quit [Quit: jerone]
18:10	\|-\|	hbaum [~hbaum@bi01p1.co.us.ibm.com] has quit [Quit: Client exiting]
18:11	\|-\|	aliguori [~anthony@cpe-70-112-17-156.austin.res.rr.com] has joined #xen
18:21	\|-\|	devastor [devastor@nepnet.fi] has quit [Ping timeout: 480 seconds]
18:36	\|-\|	cdub [~chrisw@wlanconf-nat-pool-bos.redhat.com] has quit [Read error: Connection reset by peer]
18:42	\|-\|	DrX0 [~DrX0@pool-71-105-109-220.lsanca.dsl-w.verizon.net] has quit [Read error: Connection reset by peer]
18:42	\|-\|	DrX0 [~DrX0@pool-71-105-109-220.lsanca.dsl-w.verizon.net] has joined #xen
18:59	\|-\|	sputhenp [~sputhenp@202.80.58.210] has quit [Quit: Leaving]
18:59	\|-\|	ahs3 changed nick to ahs3_away
18:59	\|-\|	devastor [devastor@nepnet.fi] has joined #xen
19:02	\|-\|	sputhenp [~sputhenp@202.80.58.210] has joined #xen
19:02	\|-\|	zez [~zez@217-133-59-42.b2b.tiscali.it] has quit [Remote host closed the connection]
19:56	\|-\|	agrif_away changed nick to agriffis
19:58	\|-\|	bestorga [~bestorga@71-222-114-20.ptld.qwest.net] has joined #xen
20:12	\|-\|	pvanhoof [~pvanhoof@d54C0EE14.access.telenet.be] has quit [Quit: Leaving]
20:41	\|-\|	sputhenp [~sputhenp@202.80.58.210] has quit [Quit: Leaving]
21:00	\|-\|	visik7 [~visi@host178-38-dynamic.3-87-r.retail.telecomitalia.it] has quit [Read error: Connection reset by peer]
21:01	\|-\|	visik7 [~visi@host178-38-dynamic.3-87-r.retail.telecomitalia.it] has joined #xen
21:19	<movement>	OK, looks to me that xenstored restartability is impossible without a new domU API. HURRAH.
21:33	<aliguori>	movement: :-/
21:33	<aliguori>	there's just break guest API compatibility, how bad could it really be :-)
21:40	<movement>	hmm, I think I came up with a plan. I've no idea if it's workable though.
21:41	<movement>	(only considering the xenstore page right now)
21:41	<movement>	only update req_cons if we've acted upon a whole message (readchn())
21:42	\|-\|	aliguori [~anthony@cpe-70-112-17-156.austin.res.rr.com] has quit [Quit: Leaving]
22:07	\|-\|	segher [~segher@dslb-084-056-129-197.pools.arcor-ip.net] has quit [Ping timeout: 480 seconds]
22:09	\|-\|	segher [~segher@dslb-084-056-173-145.pools.arcor-ip.net] has joined #xen
22:20	\|-\|	donour [~donour@71-37-150-57.albq.qwest.net] has joined #xen
22:20	\|-\|	donour [~donour@71-37-150-57.albq.qwest.net] has left #xen []
22:29	\|-\|	agriffis [~agriffis@156.153.254.66] has quit [Quit: leaving]
22:29	\|-\|	agriffis [~agriffis@156.153.254.66] has joined #xen
22:37	\|-\|	aliguori [~anthony@cpe-70-112-17-156.austin.res.rr.com] has joined #xen
22:59	\|-\|	VS_ChanLog [~stats@ns.theshore.net] has left #xen [Rotating Logs]
22:59	\|-\|	VS_ChanLog [~stats@ns.theshore.net] has joined #xen
23:31	<icblenke>	http://fosdem.org/2007/interview/andrew+morton <-- Andrew Morton says common paravirtualization interface: "expect support for Xen's "Domain U" function to be merged in 2.6.21 or 2.6.22"
23:33	\|-\|	bestorga [~bestorga@71-222-114-20.ptld.qwest.net] has quit [Quit: Leaving]
23:48	\|-\|	paelzer [~kvirc@blueice1n1.de.ibm.com] has joined #xen
23:53	\|-\|	kalden [~kalden@fac34-3-82-231-26-11.fbx.proxad.net] has quit [Quit: Parti]
---	Log	closed Wed Feb 07 00:00:45 2007