TheGrebs.com:IRC Logs:#xen:2006:07:03

---	Log	opened Mon Jul 03 00:00:46 2006
00:04	\|-\|	Netsplit over, joins: oelewapperke, weasel, deejoe, riel, demon, killefiz, dst, tessier, devastor, mastermind (+4 more)
00:04	\|-\|	Hunger [Hunger.hu@213.163.11.138] has joined #xen
00:04	\|-\|	Netsplit over, joins: womble, lcantey___, segher, damm_, movement
00:04	\|-\|	Netsplit over, joins: SNy, jforbes, zwane_work, bored2sleep
00:04	\|-\|	Netsplit over, joins: wenchien, Dougie
00:04	\|-\|	Netsplit over, joins: robin_
00:04	\|-\|	Netsplit over, joins: DV
00:04	\|-\|	Netsplit over, joins: DV_, ag-, pea
00:04	\|-\|	Netsplit over, joins: schultmc, xai, tonyb, nick125, anticw, am88b, zwane
00:04	\|-\|	Netsplit over, joins: ns
00:04	\|-\|	Netsplit over, joins: msinhore, gerrit_, murb, dezent_, Sir_Ahzz, knewt, tessier__, dykman, koki
00:04	\|-\|	Netsplit over, joins: lucca, lilo2, unriel, um, cdub, rpg_, hensema, klapmuetz, hbaum
00:04	\|-\|	Netsplit over, joins: netman1, buggs, mooli
00:04	\|-\|	Netsplit over, joins: dhendrix, katzj, grifferz
00:04	\|-\|	NetHack #xen [+oo anticw Sir_Ahzz] by [charon.oftc.net]
00:07	\|-\|	VS_ChanLog [~stats@ns.theshore.net] has joined #xen
00:29	\|-\|	psuresh [~psuresh@125.16.129.16] has joined #xen
00:31	\|-\|	zwane_work [~zwane@S0106001217d33fa6.vc.shawcable.net] has quit [Quit: Leaving]
00:35	\|-\|	zwane_work [~zwane@S0106001217d33fa6.vc.shawcable.net] has joined #xen
00:44	\|-\|	horms [~horms@vagw.valinux.co.jp] has joined #xen
00:53	\|-\|	mastermind [~mastermin@mastermind.kaltenbrunner.cc] has quit [Ping timeout: 480 seconds]
01:02	\|-\|	muli [~muli@87.69.40.180.cable.012.net.il] has quit [Remote host closed the connection]
01:11	\|-\|	psuresh [~psuresh@125.16.129.16] has quit [Quit: Leaving]
01:29	\|-\|	Q3aiml_ [~Q3aiml@c-67-190-251-183.hsd1.mn.comcast.net] has joined #xen
01:33	\|-\|	Q3aiml [~Q3aiml@c-67-190-251-183.hsd1.mn.comcast.net] has quit [Ping timeout: 480 seconds]
02:33	\|-\|	lcantey___ [~lcantey@ppp-67-126-40-8.dialup.pltn13.pacbell.net] has quit [Quit: lcantey___]
02:34	\|-\|	lcantey___ [~lcantey@ppp-67-126-40-8.dialup.pltn13.pacbell.net] has joined #xen
02:39	\|-\|	Shoragan_ [~shoragan@d072.apm.etc.tu-bs.de] has joined #xen
02:42	\|-\|	lcantey___ [~lcantey@ppp-67-126-40-8.dialup.pltn13.pacbell.net] has quit [Quit: lcantey___]
02:48	\|-\|	Shoragan_ [~shoragan@d072.apm.etc.tu-bs.de] has quit [Ping timeout: 480 seconds]
02:51	\|-\|	Shoragan_ [~shoragan@d072.apm.etc.tu-bs.de] has joined #xen
03:02	\|-\|	Shoragan_ [~shoragan@d072.apm.etc.tu-bs.de] has quit [Quit: Leaving]
03:03	\|-\|	Shoragan_ [~shoragan@d072.apm.etc.tu-bs.de] has joined #xen
03:34	\|-\|	rusty [~rusty@CPE-58-169-49-163.nsw.bigpond.net.au] has joined #xen
03:41	\|-\|	rusty [~rusty@CPE-58-169-49-163.nsw.bigpond.net.au] has quit [Quit: Ex-Chat]
03:51	\|-\|	visik7 [~visi@155.185.28.178] has joined #xen
03:58	\|-\|	athomas [~athomas@host86-136-84-147.range86-136.btcentralplus.com] has joined #xen
03:58	\|-\|	womble [~mpalmer@106.135.233.220.exetel.com.au] has quit [Quit: What's behind the round window...]
04:08	\|-\|	mooli [~muli@nesher3.haifa.il.ibm.com] has quit [Ping timeout: 480 seconds]
04:20	\|-\|	parabelboi [~parabelbo@blueice2n1.de.ibm.com] has joined #xen
04:34	\|-\|	zwane_work [~zwane@S0106001217d33fa6.vc.shawcable.net] has quit [Ping timeout: 480 seconds]
04:52	\|-\|	pvanhoof [~pvanhoof@mailhost.newtec.be] has joined #xen
04:58	\|-\|	visik7 [~visi@155.185.28.178] has quit [Remote host closed the connection]
05:26	\|-\|	horms [~horms@vagw.valinux.co.jp] has quit [Quit: Leaving]
05:35	\|-\|	sdague [~sdague@user-12lc6iq.cable.mindspring.com] has joined #xen
06:08	\|-\|	muli [~muli@nesher3.haifa.il.ibm.com] has joined #xen
06:15	\|-\|	dst [~dst@synyx.org] has quit [Ping timeout: 480 seconds]
06:18	\|-\|	SLot [~SLot___@gustavo.gruposim.com.br] has joined #xen
06:29	\|-\|	horms [~horms@p7209-ipbf22marunouchi.tokyo.ocn.ne.jp] has joined #xen
06:29	\|-\|	Shoragan_ [~shoragan@d072.apm.etc.tu-bs.de] has quit [Read error: Connection reset by peer]
06:42	\|-\|	visik7 [~visi@155.185.28.178] has joined #xen
06:44	\|-\|	rusty [~rusty@ppp60-216.lns1.cbr1.internode.on.net] has joined #xen
06:45	\|-\|	Shoragan_ [~shoragan@d072.apm.etc.tu-bs.de] has joined #xen
06:46	\|-\|	visik7 [~visi@155.185.28.178] has quit [Remote host closed the connection]
06:57	\|-\|	rusty [~rusty@ppp60-216.lns1.cbr1.internode.on.net] has quit [Quit: Ex-Chat]
07:19	\|-\|	THunder [~THunder@mvil2.bb.netvision.net.il] has joined #xen
07:19	\|-\|	THunder [~THunder@mvil2.bb.netvision.net.il] has left #xen []
07:21	<oelewapperke>	that's more a distro problem
07:21	\|-\|	visik7 [~visi@155.185.28.178] has joined #xen
07:41	\|-\|	bernarde [~bernarde@201.44.214.48] has joined #xen
07:48	\|-\|	visik7 [~visi@155.185.28.178] has quit [Remote host closed the connection]
07:50	\|-\|	dezent_ [dezent@unixgeek.biz] has quit [Quit: leaving]
08:41	\|-\|	horms [~horms@p7209-ipbf22marunouchi.tokyo.ocn.ne.jp] has quit [Remote host closed the connection]
08:42	\|-\|	Shoragan_ [~shoragan@d072.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
08:55	\|-\|	lcantey___ [~lcantey@ppp-67-126-40-237.dialup.pltn13.pacbell.net] has joined #xen
08:55	\|-\|	visik7 [~visi@155.185.28.43] has joined #xen
09:03	\|-\|	lcantey___ [~lcantey@ppp-67-126-40-237.dialup.pltn13.pacbell.net] has quit [Quit: lcantey___]
09:12	\|-\|	elisboa [~elisboa@201.44.96.130] has joined #xen
09:42	\|-\|	elisboa [~elisboa@201.44.96.130] has quit [Remote host closed the connection]
10:16	\|-\|	lcantey___ [~lcantey@64-60-250-34.cust.telepacific.net] has joined #xen
10:21	\|-\|	pvanhoof [~pvanhoof@mailhost.newtec.be] has quit [Remote host closed the connection]
10:43	\|-\|	lcantey___ [~lcantey@64-60-250-34.cust.telepacific.net] has quit [Quit: lcantey___]
10:43	\|-\|	zwane_work [~zwane@S0106001217d33fa6.vc.shawcable.net] has joined #xen
10:47	\|-\|	Marco [~marco@249-174.surfsnel.dsl.internl.net] has joined #xen
10:48	<Marco>	hi
10:52	\|-\|	visik7 [~visi@155.185.28.43] has quit [Quit: ]
11:04	\|-\|	Shoragan_ [~shoragan@d072.apm.etc.tu-bs.de] has joined #xen
11:07	\|-\|	elisboa [~elisboa@201.44.96.130] has joined #xen
11:10	\|-\|	mastermind [~mastermin@mastermind.kaltenbrunner.cc] has joined #xen
11:14	\|-\|	Shoragan_ [~shoragan@d072.apm.etc.tu-bs.de] has quit [Ping timeout: 480 seconds]
11:21	\|-\|	Shoragan_ [~shoragan@d072.apm.etc.tu-bs.de] has joined #xen
11:26	\|-\|	Shoragan_ [~shoragan@d072.apm.etc.tu-bs.de] has quit [Read error: Connection reset by peer]
11:27	\|-\|	Shoragan_ [~shoragan@d072.apm.etc.tu-bs.de] has joined #xen
11:37	\|-\|	athomas [~athomas@host86-136-84-147.range86-136.btcentralplus.com] has quit [Quit: Leaving]
11:48	\|-\|	_bernarde [~bernarde@201.44.214.48] has joined #xen
11:52	\|-\|	Basic_py [~Basic@warden.real-time.com] has joined #xen
11:53	\|-\|	bernarde [~bernarde@201.44.214.48] has quit [Ping timeout: 480 seconds]
11:55	\|-\|	Shoragan_ [~shoragan@d072.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
11:56	\|-\|	Shoragan_ [~shoragan@d072.apm.etc.tu-bs.de] has joined #xen
11:59	\|-\|	FloTiX [~flotix@AAmiens-157-1-20-96.w86-196.abo.wanadoo.fr] has joined #xen
12:02	\|-\|	visik7 [~visi@host39-8.pool8260.interbusiness.it] has joined #xen
12:12	\|-\|	parabelboi [~parabelbo@blueice2n1.de.ibm.com] has quit [Read error: Connection reset by peer]
12:14	\|-\|	zwane_work [~zwane@S0106001217d33fa6.vc.shawcable.net] has quit [Quit: Leaving]
12:16	\|-\|	visik7 [~visi@host39-8.pool8260.interbusiness.it] has quit [Quit: ]
12:20	\|-\|	zwane_work [~zwane@S0106001217d33fa6.vc.shawcable.net] has joined #xen
12:21	\|-\|	mook [~muli@87.69.40.180.cable.012.net.il] has joined #xen
12:32	\|-\|	visik7 [~visi@host39-8.pool8260.interbusiness.it] has joined #xen
12:47	\|-\|	Q3aiml_ [~Q3aiml@c-67-190-251-183.hsd1.mn.comcast.net] has quit [Ping timeout: 480 seconds]
12:47	\|-\|	Q3aiml [~Q3aiml@c-67-190-251-183.hsd1.mn.comcast.net] has joined #xen
12:55	\|-\|	FloTiX [~flotix@AAmiens-157-1-20-96.w86-196.abo.wanadoo.fr] has quit [Quit: m0w]
13:15	<movement>	the xen api pdf is 403 forbidden, darn
13:16	<movement>	ah, only via wget
13:28	\|-\|	Q3aiml [~Q3aiml@c-67-190-251-183.hsd1.mn.comcast.net] has quit [Ping timeout: 480 seconds]
13:29	\|-\|	Q3aiml [~Q3aiml@c-67-190-251-183.hsd1.mn.comcast.net] has joined #xen
14:27	\|-\|	tobi [~tobi@zux006-049-116.adsl.green.ch] has joined #xen
14:33	\|-\|	msinhore [~msinhore@mail.samurai.com.br] has quit [Ping timeout: 480 seconds]
14:43	\|-\|	DV [~veillard@veillard.com] has quit [Ping timeout: 480 seconds]
14:43	\|-\|	DV_ [~veillard@veillard.com] has quit [Ping timeout: 480 seconds]
14:45	\|-\|	msinhore [~msinhore@mail.samurai.com.br] has joined #xen
14:47	\|-\|	pvanhoof [~pvanhoof@d54C0E27E.access.telenet.be] has joined #xen
15:25	\|-\|	_bernarde [~bernarde@201.44.214.48] has quit [Remote host closed the connection]
15:39	\|-\|	ns [~nivedita@c-67-171-181-157.hsd1.or.comcast.net] has quit [Ping timeout: 480 seconds]
15:49	\|-\|	ns [~nivedita@c-67-171-181-157.hsd1.or.comcast.net] has joined #xen
16:06	\|-\|	Shoragan_ [~shoragan@d072.apm.etc.tu-bs.de] has quit [Quit: Leaving]
16:10	\|-\|	DV [~veillard@veillard.com] has joined #xen
16:10	\|-\|	DV_ [~veillard@veillard.com] has joined #xen
16:14	\|-\|	parabelboi [~parabelbo@port-87-234-138-223.dynamic.qsc.de] has joined #xen
16:17	\|-\|	msinhore [~msinhore@mail.samurai.com.br] has quit [Remote host closed the connection]
16:24	\|-\|	msinhore [~msinhore@mail.samurai.com.br] has joined #xen
16:26	\|-\|	visik7 [~visi@host39-8.pool8260.interbusiness.it] has quit [Remote host closed the connection]
16:27	\|-\|	tobi [~tobi@zux006-049-116.adsl.green.ch] has quit [Quit: tobi]
16:33	\|-\|	visik7 [~visi@host39-8.pool8260.interbusiness.it] has joined #xen
16:34	\|-\|	msinhore [~msinhore@mail.samurai.com.br] has quit [Remote host closed the connection]
16:36	\|-\|	ivan [~ikelly@213.228.247.71] has joined #xen
16:43	\|-\|	msinhore [~msinhore@correio.samurai.com.br] has joined #xen
17:08	\|-\|	netman1 [~netman1@cc275267-a.deven1.ov.home.nl] has quit [Ping timeout: 480 seconds]
17:57	<riel>	Due to lack of Datacenter space,
17:57	<riel>	unixshell# has suspended ordering until further notice
17:57	<riel>	*We apologize for any inconvenience*
17:57	<riel>	If you are still interested in a low cost, unmanaged plan, please check out Tektonic
17:57	<riel>	nice illustration of why some people really want virtualization :)
17:58	<nick125>	lol
17:59	<knewt>	hopefully i'm going to manage to put some time aside to start playing with xen again soon. what would be a good bit of code to look at to figure out how to put drivers together now? that is, the xen side of drivers, not driver-specific stuff
18:00	<ivan>	knewt: you mean the backend driver?
18:00	<knewt>	both sides
18:00	<knewt>	all the xenbus stuff and whatnot
18:01	<ivan>	hmmm, console seems to be the simplest on xen 2
18:01	<ivan>	dunno about with 3/xenbus though
18:07	<tessier__>	riel: That message is misleading.
18:07	<tessier__>	If you email them and ask for an account you can get one
18:07	<tessier__>	I got a xen vps a couple weeks ago from them. That message has been there for ages now.
18:07	<nick125>	they just don't have enough server capacity to accept more automated orders :P
18:07	<tessier__>	I am not sure why they are doing that.
18:07	<tessier__>	right
18:07	<tessier__>	But they are actively trying to move people off of xen to virtuozzo
18:08	<nick125>	I don't know why they want to do that
18:08	<tessier__>	They seem to have some xen bug they cannot resolve
18:08	<tessier__>	involving spontaneous reboots
18:08	<nick125>	I've never had xen reboot without me telling it
18:08	<ivan>	tessier__: i haven't had one of them in quite a while
18:08	<nick125>	and I haven't heard about anyone else experiencing that issue
18:09	<ivan>	it was over a month ago
18:12	<nick125>	I have a feeling that the whole "out of datacenter space" thing is just bs, and they are doing it so they don't put more users on xen-based systems
18:14	<ivan>	i've seem matta on here a few times, he could probably better inform people
18:16	<nick125>	because, aren't tektonic and unixshell hosted in the same DC?
18:34	\|-\|	Q3aiml_ [~Q3aiml@c-67-190-251-183.hsd1.mn.comcast.net] has joined #xen
18:35	\|-\|	Q3aiml [~Q3aiml@c-67-190-251-183.hsd1.mn.comcast.net] has quit [Ping timeout: 480 seconds]
18:51	\|-\|	womble [~mpalmer@106.135.233.220.exetel.com.au] has joined #xen
18:58	\|-\|	visik7 [~visi@host39-8.pool8260.interbusiness.it] has quit [Remote host closed the connection]
19:07	<riel>	tessier__: I suspect that with virtuozo they can fit way more virtual machines per physical system
19:08	<tessier__>	riel: Why is that? I'm not familiar with virtuozzo
19:09	<riel>	memory is not partitioned hard
19:09	<riel>	but each guest has some minimum and maximum, with global page replacement across it all
19:10	<tessier__>	Why can't xen do that?
19:10	<knewt>	plus there's only one actual kernel, right?
19:10	<ivan>	just looked at the forums. matta said he's not going to start offering xen until 3.0.3 has been tested for a few weeks
19:10	<riel>	knewt: exactly
19:11	<riel>	with virtuozo you can also share page cache - ie. cache glibc, etc just once in memory
19:11	<ivan>	tessier__: can't xen do similar with the balloon driver?
19:12	<tessier__>	ivan: Not really. You are still limited by the amount of physical memory in the machine. Sounds like virtuozzo shares a lot more.
19:13	<tessier__>	Seems like that would be more of a security problem than you could have with xen.
19:14	<ivan>	sure
19:14	<knewt>	virtuozzo is the same sort of virtualization level as vserver
19:23	\|-\|	msinhore [~msinhore@correio.samurai.com.br] has quit [Quit: Fui embora]
19:37	<tessier__>	Is it more like user mode linux?
19:38	<ivan>	tessier__: no, it's partitioning
19:38	<ivan>	like solaris zones
19:44	\|-\|	horms [~horms@vagw.valinux.co.jp] has joined #xen
19:46	<nick125>	eek, virtuozzo spits
19:49	<tessier__>	nick125: What's wrong with it?
19:55	<nick125_lappy>	tessier__: one main issue I don't like about it is the evil ethics it enables
19:56	<nick125_lappy>	it allows for easy overselling on RAM
20:04	<nick125_lappy>	plus, it's not nearly as efficent as xen (from what I've heard), and, it will only run whatever the host os is running (if the host is running linux, the VMs will run linux)
20:13	\|-\|	pvanhoof [~pvanhoof@d54C0E27E.access.telenet.be] has quit [Remote host closed the connection]
20:40	<tessier__>	Nothing wrong with overselling RAM. I wish I could oversell RAM on my personal box.
20:41	<tessier__>	I have some domains which I need to be there but are rarely active.
20:56	\|-\|	parabelboi [~parabelbo@port-87-234-138-223.dynamic.qsc.de] has quit [Ping timeout: 480 seconds]
21:04	\|-\|	parabelboi [~parabelbo@port-87-234-138-223.dynamic.qsc.de] has joined #xen
21:05	\|-\|	wenchien [~wenchien@221-169-69-23.adsl.static.seed.net.tw] has quit [Ping timeout: 480 seconds]
21:17	\|-\|	parabelboi [~parabelbo@port-87-234-138-223.dynamic.qsc.de] has quit [Ping timeout: 480 seconds]
21:19	<riel>	tessier__: read my OLS paper :)
21:19	<riel>	tessier__: I came up with a way to measure how much RAM a VM needs
21:34	<tessier>	riel: ooh...cool. URL?
21:35	<riel>	none yet, the OLS papers are not publically available yet
21:35	<aliguori>	riel, can you give a quick summary of your basic technique?
21:37	<riel>	sure
21:37	<riel>	I measure refaults, meaning when the system faults on recently evicted memory
21:37	<riel>	I measure those against the "distance"
21:38	<riel>	say we fault on a page that is 100 MB behind the LRU exit point, meaning 100MB of pages got evicted after this one page got evicted
21:38	<riel>	if we had 100MB more memory, that fault would have been avoided
21:38	<riel>	so we create a (distance, number of faults) histogram
21:39	<riel>	and we can see that eg. if we had 100MB more memory in this VM, we would reduce the number of faults by 20%
21:39	<riel>	if we had 500MB more memory, we would reduce the number of faults by 80%
21:39	<riel>	if we had 1000MB more memory, the reduction would be 85%
21:39	<riel>	etc...
21:52	\|-\|	ivan [~ikelly@213.228.247.71] has quit [Quit: leaving]
21:59	<aliguori>	riel, very interesting. have you found any interesting results (for instance, what the sweet spot tends to be)?
22:00	<riel>	you measure that
22:00	<riel>	come to my OLS talk :)
22:01	<aliguori>	:-)
22:01	<aliguori>	i will
22:02	<tessier>	http://pastebin.ca/78460
22:02	<riel>	I'm pretty sure the sweet spot varies by workload
22:02	<tessier>	uh oh
22:02	<riel>	and by time of day
22:02	<tessier>	Just got that in two of my domainU's
22:02	<riel>	tessier: dom0 has a higher scheduler priority than domU ?
22:03	<tessier>	And I still get a lot of "Time went backwards" messages in my dom0 event hough a patch a while back was supposed to fix that
22:03	<tessier>	riel: No
22:03	<riel>	mmmm ok
22:03	<tessier>	I can't play mp3's on this box anymore. xmms just stops. I bet it is related to the timer thing.
22:04	<riel>	no doubt
22:04	<riel>	my dom0 works fine as a desktop still
22:04	<riel>	using SEDF, since my system crashed with the credit scheduler
22:04	<tessier>	Mine does also for the most part.
22:04	<riel>	now if only IO to HVM domains was more efficient
22:05	<tessier>	Say, would any of you know of a way to find out how much time a system spends waiting for seeks on the HD?
22:05	<riel>	look at the iowait time
22:06	<tessier>	Yeah but that is not a very good quantifier
22:06	<riel>	what else do you want to know?
22:06	<tessier>	For one thing it does not tell me which disk is going the seeking
22:06	<tessier>	And it does not tell me what process is causing it
22:06	<riel>	iostat -x then?
22:06	<riel>	well no, you cannot tell that
22:06	<riel>	it is not possible to tell which process causes disk writes, most of the time
22:07	<riel>	unless you're fine blaming everything on pdflush
22:08	<riel>	ok, I need to finish this PV kernel
22:08	<aliguori>	riel, fwiw, i don't trust the async dma patch that's in the xen tree very much. the ones that have floated around qemu-devel have been much more sophisticated and fabrice has been reluctant to apply any of them
22:08	<riel>	HVM is just too slow for all kinds of reasons
22:08	<riel>	aliguori: heh, no surprise
22:08	<riel>	well, if there is any kind of data corruption, I _will_ find it
22:09	<aliguori>	riel, you may want to try backing it out and seeing if that enables dma again...
22:09	[~]	riel does 24 100k row postgresql inserts per day
22:09	<aliguori>	nice :-)
22:09	<riel>	and one 2+ million row query
22:09	<riel>	that moves data from the hourly summaries into a daily summary
22:09	<riel>	in one huge query
22:09	<riel>	if there is data corruption, this workload will catch it, guaranteed
22:10	<riel>	btw, there's no DMA being done right now from my VT guest :(
22:10	<aliguori>	riel, well, there is potential for data corruption in the current code. raw_open doesn't use O_DIRECT so writes are going through the page cache
22:11	<riel>	FC5 is still on changeset 10278 though
22:11	<riel>	oh, my VM is LVM backed, not file backed
22:11	<riel>	not sure how much of a difference that'll make though
22:11	<aliguori>	riel, i dunno.. does a device opened through the file system not use the page cache?
22:13	<aliguori>	it's kind of scary too b/c the raw_{read,write} doesn't check for complete read writes.. it just fails
22:13	<riel>	ugh
22:13	<aliguori>	i reckon vt is too slow to ever exhaust the io buffers :-) but it was faster, it could be a problem
22:14	<riel>	well, with IDE you can only ever have one I/O in flight at a time
22:14	<aliguori>	riel, the best part is that the ide code just ignores the value from a read or write... so even with an error things will go bad
22:14	<riel>	I imagine the SCSI emulation will be limited to 32 or so
22:15	<aliguori>	riel, not necessarily. again, open() w/o O_DIRECT so write's will return immediately
22:15	<aliguori>	so you could have multiple outstanding writes (even though userspace doesn't realize that)
22:15	<riel>	uh oh
22:15	<riel>	the thing is too stupid to wait on write completion ?!
22:15	<aliguori>	riel, well, the normal write() call doesn't necessarily return on write completion...
22:16	<aliguori>	riel, i think just adding O_DIRECT or O_SYNC would help a lot
22:16	<aliguori>	also updating the read/write calls to loop
22:16	<aliguori>	doesn't even check for EINTR right now...
22:17	<riel>	crap
22:17	<riel>	why did we ever consider Xen to be viable? :)
22:18	<aliguori>	tools/ioemu/block.c:raw_{read,write,open}, tools/ioemu/hw/ide.c:ide_sector_write()
22:18	<aliguori>	because it's the only enterprise ready open source virtualization solution :-)
22:19	<aliguori>	at least, that's what all the press releases are saying :-)
22:19	<riel>	s/enterprise ready/enterprise quality/
22:19	<aliguori>	hehe
22:19	<riel>	intuhprize kwalitie
22:20	<aliguori>	if it makes you feel any better, vt support is much better than it was a few months ago.. it was unbelievably slow. slower than qemu doing full emulation
22:20	<aliguori>	apparently, lists.xensource.com doesn't like my email address b/c my note on christian's latest device model patch hasn't gone through. his patch introduces quite a few nasty bugs
22:21	<aliguori>	:-( i guess i'll resend from a different account
22:22	<katzj>	aliguori: actually, all of your email addresses have been blacklisted ;-)
22:23	<riel>	conspiracy, I tell you
22:23	<riel>	(I didn't tell you that!)
22:26	<aliguori>	hehe
22:26	<aliguori>	katzj, i guess they finally had enough :-)
22:28	<aliguori>	my block is on one blacklist and some stupid spam filters that use blacklists as a primary filtering mechanism end up causing problems...
22:31	<riel>	you're sending out email directly from a dynamic IP address, and not via a mail server?
22:31	<riel>	OTOH - RR :)
22:31	<aliguori>	:-)
22:31	<aliguori>	yeah, I know
22:31	<aliguori>	it's evil
22:32	<riel>	and their residential service too, not a business line
22:32	<aliguori>	yup :-)
22:32	<riel>	the business lines with static IP are rumored to be ok blocklist-wise
22:32	<aliguori>	wasn't a problem until recently
22:33	<riel>	query zone: '243.9.116.70.dnsbl.sorbs.net'
22:33	<riel>	243.9.116.70.dnsbl.sorbs.net. 3600 IN A 127.0.0.10
22:33	<riel>	243.9.116.70.dnsbl.sorbs.net. 3600 IN TXT "Dynamic IP Addresses See: http://www.sorbs.net/lookup.shtml?70.116.9.243"
22:35	\|-\|	segher [~segher@dslb-084-056-141-254.pools.arcor-ip.net] has quit [Ping timeout: 480 seconds]
22:35	<aliguori>	i've been hoping i could setup spf and that would appease the black lists
22:35	<riel>	LOL
22:35	<riel>	it's a good thing you're not an email developer :)
22:35	<aliguori>	uh oh :-)
22:36	<aliguori>	i reckon that's a silly idea then :-)
22:36	<riel>	about as bad as they get, sorry :)
22:36	<aliguori>	hehe
22:36	<riel>	1) your email does not travel via the blocklist
22:36	<riel>	2) your email travels directly to the destination server
22:36	<riel>	3) the destination server's policy would have to be changed, not the blocklist
22:36	<riel>	4) SPF is generally a bad idea
22:36	\|-\|	segher [~segher@dslb-084-056-143-048.pools.arcor-ip.net] has joined #xen
22:37	<aliguori>	what i meant, is that spf would appease the spam filter (even though it was on the black list)
22:37	<aliguori>	i didn't know that spf was frowned upon
22:37	<riel>	SPF applies to more spam than legitimate email
22:37	<aliguori>	really? i guess that makes sense
22:37	<riel>	SPF is incompatible with email forwarders like alumni.mit.edu
22:37	<aliguori>	yeah
22:38	<riel>	also, which IP addresses would you designate as being allowed to send email for your domain, considering that you are on DYNAMIC IP SPACE ?!
22:38	<aliguori>	:-)
22:38	<riel>	you don't have a static IP address, how can you do SPF ?
22:38	<aliguori>	while it's dynamic, it's rather fixed.. it only changes every few months
22:38	<aliguori>	so updating the spf record wouldn't be too painful
22:38	<aliguori>	yeah, i know, i should get a static ip :-)
22:39	<riel>	that would also solve the "I am on a dynamic address list" problem :)
22:39	<aliguori>	yeah, i know
22:39	<tessier>	I have a static IP but a lot of mail servers still have it on a dynamic list
22:39	<tessier>	Just because it is DSL
22:39	<aliguori>	:-) it's just laziness on my part
22:40	<tessier>	So there is at least one machine out there blocking mail from me every day according to my mail logs
22:40	<riel>	tessier: what blocklist is that?
22:40	<aliguori>	i've been meaning to get a xen vps somewhere anyway
22:40	\|-\|	SLot [~SLot___@gustavo.gruposim.com.br] has quit [Read error: Connection reset by peer]
22:41	<tessier>	riel: I don't know what blocklist the other mailserver is using
22:41	<riel>	tessier: it's not showing up on http://openrbl.org/client/#75.5.99.178
22:41	\|-\|	SLot [~SLot___@gustavo.gruposim.com.br] has joined #xen
22:41	<riel>	maybe they're using an OLD copy of the bogons list?
22:41	<riel>	not a DNSBL at all
22:41	<tessier>	75.5.99.177 is actually my mail server
22:41	[~]	tessier looks up the error
22:41	<tessier>	Jul 3 19:46:44 mail postfix/smtp[27117]: 4F3544A2E2: to=<cgs@ldsys.net>, relay=ext1-chi.ldsys.net[208.176.63.109], delay=340219, status=deferred (host ext1-chi.ldsys.net[208.176.63.109] refused to talk to me: 554 <adsl-75-5-99-177.dsl.sndg02.sbcglobal.net[75.5.99.177]>: Client host rejected: Use a mail exchanger or a public web mail account (Yahoo, Hotmail, Excite, etc.). DSL and dialup are not allowed.)
22:42	<tessier>	I guess if they don't want to talk to DSL it's their perogative.
22:42	<riel>	ahhhhh, you're running into a regexp on their side I guess
22:42	<tessier>	They prefer to talk to webmail? Sheesh....
22:42	<riel>	I guess that's not somebody you care about
22:43	<tessier>	No, it isn't. They can jump in a lake.
22:43	<riel>	they sure can, they're in IL
22:43	<riel>	big lake
22:43	<tessier>	I wonder if I can get SBC to change my reverse. SBC wasn't my idea. The guy I rent this place from chose them.
22:46	\|-\|	Q3aiml_ [~Q3aiml@c-67-190-251-183.hsd1.mn.comcast.net] has quit [Ping timeout: 480 seconds]
22:59	\|-\|	VS_ChanLog [~stats@ns.theshore.net] has left #xen [Rotating Logs]
22:59	\|-\|	VS_ChanLog [~stats@ns.theshore.net] has joined #xen
---	Log	closed Tue Jul 04 00:00:59 2006