Back to Home / #linode / 2006 / 10 / Prev Day | Next Day
#linode IRC Logs for 2006-10-23

---Logopened Mon Oct 23 00:00:56 2006
00:07|-|kokoko [] has joined #linode
00:13|-|konoko [] has quit [Ping timeout: 480 seconds]
00:17|-|Newsome [] has quit [Quit: Linux: Now with employee pricing!]
00:17|-|tasaro [] has quit [Quit: Snak 5.1.5 IRC For Macintosh -]
00:30|-|internat [] has quit [Quit: This computer has gone to sleep]
00:31|-|kokoko [] has quit [Ping timeout: 480 seconds]
00:34<taupehat>new rtorrent/libtorrent out for debian testing
00:35<taupehat>hopefully this one won't crash whenever a torrent finishes downloading
00:36<erikh>cool, thanks for the tip.
00:36<taupehat>you're welcome
00:36[~]erikh officially hasn't slept for 24 hours now
00:36<taupehat>it's on heldback, but I'm sure you know how to work around that
00:36<erikh>ah, it's kept back
00:36<erikh>yeah, I think I'll just wait
00:36<taupehat>apt-get install rtorrent libtorrent9
00:36<erikh>don't use torrent often enough
00:36<taupehat>nothing bad will happen, i promise
00:37<erikh>eh, generally there's a reason things are held back in testing.
00:37<erikh>I'm not saying they're major, but until I have a need for it..
00:37<taupehat>I'll deal with the problems, if there are any
00:38<taupehat>it's hard to make worse what it does now
00:38<taupehat>plus, generally stuff gets kept back for non-x86 arch
00:38<erikh>I really need to slim down my mail filter chain.
00:38<erikh>I think I'm going to gut spamd and use postgrey
00:39<erikh>and just do my RBL's from postfix.
00:39<taupehat>"zomg, there's a transient segfault on the mipsel version!!!"
00:39<taupehat>erikh: that's what I use
00:39<taupehat>postgrey + RBL
00:39<taupehat>I hardly get anything
00:39<erikh>you use amavis/clam at all?
00:39<taupehat>but then again
00:39<erikh>I need to, but I understand why others wouldn't
00:41<erikh>hmm, although, that may coerce a certain freeloader on my box to either start helping me with server costs or finding a new disk to fill up..
00:41<erikh>I make minimum wage and can afford my cost, he can certainly afford to pay half of it.
00:42<erikh>oh, btw... wahoo. I got a raise
00:42<erikh>thanks to the state of oregon.
00:42<taupehat>erikh: nevermind
00:42<erikh>taupehat: eh?
00:43<erikh>I'm pretty grumpy right now, not intending to be a jerk. Anyways, it's time for work! cya.
01:10|-|SpaceHobo [] has joined #linode
01:12|-|konoko [~Vampire@] has joined #linode
01:36|-|konoko [~Vampire@] has quit [Ping timeout: 480 seconds]
02:21|-|andrew_j_w [] has joined #linode
02:29|-|donnchao1aoimh [] has joined #linode
02:30|-|donnchaocaoimh [~donncha@] has quit [Ping timeout: 481 seconds]
02:39|-|konoko [~Vampire@] has joined #linode
02:42|-|SpaceHobo [] has quit [Ping timeout: 480 seconds]
02:49|-|internat [] has joined #linode
03:22|-|SpaceHobo [~spacehobo@] has joined #linode
05:30|-|jekil [] has joined #linode
06:41|-|linville [] has joined #linode
06:52|-|konoko [~Vampire@] has quit [Ping timeout: 480 seconds]
07:17|-|konoko [~Vampire@] has joined #linode
07:46|-|andrew_j_w [] has quit [Ping timeout: 480 seconds]
08:21|-|ldexter_ [] has quit [Quit: BitchX: it isn't only my script, its also my client]
08:29|-|konoko [~Vampire@] has quit [Ping timeout: 480 seconds]
08:44|-|konoko [] has joined #linode
09:12<tronix>another DoS again?
09:12<tronix>75% packet loss on two different hosts at TP
09:12<tronix>heh well, wait for newcomers soon. ;)
09:13<tronix>(to the channel)
09:14<tronix>hmm hovering around 85% now.
09:15|-|Beirdo [] has quit [Remote host closed the connection]
09:15|-|Beirdo [] has joined #linode
09:15<tronix>ahh just cleared up. weird.
09:16[~]tronix cancels ticket submission and goes back to slumber
09:16|-|konoko [] has quit [Quit: taisetsu to yuukoto wa subarashii tsukashi subarashii to yuukoto wa motto taisetsu desu]
09:17|-|konoko [] has joined #linode
09:17|-|Ciaran [~ciaran@] has joined #linode
09:17<Ciaran>Hi there.
09:18<Ciaran>Just wondering - are caker or mikegrb around? Just wanted to PM them for a bit...
09:18<bendy24>not yet
09:45<JasonF>k guys... I just cancelled my linode account
09:45<JasonF>it was cool knowing all you peoples, and I'll be around on #linuxchics here or on a multitude of channels on freenode
09:46<JasonF>nice knowing everyone :)
09:46|-|JasonF [] has left #linode []
09:46[~]bendy24 waves
10:10|-|FireSlash [~FireSlash@] has joined #linode
10:18|-|spr [] has joined #linode
10:23|-|cmantito [] has joined #linode
10:41|-|spr [] has quit [Ping timeout: 480 seconds]
10:43|-|Dreamr3 [] has joined #linode
10:47|-|Dreamr_3 [] has quit [Ping timeout: 480 seconds]
10:56|-|FireSlash_ [~FireSlash@] has joined #linode
10:58|-|flint [] has quit [Ping timeout: 480 seconds]
11:01|-|FireSlash [~FireSlash@] has quit [Ping timeout: 480 seconds]
12:00|-|Newsome [] has joined #linode
12:03|-|afv-13 [] has joined #linode
12:27|-|fo0bar [] has joined #linode
12:44|-|flint [] has joined #linode
12:56<afv-13>was everyones network down for 2 hours earlier?
12:59|-|SpaceHobo [~spacehobo@] has quit [Ping timeout: 480 seconds]
13:51|-|cmantito [] has quit [Ping timeout: 480 seconds]
14:14|-|flint [] has quit [Quit: Leaving]
14:37|-|spr [] has joined #linode
14:58|-|tasaro [~tasaro@] has joined #linode
15:08|-|LinodeJavaUser [] has joined #linode
15:09<LinodeJavaUser>anyone knows when the linode 200's will be available ?
15:09|-|tasaro [~tasaro@] has left #linode []
15:09<LinodeJavaUser>i need one kinda urgently ...
15:11<LinodeJavaUser>anyone from linode here ?
15:18|-|SpaceHobo [] has joined #linode
15:19<linbot>afv-13: Linode100 - 1, Linode150 - 0, Linode200 - 0, Linode300 - 24, Linode400 - 24
15:19<afv-13>no eta
15:20<@caker>one moment
15:20<afv-13>maybe the site says?
15:20<linbot>caker: Linode100 - 1, Linode150 - 0, Linode200 - 2, Linode300 - 24, Linode400 - 24
15:21<LinodeJavaUser>how do i sign up for a linode 200 if its not on the webpage ?
15:21|-|spr [] has quit [Quit: Spoon!]
15:22<@caker>LinodeJavaUser: they're available now
15:22<LinodeJavaUser>oh ok. kewl.
15:22|-|LinodeJavaUser [] has left #linode []
15:34|-|hirstp [] has joined #linode
15:34|-|Newsome [] has left #linode [Linux: Now with employee pricing!]
15:39|-|hirstp [] has quit [Quit: CGI:IRC (Ping timeout)]
15:53|-|bendy24 changed nick to scott
15:59|-|FireSlash_ [~FireSlash@] has quit [Read error: Connection reset by peer]
16:01|-|scott changed nick to bendy24
16:01|-|bendy24 changed nick to scott
16:14|-|internat [] has quit [Quit: This computer has gone to sleep]
16:38|-|linville [] has quit [Quit: Leaving]
16:40[~]warewolf fg
16:51|-|Bobo [] has joined #linode
16:53<Bobo>My token_refill has gone down to 50 and my server is basically toast. Any ideas?
16:56<@caker>Bobo: I'll check, one sec
16:56<@caker>Filename Type Size Used Priority
16:56<@caker>/dev/ubdb partition 263160 113056 -1
16:56<@caker>OK io_count=2172239 io_rate=34 io_tokens=-1 token_refill=50 token_max=2000
16:56<@caker>yeah, a little swap heavy there
16:57<@caker>I've increased them a bit, hopefully you can figure out what's eating your VM
16:57<Bobo>Thanks. I'm comfortable using a little swap, what do you folk consider reasonable?
16:58<@caker>As long as it's not paging too often, swap can be OK. around 70M is typically when I see paging, worse past 80-100
17:00|-|Newsome [] has joined #linode
17:01[~]warewolf takes a stroll down memory lane
17:01<warewolf><mr_bill> office-wide email: "anybody lose a ring? c leaning crew found one in the hallway"
17:01<warewolf><mr_bill> developer replies: "Look for Gollum, he's in a windowless office somewhere and REALLY wants it back"
17:08<Bobo>Aha - found out apache was using all of my I/O's. Any chance I could have my 512 back?
17:10<Bobo>Woo hoo thanks. I appreciate the help, but it might be an idea to warn customers next time - I've spent a LOOONNNGG time finding out why my Linode suddenly became real slow.
17:11<@caker>indeed, we need to something
17:23|-|Shaun [] has quit [Ping timeout: 480 seconds]
17:28|-|Shaun [] has joined #linode
17:31|-|afv-13 [] has quit [Quit: The conversation has been moved to the Trash.]
17:36|-|Bobo [] has quit [Quit: CGI:IRC (Ping timeout)]
17:37|-|cmantito [] has joined #linode
18:00|-|encode [] has quit [Quit: leaving]
18:03|-|encode [] has joined #linode
18:07|-|SpaceHobo [] has quit [Ping timeout: 480 seconds]
18:31|-|Battousai [~bryan@2002:d8a4:1c9e:2:20e:35ff:febc:b116] has quit [Quit: Leaving]
18:58|-|jekil [] has quit [Ping timeout: 480 seconds]
19:20|-|pingu [] has joined #linode
19:20|-|FireSlash [~FireSlash@] has joined #linode
19:21<pingu>Does anyone know if it's possible to get a LAN going between linodes, across hosts? For clustering and such? It doesn't seem to be a service offered.
19:27<@caker>pingu: you can, using your public IPs, but bandwidth would be accounted
19:28<@caker>(not intentionally, just by the current design)
19:29<pingu>Yeah, doesn't work that well then. You can't suggest a solution?
19:30<pingu>I mean, there just isn't that much scalability or reliably in a single linode, nomatter how good a job linode may do.
19:30<@caker>not until we can implement something, which is about medium-low on the todo list
19:31<pingu>OK, thanks.
19:31<@caker>what are you trying to accomplish?
19:31<@caker>or rather, what problem is it you're faced with?
19:31<pingu>Well, I really just needed high avaliability, and scalability.
19:32<@caker>yeah, being able to virtual-LAN connect nodes would be nice
19:32<@caker>loadbalancing, etc
19:32<pingu>On a more technical level... What would be ideal is.
19:32<pingu>being able to pay for all the bandwith on one front end host.
19:32<pingu>The load balancer.
19:33<pingu>Then just the hardware on the back end hosts, and maybe a flat rate for the connection between them.
19:33<pingu>So, really just a way to stick a crossover between all of my linodes.
19:33<@caker>Yes, when I implement this, the local traffic wouldn't count against your b/w, and you'd be able to pool the xfer resources of your accounts
19:34<pingu>Definately doesn't sound like a small task though :)
19:34<@caker>it's not that bad -- the recent updates I did to the back-end host libraries had provisioning to implement that
19:34<@caker>so more than half the work is already done
19:35<@caker>minus the bean counting part, it's possible to do now just by assigning more IPs from a non-routable IP space to the nodes in question
19:36<pingu>bean counting sorry?
19:36<@caker>transfer accounting (bits/beans)
19:36<pingu>Were you planning on charging extra for the crossover?
19:36<pingu>Wow, great :)
19:37<@caker>if you've got multiple nodes in the same DC now, we can give it a shot
19:37<pingu>No, just one :\
19:38<@caker> <-- progress
19:38<pingu>Oh, so.
19:39<@caker>render zone (into bind's format) done, can checkzone it, domain status can be changed from on/off, edit mode (continue serving old domain, but don't render new one yet)
19:39<pingu>You're providing free DNS?
19:39<@caker>Yes, soon
19:39<pingu>That's definately a perk!
19:39<@caker>it can also import a domain via an axfr :)
19:42<pingu>So, without any pressure here.
19:42<pingu>Would you say this crossover functionality could be implemented within say, six months?
19:43<@caker>it would work now (hint hint), just that the bandwidth accounting would still count it
19:44<@caker>also, the bandwidth overages are handled manually, so I could make a note on your account regarding your setup...
19:46<pingu>Yeah, sure. But that isn't all that maintainable for you.
19:46<@caker>not for long term, no, but special-for-you, today only!
19:47<pingu>haha :)
19:47<@caker>yeah, it'll be a few months for the real deal
19:47<pingu>You just want an excuse to test it!
19:47<@caker>damn, busted.
19:47<pingu>Alas, a few months is fine.
19:49<pingu>Oh, did you know the stability of selinux and xen? And UML for that matter?
19:50<@caker>Xen is getting more stable -- our test box has been up for over(?) a month now, it's required monthly reboots (at least) since then
19:50<@caker>selinux and UML I haven't played with
19:51<pingu>Yeah, we use xen reasonably extensively and the newest releases are seeming more stable.
19:52<pingu>Wait, so you're runing selinux on your dom-0? Or did you mis-read my question?
19:52<linbot>New news from forums: Sticky: DNS Manager README [Updated] in DNS Manager (pre-beta) <>
19:52<@caker>not running selinux
19:53<@caker>now, if I could only find out the rule for escaping/allowing quotes in TXT and SRV records
19:53<pingu>It looks like selinux runs fine on UML, there's no facility to build our own kernels and use those personally is there?
19:54<@caker>no -- I haven't been following selinux, is it in mainline now?
19:54<pingu>I don't think so.
19:54<pingu>Quite stable though.
19:54<@caker>can it can be turned off (or is by default?) and function normally?
19:55<pingu>It can be turned off, it isn't on by default.
19:55<pingu>Yes, it will function perfectly normally.
19:55<@caker>I'll look at including it in the next 2.6-um release
19:56<pingu>Oh, did you mean in the main kernel?
19:56<pingu>I don't think it will play nice with normal systems, sorry I didn't quite read that.
19:56<pingu>The patchset is off by default.
19:56<@caker>I meant is it distributed via, or a patch
19:56<pingu>Well, it isn't a patchset but anyway!
19:56<pingu>It's it's own kernel :)
19:57<pingu>I wouldn't put it in your main kernel.
19:57<@caker>never mind :)
19:57<pingu>Never mind what?
19:58<@caker>SELinux support is already included in the mainline Linux 2.6 kernel
19:58<@caker>so it IS in kernels; it's a .config option
19:58<pingu>Wow, I had no idea.
19:58<pingu>I don't actually use it in production.
19:59<pingu>So, this makes it fine by linode's policy?
19:59<@caker>no, just easier to maintain
19:59<@caker>and gives me warm and fuzzies that it's in mainline, too, but...
20:00<@caker>could everyone live with an SElinux enabled 2.6 kernel, or would I need both (non-se and se) kernels? (Like I said, I've never used it :)
20:01<pingu>I'd say you would want non-se and se.
20:01<pingu>I'm not sure if there's an easy way to pass.
20:02<pingu>Some kind of a... turn_me_on switch to the kernel at boot.
20:02<@caker>enforce=0 and selinux=0, looks like
20:03<pingu>Great, even easier :)
20:03<pingu>Can I specify kernel options in my control area?
20:04<@caker>no, we've got to specifically add support for kernel args (imagine: ubdg=/etc/passwd, or something)
20:04<pingu>Right :)
20:06<pingu>Well, I'd better get some work done. Thanks again.
20:06<@caker>thanks for your input
20:37|-|FireSlash [~FireSlash@] has quit [Read error: Connection reset by peer]
21:04|-|schultmc_ [] has joined #linode
21:31|-|cmantito [] has quit [Quit: Outta here, yo.]
21:58<encode>so, im having an interesting issue with postfix on my ubuntu 6.06 linode
21:59<encode>the timezone is set correctly (yes i have rebooted after doing so)
21:59<encode>and `date` reports the correct date, time and timezone
21:59<encode>yet emails sent / received do not
21:59<encode>have the correct time info
21:59<encode>any ideas
21:59<@caker>what info do they have?
22:00<encode>what info does what have?
22:00<@caker>how off are the wrong dates?
22:00<encode>8 hours
22:01<encode>my local timezone is GMT+10
22:01<@caker>is postfix chrooted?
22:02[~]efudd podners
22:02<encode>Received: from <remote_mailserver> (<remote_mailserver> [<ip address>]) by <local_mailserver> (Postfix) with ESMTP id 3A301C4AF for <email@account>; Tue, 24 Oct 2006 12:53:31 +1000 (EST)
22:02<efudd>so, whatever was pissing off my node a couple of weeks back has gone away I guess caker. :)
22:02<encode>Subject: test
22:02<encode>To: email@account
22:02<encode>X-Mailer: Lotus Notes Release 7.0 August 18, 2005
22:02<encode>Message-ID: <OF92A97CAE.49E6720A-ONCA257211.000FD4F7-4A257211.000FE25F@<remote_mailserver>>
22:02<efudd>well, i guess a month ago now.
22:02<encode>Date: Tue, 24 Oct 2006 12:53:03 +1000
22:02<encode>no, postfix isnt chrooted
22:03<encode>thers a copy of the mail headers from an email i just sent
22:03<encode>headers look ok to me
22:03<encode>ooh, hang on a sec
22:03<@caker>efudd: nice
22:04<encode>ok, so now i know what the problem was - webmail timezone setting
22:04<encode>but, recently people have complained that my emails arrived in the past
22:04<encode>which is why i thought config of mail server or soemthing
22:04<encode>oh well, never mind
22:05<encode>thanks for your brain-exertions anyway
22:45|-|Dreamr_3 [] has joined #linode
22:49|-|Dreamr3 [] has quit [Ping timeout: 480 seconds]
22:59|-|VS_ChanLog [] has left #linode [Rotating Logs]
22:59|-|VS_ChanLog [] has joined #linode
23:04|-|schultmc_ [] has quit [Quit: Client exiting]
---Logclosed Tue Oct 24 00:00:23 2006