Back to Home / #linode / 2004 / 10 / Prev Day | Next Day
#linode IRC Logs for 2004-10-12

---Logopened Tue Oct 12 00:00:13 2004
00:11--- ---> claviola [claviola@] has joined #linode
00:20--- ---> claviola_ [claviola@] has joined #linode
00:21--- ---> claviola__ [claviola@] has joined #linode
00:21--- <<-- claviola__ [claviola@] has quit (Quit: )
00:28--- <<-- claviola [] has quit (Ping timeout: 480 seconds)
00:29--- <<-- spr [] has quit (Quit: ..must sleep..)
00:32--- <<-- Zymurgy [] has quit (Quit: Leaving)
00:41--- ---> claviola__ [claviola@] has joined #linode
00:48--- <<-- claviola_ [claviola@] has quit (Ping timeout: 480 seconds)
03:52--- ---> Guest43 [] has joined #linode
03:52--- <<-- adamg [] has quit (Read error: Connection reset by peer)
04:33--- Channel: changed the topic of #linode to: -- The Linode Virtual Server hang out | | |
05:11--- <<-- Shaitan [~Shaitan@] has quit (Quit: Leaving)
05:14--- User: *** Ciaran[zzz] is now known as Ciaran
05:14Ciaran Yay, new Linodes! :D
06:36--- <<-- adapter [] has quit (Quit: The lamest quit message in the universe.)
06:36--- ---> adapter [] has joined #linode
07:32--- <--- alnr [] has left #linode (Leaving)
07:43--- <<-- Qball [] has quit (Ping timeout: 480 seconds)
08:09--- ---> RagnaRok [] has joined #linode
08:16--- <<-- RagnaRok [] has quit (Quit: CGI:IRC (Ping timeout))
08:43Ciaran Okay. How cool is this?
08:43Ciaran Multiplayer Bubble Bobble. :D
09:32--- <<-- claviola__ [claviola@] has quit (Remote host closed the connection)
10:34--- <<-- anderiv [] has quit (Quit: leaving)
10:34--- ---> anderiv [] has joined #linode
10:35--- ---> spr [] has joined #linode
10:35--- <<-- spr [] has quit (Remote host closed the connection)
11:04--- <<-- ossido [] has quit (Remote host closed the connection)
11:04--- ---> ossido [] has joined #linode
12:01--- User: *** tj-tp is now known as tj|work
12:03--- User: *** tj|work is now known as tj-tp
12:24--- Netsplit <-> quits: ElectricElf, BB, cow, chris, Efudd, ossido, kso, jh_, UML_ChanLog
12:26--- Netsplit over, joins: jh_
12:26--- Netsplit over, joins: chris
12:27--- Netsplit over, joins: kso
12:27--- ---> ElectricElf [] has joined #linode
12:34--- ---> Efudd [] has joined #linode
12:35--- ---> BB [] has joined #linode
12:35--- ---> ossido [] has joined #linode
12:36--- ---> UML_ChanLog [] has joined #linode
12:36--- ---> cow [] has joined #linode
12:37--- ---> tierra|w [] has joined #linode
12:40--- ---> Redgore [Redgore@] has joined #linode
12:59--- ---> bji [] has joined #linode
12:59bji Hello
13:00bji host31 is really slow right now
13:00bji load is > 5
13:00bji > 6
13:01bji > 7
13:01bji > 8
13:02BB have you sent a msg to caker?
13:02bji > 10
13:02bji > 11
13:02tjfontaine supprt ticket?
13:02bji > 12
13:02SupaDongzu haha
13:02tjfontaine we get the picture
13:02SupaDongzu she's guna BLOW!
13:02* SupaDongzu hits the decks
13:02bji no haven't sent a ticket
13:03bji usually caker is in here and can fix the problem pretty quickly
13:03bji almost certainly it's a Linode on the host that's swap thrashing
13:03bji > 13 :)
13:03* SupaDongzu puts on his protective goggles
13:03bji I guess I'll submit a ticket
13:04bji but I have a feeling that the swap thrashing Linode will resolve itself before the ticket is even seen by caker
13:05bji Hm, it's back down to 15.40 from 16.17
13:06SupaDongzu Gents, we may yet survive this!
13:06SupaDongzu if so, I want you to promise that nothing said in the past few minutes leaves this bunker!
13:07bji SupaDongzu, what Linode plan do you have?
13:07bji 64? 96?
13:07tjfontaine SupaDongzu: I swear to always obey your wishes if we do, I know I have made this promise before but I mean it this time
13:07SupaDongzu bji: I'm just here for the free drinks
13:07bji heh
13:07* BB will sell the info for $10
13:07tjfontaine 192 on host 34
13:07tjfontaine and 128 on host 10
13:07bji So how is the performance of the 192s and 128s?
13:08bji Do you also get the occasional awful load spikes?
13:08* tjfontaine shrugs
13:08tjfontaine I've not had any real problems
13:08bji I've experienced this on host5, and then was moved to host31 which solved it temporarily - until host31 filled up. Now I get it there too.
13:08tjfontaine any swap thrashers are a few and far between
13:08bji The ironic thing is that I'm trying to install munin to keep track of the load problems
13:09bji And I can't get very far into the rpm build because of this load issue
13:09Efudd my 192 is fine.. but eh, it varies sometimes.. my load is never really bad, but I occasionaly see IO issues with stuff on the host.
13:09bji What do you mean by IO issues?
13:09bji Limiting?
13:10bji Or high load on your Linode caused by swap thrashers on the host?
13:10bji Or something else entrirely?
13:11bji holding steady at 8.67
13:12bji I wonder if Linode should just disallow swap
13:12bji I would sign up for a Linode plan hosted on a server on which swap was disallowed
13:13bji You'd have to keep all of your processes within the bounds of your RAM
13:13tjfontaine ...
13:13bji But there'd be no problem with swap thrashers
13:13BB that would be a bad idea
13:13bji Why?
13:13iggy great idea
13:13BB he would loose business
13:13bji I don't mean enforce this on everyone
13:13bji I just mean, give a choice
13:13iggy special host
13:13bji Join a host with no swap allowed, or join a normal host, your choice
13:20bji oh well
13:20bji I guess I'll just have to wait for the host to calm down on its own
13:20bji by all
13:20bji bye all
13:20--- <--- bji [] has left #linode ()
13:42--- ---> bji [] has joined #linode
13:42bji Is caker here yet?
13:42tjfontaine not yet
13:42bji ok thanks!
13:42--- <--- bji [] has left #linode ()
13:42BB have you sent him a support ticket yet?
13:42BB or maybe not
13:43tjfontaine *sigh*
13:57tjfontaine linbot: weather lhr
13:57linbot tjfontaine: The current temperature in Heathrow, United Kingdom is 55\xB0F (6:20 PM BST on October 12, 2004). Conditions: Light Rain Drizzle. Humidity: 88%. Dew Point: 52\xB0F. Visibility: 3.1 miles (5.0 kilometers). UV: 0 out of 12
14:02--- User: *** Guest43 is now known as adamg
14:05--- ---> wferrel [] has joined #linode
14:06--- ---> bji [] has joined #linode
14:06bji Hi, any sign of caker yet?
14:07tjfontaine no, did you file the ticket?
14:07bji yes
14:07bji no response
14:07tjfontaine welcome to the waiting game
14:07bji heh
14:07bji are you all waiting for him?
14:07tjfontaine just have patience
14:07tjfontaine no, just you so far
14:07bji cool
14:07bji thanks
14:08bji If you don't mind
14:08tjfontaine "I'm sorry all our other operators are busy right now, please stay online, your call is important to us"
14:08bji I'll check back in a little while
14:08tjfontaine no
14:08bji ok see ya!
14:08cow bji: its been slow for a good few hours already
14:08tjfontaine stay online
14:08cow im getting load of 3
14:08bji lucky you
14:08bji my load is > 9
14:08bji 10.91 to be exact
14:09bji I can stay online but
14:09bji I'll be switching to and from work consoles
14:09bji so I hope I don't miss anything important :)
14:09tjfontaine its not going anywhere
14:10tjfontaine plus if you time out the topic points you to the irc logs
14:10cow wow
14:10cow my load went down to 0.70
14:10bji what host are you on?
14:10bji I'm on host31
14:10cow okay... load average is in the last 15, 5, and 1 minutes?
14:10cow i can never remember
14:10cow host31
14:11bji 1, 5, and 15 minutes in that order
14:11bji bji$ uptime 14:11:10 up 12 days, 23:52, 2 users, load average: 9.89, 8.86, 6.56
14:11tjfontaine 14:11:32 up 36 days, 22:14, 4 users, load average: 0.00, 0.01, 0.00
14:11bji tjfontaine, what host is that?
14:12tjfontaine thats 10
14:12tj-tp 14:12:16 up 18:42, 2 users, load average: 0.38, 0.37, 0.48
14:12tj-tp thats 34 iirc
14:12tierra|w sighup, linode avail
14:12@sighup Linode availability -- [Linode 64: 42] [Linode 96: 30] [Linode 128: 15] [Linode 192: 10] [Linode 256: 10]
14:13cow does sighup have a host load checker?
14:13cow sighup, host31 load
14:13@sighup cow: excuse me?
14:13tjfontaine sighup: load
14:13@sighup tjfontaine: sorry...
14:13tierra|w that's a first in a long time there's been slots open for all
14:13bji yeah he added lots of new hosts recently
14:13tjfontaine 3 new boxes
14:13bji up to host37 if I recall correctly
14:15bji sigh
14:15bji I really wish I'd installed debian instead of redhat
14:15bji I have a feeling that installing munin would be much easier on debian
14:16bji also, just to ruffle a few more feathers, I absolutely hate, more than anything else on this great green planet, PERL AND IT'S STUPID CPAN SYSTEM
14:16bji has any software package ever written had more of a strung-together-out-of-bits-of-string-and-chewing-gum feel than Perl?
14:16bji I don't think so
14:16tjfontaine you're silly
14:17bji I actually met a guy here in NYC who wrote stock trading systems in Perl
14:17bji He was really proud of it, like Perl was something cool
14:17tierra|w bji, how long has host31 been acting up?
14:17bji host31 has been acting up for at least an hour and half, as far as I know
14:17bji I wasn't really using it much before that
14:18bji It could have been having problems before that too, I don't know
14:18tierra|w damn
14:18bji but the load average is coming down again
14:18bji it's only 4.95 now
14:19bji it will probably be OK for a couple of minutes
14:19bji then shoot up again
14:19bji that seems to be the pattern
14:19tierra|w my services are all still up and working, but I can't ssh in
14:20bji I guess the load must be spiking up again
14:20bji my bash shell won't even respond
14:20bji sigh
14:21bji why doesn't anyone offer really cheap true dedicated servers?
14:21bji I'd be happy with a Pentium 100 as long as I was the only one on it
14:21bji Hey the load hit a new all time high
14:21bji 22.65
14:21bji whee
14:21tierra|w damn
14:22tierra|w I can't even check if my linode is contributing
14:22tjfontaine sighup: ping caker
14:22@sighup caker: ping! ping! ping!
14:22@sighup caker was last seen on the network 19 hours, 3 minutes and 37 seconds ago, saying: <private message> [1097536754]
14:23* caker is working on it
14:23bji I'm sure mine is contributing, I'm running a perl -m CPAN command that seems to be installing the whole world
14:23tjfontaine o0
14:23* tjfontaine goes to hide in the corner
14:23bji But, my perl command alone is just a contributer to the problem, it's not the root cause, which is probably someone's swap thrashing Linode
14:24tierra|w I just remembered I did an emerge -uU world that included glibc, but that was yesterday... it should of been done within an hour or 2 at most
14:25bji ah, a gentoo user
14:25* tierra|w waits for bashing
14:25bji I guess you suffer as much as the rest of us anyway
14:25* caker finds a user with a 400M in swap ...
14:25* tierra|w uses maybe 2% of the CPU every month at most...
14:26bji what fun, my perl -m CPAN thing failed because one of the tests of one of the modules it was installing failed probably because the server was too slow
14:26tjfontaine 400mb swap?
14:26bji I wonder if my entire perl installation is now fuxored?
14:26@caker yeah .. I know what I'm working on this week :)
14:26bji it's not CPU that's the problem, it's I/O
14:26bji hey, it's the man himself
14:26bji I seem to be heavily into swap myself
14:27bji 98 MB in
14:28bji I don't know why spamd processes stick around for days and grow to godawful sizes
14:28bji Is that normal?
14:28tierra|w I stopped using it because of that
14:28tjfontaine you should restart it every 48hrs
14:28tierra|w it was doing it to me... I took that off sometime 2 weeks ago though
14:28tjfontaine if you actually use spamd
14:29tjfontaine same with clamd
14:29@caker bji: your node seems to be part of the problem
14:29tierra|w how about mine?
14:29@caker bji: but by far not the worst offender
14:29tierra|w I can't even login...
14:30@caker better now?
14:30tierra|w I got in!
14:30bji much better now
14:30bji In what way was I part of the problem?
14:30tierra|w io_tokens=399984
14:30tierra|w hmm... I'm fine
14:30@caker bji: I retract my statement -- it seems to be ok now
14:31bji yes, I know, but
14:31bji I just want to understand how I contributed
14:31bji I know I didn't help by running a perl -m CPAN when I knew that the host was having issues
14:31@caker bji: I saw a constant i/o rate, which usually means swapping
14:31@caker ahh
14:31bji I suppose I should have held off
14:31bji but once it was running, I was loathe to stop it
14:32bji it's zinging along now though
14:32tierra|w ok, yeah, my emerge update finished some time ago with no problems... my linode currently sitting at 60mb into swap, but no IO going on
14:33bji I'm 76 MB into swap, which sucks
14:33bji It's almost entirely httpd and spamd
14:33* tierra|w goes to check what's using that much as 60 isn't that normal for tierra|w
14:34@caker I've notified the repeat offender on host31
14:34@caker sorry about that
14:34bji that's ok, thank you for helping out
14:34bji it's amazing how quickly the problem is solved once you are notified
14:37bji Jeeze, restarting spamd reduced my swap usage by 60 MB
14:37tjfontaine caker++
14:38--- <<-- Redgore [Redgore@] has quit (Quit: )
14:42tierra|w umm... I still can't login except with Lish
14:44tierra|w the second I send a username, ssh drops the connection (well, assuming putty connected in the first place)
14:44tierra|w I can't see and memory allocation errors in the logs
14:44--- ---> Redgore [] has joined #linode
14:45@caker tierra|w: which distro, btw (does restarting sshd make it work?)
14:45tjfontaine thats what I was going to suggest
14:45tierra|w good question... gentoo
14:45tierra|w nope, restarting didn't work
14:46tjfontaine nothing in logs about sshd?
14:46@caker tierra|w: I see the same thing trying to ssh into your IP
14:46tjfontaine death on reverse?
14:46tierra|w wait, there's some logs
14:47tierra|w Oct 12 14:45:24 ibaku sshd[1762]: Received signal 15; terminating.
14:47tierra|w Oct 12 14:45:26 ibaku sshd[30282]: error: Bind to port 22 on failed: Address already in use.
14:47tierra|w Oct 12 14:45:26 ibaku sshd[30282]: fatal: Cannot bind any address.
14:47tierra|w Oct 12 14:46:11 ibaku sshd[30291]: Illegal user caker from ::ffff:
14:48@caker the bind to port 22 on is normal
14:49tierra|w oh, those are restart messages
14:49BB i'd restart it again if it cant bind to port 22 on
14:49tierra|w ibaku root # /etc/init.d/sshd restart
14:49tierra|w * Stopping sshd... [ !! ]
14:50tierra|w now it had a problem this time
14:50BB hmm "rm /var/lib/init.d/started/sshd" and try a start?
14:50tierra|w tcp 0 0 *:ssh *:* LISTEN 1169/sshd
14:50tierra|w netstat shows it's running and bound
14:51BB hmmm
14:51BB weird
14:51tjfontaine lsof
14:51tjfontaine kill the process thats still binding to it
14:51tierra|w yeah, that was my next step
14:51--- <--- bji [] has left #linode ()
14:52@caker the "can't bind" message is normal -- it's caused by multiple ListenAddress lines in sshd_config
14:52@caker #ListenAddress
14:52@caker #ListenAddress ::
14:52tierra|w * Starting sshd... [ ok ]
14:52tierra|w excellent, fixed
14:53tierra|w thanks all
14:57* adamg doesnt like his io settings, since he rebooted
15:16iggy quit being an io thrasher then
15:16iggy ;)
15:16--- <<-- anderiv [] has quit (Quit: leaving)
15:16* Efudd is thrashing IO!
15:17Efudd io_count=158507511 io_rate=0 io_tokens=400000 token_refill=512 token_max=400000
15:18--- ---> schwuk [] has joined #linode
15:18BB hmmm schwuk from cumbria?
15:19schwuk Yes. You are?
15:19BB a member of lugmaster list ;) (from Tyneside)
15:19schwuk Chunky?
15:19schwuk Ah
15:20BB thought i recognised the nick
15:20schwuk You got yourself a Linode then?
15:20BB yeah had one a while
15:20schwuk & what do u think?
15:20BB errr... it works for me,
15:20tjfontaine its fantastic
15:20BB -,
15:21BB what tjfontaine said, except i cant spell :)
15:22schwuk BB - that's pretty much my response - "it just works"
15:23--- ---> Ebox [] has joined #linode
15:26--- <--- Ebox [] has left #linode ()
15:28Efudd updating glibc==tehsux
15:30--- ---> pedro [] has joined #linode
15:37tierra|w Efudd, just did that yesterday... lots of fun
15:37--- <<-- schwuk [] has quit (Quit: Exits stage right)
16:29--- ---> anderiv [] has joined #linode
16:32You_Wish Linode Rules!
16:32--- ---> sh8d0w [] has joined #linode
16:32--- <<-- You_Wish [] has quit (Quit: )
16:33--- <<-- anderiv [] has quit (Quit: )
16:34--- ---> anderiv [] has joined #linode
16:42--- ---> Newsome [~sorenson@] has joined #linode
16:45--- ---> Redgore_ [] has joined #linode
16:45--- <<-- Redgore [] has quit (Read error: Connection reset by peer)
16:48--- ---> dtrumb [] has joined #linode
16:48dtrumb How do I check for an open mail relay?
16:56dtrumb I have a message sent to a yahoo account in my mail.log file, status = sent.
16:57dtrumb No one has access to send messages yet.
16:57BB who does it say sent it? or where does it say it came from?
16:58BB thought u said thats where it was sent to?
16:58--- <<-- Redgore_ [] has quit (Quit: )
16:59dtrumb I did.
16:59@caker outlook is pissing me off ...
17:00* tjfontaine kills outlook for caker
17:00@caker ty
17:00tjfontaine yw
17:00@caker my box crashed, and now it wants me to compact all 100 folders
17:00@caker can't check email
17:00tjfontaine ah
17:03dtrumb BB: nm, figured it out.
17:03BB :)
17:04dtrumb Didn't realize people where signing up for the site. Now I have to figure out how many signed up before I had email running. ;)
17:05BB heh
17:07--- <<-- dtrumb [] has quit (Quit: dtrumb has no reason)
17:48linbot New news from forums: Mysql Crash in Linux, Apache, Mysql and PHP (LAMP) Forum || Newb Question - updatedb / locate in General Discussion
18:00--- ---> claviola [claviola@] has joined #linode
18:00pedro Fala, vivi.
18:01--- User: *** pedro is now known as plamarao
18:04--- <<-- sh8d0w [] has quit (Quit: ...)
18:11--- ---> claviola_ [claviola@] has joined #linode
18:19--- <<-- claviola [] has quit (Ping timeout: 480 seconds)
18:22--- ---> claviola__ [claviola@] has joined #linode
18:30--- <<-- claviola_ [claviola@] has quit (Ping timeout: 480 seconds)
18:42--- ---> spr [] has joined #linode
19:04--- ---> claviola_ [claviola@] has joined #linode
19:11--- <<-- claviola__ [claviola@] has quit (Ping timeout: 480 seconds)
19:12--- <<-- ossido [] has quit (Ping timeout: 480 seconds)
19:20--- ---> ossido [] has joined #linode
19:42mikegrb lolz
19:42npmr lol
19:42mikegrb lolz
19:42npmr lol
19:42mikegrb lolz
19:42npmr lol
19:42mikegrb lolz
19:42npmr lol
19:42npmr hey mikegrb
19:42BB ut oh
19:46mikegrb roflz
19:46tierra|w rofl
19:46mikegrb roflz
19:46tierra|w rofl
19:52--- ---> Zymurgy [] has joined #linode
20:07mikegrb lolz
20:07linbot lol
20:12mikegrb lolz
20:12SupaDongzu lol
20:13mikegrb roflz
20:13SupaDongzu roflz
20:13SupaDongzu linbot: lol
20:13SupaDongzu !lol
20:13SupaDongzu 17:13 <linbot> Error: 'lol' is not a valid command.
20:13--- <--- BB [] has left #linode (mikegrb - scripted tartness)
20:18--- <<-- claviola_ [claviola@] has quit (Read error: Connection reset by peer)
20:19--- ---> claviola [claviola@] has joined #linode
20:20SupaDongzu hi.
20:20--- ---> claviola_ [claviola@] has joined #linode
20:21SupaDongzu I see.
20:28--- <<-- claviola [] has quit (Ping timeout: 480 seconds)
20:32--- ---> claviola__ [claviola@] has joined #linode
20:37--- ---> claviola [claviola@] has joined #linode
20:37--- <<-- claviola__ [claviola@] has quit (Read error: Connection reset by peer)
20:38--- ---> claviola__ [claviola@] has joined #linode
20:40--- <<-- claviola_ [claviola@] has quit (Ping timeout: 480 seconds)
20:40--- ---> claviola_ [claviola@] has joined #linode
20:43--- <<-- claviola_ [claviola@] has quit (Quit: )
20:45--- <<-- claviola [] has quit (Ping timeout: 480 seconds)
20:45--- <<-- tierra|w [] has quit (Quit: In God we trust; all others must pay cash.)
20:46--- <<-- claviola__ [claviola@] has quit (Ping timeout: 480 seconds)
21:11--- ---> Shaitan [~Shaitan@] has joined #linode
21:13--- <<-- npmr [] has quit (Ping timeout: 480 seconds)
21:31--- ---> hurg [] has joined #linode
21:35--- ---> kingmob [] has joined #linode
21:46linbot New news from forums: IO Tokens on different linode levels in Performance and Tuning
21:49--- ---> malaria444 [] has joined #linode
21:51--- <--- malaria444 [] has left #linode ()
22:31--- ---> dtrumb [] has joined #linode
22:32--- <--- dtrumb [] has left #linode ()
23:39--- <<-- hurg [] has quit (Ping timeout: 480 seconds)
23:43--- ---> hurg [] has joined #linode
23:59--- <--- UML_ChanLog [] has left #linode (Rotating Logs)
23:59--- ---> UML_ChanLog [] has joined #linode
---Logclosed Wed Oct 13 00:00:34 2004