1338951434 J * fisted_ ~fisted@xdsl-84-44-225-236.netcologne.de
1338951847 Q * fisted Ping timeout: 480 seconds
1338956160 M * Bertl off to bed now ... have a good one everyone!
1338956167 N * Bertl Bertl_zZ
1338958385 J * ghislain ~AQUEOS@adsl2.aqueos.com
1338964790 M * ncopa the 3.3.7 patch applies on 3.3.8
1338964797 M * ncopa building it now and will test
1338966127 J * petzsch ~markus@dslb-094-222-075-108.pools.arcor-ip.net
1338968474 Q * petzsch Read error: Connection reset by peer
1338971416 M * ncopa boots just fine in qemu
1338971591 J * Esmil esmil@horus.0x90.dk
1338972976 Q * cuba33ci Remote host closed the connection
1338972990 J * cuba33ci ~cuba33ci@114-36-236-70.dynamic.hinet.net
1338976834 N * disposab1e disposable
1338980203 Q * PowerKe Ping timeout: 480 seconds
1338981649 J * arthus ~arthus@reverse-232.fdn.fr
1338983234 J * admint ~admint@LLagny-156-36-29-11.w80-14.abo.wanadoo.fr
1338983640 J * PowerKe ~tom@d54C686CE.access.telenet.be
1338985418 N * Bertl_zZ Bertl
1338985422 M * Bertl morning folks!
1338986079 M * Esmil Hi, have you ever seen something like this before: http://old.nabble.com/Very-High-Load-on-Dovecot-2-and-Errors-in-mail.err.-tt33856207.html#a33856207
1338986150 M * Esmil We have the same problem on our servers, but we haven't yet been able to establish if it actually the vserver patches that does it
1338986175 M * Bertl kernel/patch/util-vserver version?
1338986177 M * Esmil It's just funny that this guy is running a vserver kernel too
1338986233 M * Esmil Well, we were running 3.0.28 and vserver 2.3.2.3 like him
1338986276 M * Esmil Now we've updated to 3.3.7 and vserver 2.3.3.4 and the symptons aren't as bad there, but the problem is still there
1338986294 M * Esmil Especially when the servers receive a lot of connections
1338986785 M * Esmil What I mean is that with the 3.3.7 kernel those imap processes left in D state tend to wake up and exit quicker, so they won't pile up as much
1338986850 M * Bertl I don't think that those 'D' states and/or the behaviour is in any way Linux-VServer specific
1338986866 M * Esmil Ok, thanks
1338986872 M * Bertl mainly because we do not modify the mainline schedulers anymore
1338986881 M * Esmil Right
1338986916 M * Bertl but I have observed that certain workloads tend to fill up queues in the kernel which result in similar effects
1338987029 M * Esmil Ahh ok
1338987184 M * Esmil Btw. whatever it is they're waiting for it seems to be pr. vserver, because the processes tend to pile up and then once in a while all processes in the same vserver wakes up and exits
1338987518 M * Bertl I more presume it is that they are waiting for the same 'event' to happen (e.g. queue to complete, etc)
1338987564 M * Bertl as processes across guests are unlikely to wait for the same event (because separate by namespaces and cgroups) it seems to be an expected behaviour
1338987612 M * Esmil Right, ok
1338987699 M * Bertl off for now .. bbs
1338987703 N * Bertl Bertl_oO
1338990529 Q * arthus Quit: "Smoke me a kipper, I'll be back for breakfast"
1338993163 J * BenG ~bengreen@cpc10-aztw24-2-0-cust114.aztw.cable.virginmedia.com
1338993767 Q * PowerKe Ping timeout: 480 seconds
1338994476 N * ensc Guest2926
1338994485 J * ensc ~irc-ensc@p54ADEA69.dip.t-dialin.net
1338994894 Q * Guest2926 Ping timeout: 480 seconds
1338996008 Q * alex3 Ping timeout: 480 seconds
1338996719 J * PowerKe ~tom@d54C686CE.access.telenet.be
1338996772 J * fisted ~fisted@xdsl-87-78-13-83.netcologne.de
1338996937 Q * fisted_ Ping timeout: 480 seconds
1338998563 Q * BenG Quit: I Leave
1338999802 J * BenG ~bengreen@cpc10-aztw24-2-0-cust114.aztw.cable.virginmedia.com
1339000335 Q * BenG Quit: I Leave
1339000542 J * bonbons ~bonbons@2001:960:7ab:0:9d8c:8601:da39:26c1
1339001738 Q * ensc|w Remote host closed the connection
1339001746 J * ensc|w ~ensc@www.sigma-chemnitz.de
1339006338 Q * nicholi Read error: Connection reset by peer
1339012206 J * hijacker_ ~hijacker@cable-84-43-134-121.mnet.bg
1339012627 Q * nou Ping timeout: 480 seconds
1339014082 Q * hijacker_ Quit: Leaving
1339016118 J * eptadone ~eptadone@2-227-125-184.ip186.fastwebnet.it
1339016590 J * eptadone_ ~eptadone@2-227-125-184.ip186.fastwebnet.it
1339016702 Q * eptadone Ping timeout: 480 seconds
1339016717 N * eptadone_ eptadone
1339018724 Q * bonbons Quit: Leaving
1339019285 N * BobR_zZ BobR
1339021208 Q * ghislain Quit: Leaving.
1339021543 Q * eptadone Ping timeout: 480 seconds
1339022179 N * BobR BobR_zZ