#ceph IRC Log

Index

IRC Log for 2014-07-31

Timestamps are in GMT/BST.

[0:02] * kevinc (~kevinc__@client65-44.sdsc.edu) Quit (Quit: This computer has gone to sleep)
[0:12] * baylight (~tbayly@204.15.85.169) has left #ceph
[0:13] * brad_mssw (~brad@shop.monetra.com) Quit (Quit: Leaving)
[0:13] * aldavud (~aldavud@213.55.184.186) Quit (Ping timeout: 480 seconds)
[0:14] * cok (~chk@46.30.211.29) has joined #ceph
[0:19] * davidz1 (~Adium@cpe-23-242-12-23.socal.res.rr.com) has joined #ceph
[0:19] * davidz (~Adium@cpe-23-242-12-23.socal.res.rr.com) Quit (Read error: Connection reset by peer)
[0:20] * dmsimard is now known as dmsimard_away
[0:21] * i_m (~ivan.miro@nat-5-carp.hcn-strela.ru) Quit (Quit: Leaving.)
[0:24] * sarob_ (~sarob@nat-dip29-wl-c.cfw-a-gci.corp.yahoo.com) Quit (Remote host closed the connection)
[0:25] * sarob (~sarob@nat-dip29-wl-c.cfw-a-gci.corp.yahoo.com) has joined #ceph
[0:30] * rweeks (~rweeks@c-24-6-118-113.hsd1.ca.comcast.net) Quit (Quit: Leaving)
[0:30] * JuanEpstein (~rweeks@c-24-6-118-113.hsd1.ca.comcast.net) Quit (Quit: Leaving)
[0:30] * Guest4328 (~rweeks@c-24-6-118-113.hsd1.ca.comcast.net) Quit (Quit: Leaving)
[0:30] <mongo> burley: how many osd's per host per core?
[0:30] <mongo> note I have never run into cpu as a limiter.
[0:30] <burley> 12 OSDs per host with 6 core (12 with HT)
[0:30] * sarob (~sarob@nat-dip29-wl-c.cfw-a-gci.corp.yahoo.com) Quit (Read error: Operation timed out)
[0:30] * aldavud (~aldavud@217-162-119-191.dynamic.hispeed.ch) has joined #ceph
[0:30] <burley> 2GHz
[0:31] <mongo> what os and what backing fs?
[0:31] <burley> ext4 Ubuntu 14.04
[0:32] <mongo> 10 GbE?
[0:32] <burley> yes
[0:33] <burley> we have dropped replicas for now for our testing, knowing that'll have other implications as well, but at least free some CPU cycles so we can see what it can do without replication
[0:33] <mongo> how many hosts? even with erasure coding I don't have a huge cpu load on my 14.04 cluster
[0:33] * alram (~alram@cpe-172-250-2-46.socal.res.rr.com) has joined #ceph
[0:33] <burley> 3 OSDs
[0:33] <burley> was testing from one node
[0:34] <mongo> and you aren't running bridging etc? is the high cpu IO wait?
[0:34] <burley> was user space CPU usage
[0:34] <burley> no bridging
[0:34] <burley> top showed the OSD proc's as using all the CPU
[0:35] <burley> disks are all crucial m500 SSDs
[0:35] <mongo> oh, is this under rados bench?
[0:36] <burley> no, using the iometer.fio fio test
[0:36] <burley> tweaked iodepth to 256
[0:36] * Midnightmyth (~quassel@93-167-84-102-static.dk.customer.tdc.net) Quit (Ping timeout: 480 seconds)
[0:37] * dspano (~dspano@rrcs-24-103-221-202.nys.biz.rr.com) Quit (Quit: leaving)
[0:38] <burley> but even with replication effectively disabled, we're saturating the CPUs
[0:38] <mongo> I would look at checking with latency top. I assume you have a choke point.
[0:41] <burley> http://pastie.org/private/c5xrqrsdbhcs2rjlxerpca
[0:41] * PerlStalker (~PerlStalk@2620:d3:8000:192::70) Quit (Quit: ...)
[0:42] <burley> http://pastie.org/private/ix6kxvm3o7vmnncghhsnga
[0:42] <burley> ^-- that has full detail
[0:43] <mongo> did you use ceph-deploy?
[0:44] <burley> no
[0:44] <burley> here's another instance with more latency:
[0:44] <burley> http://pastie.org/private/yuk9ilznzouuqcappwfliq
[0:47] <mongo> ya, blocking on directio, sorry both my clusters are btrfs but I would revisit your mkfs and mount args.
[0:48] <burley> the blocking on directio looks like a small fraction of the delay
[0:48] <burley> at least as far as I am reading the output -- haven't used latencytop before
[0:49] * darkfader (~floh@88.79.251.60) Quit (Ping timeout: 480 seconds)
[0:50] * danieagle (~Daniel@179.184.165.184.static.gvt.net.br) has joined #ceph
[0:52] <mongo> ya, without being local it is hard but if you just did a mkfs.xfs with no options and the defaults for mounts it may be pretty expensive waiting for metadata though. what type of storage controller are you using though?
[0:54] <mongo> you have noatime and nodiratime on the mounts right?
[0:54] <burley> noatime
[0:55] <burley> lsi 9207
[0:55] * cok (~chk@46.30.211.29) Quit (Quit: Leaving.)
[0:55] <burley> noatime implies nodiratime
[1:01] * thomnico (~thomnico@sccc-66-78-236-243.smartcity.com) has joined #ceph
[1:05] * gregsfortytwo1 (~Adium@126-206-207-216.dsl.mi.winntel.net) Quit (Quit: Leaving.)
[1:09] * joef (~Adium@2620:79:0:131:a010:ae60:702a:d1e9) Quit (Remote host closed the connection)
[1:10] * hasues (~hasues@kwfw01.scrippsnetworksinteractive.com) Quit (Quit: Leaving.)
[1:12] * joef (~Adium@2601:9:280:f2e:b183:3e43:193e:9d8c) has joined #ceph
[1:12] * joef (~Adium@2601:9:280:f2e:b183:3e43:193e:9d8c) has left #ceph
[1:13] * Nats (~natscogs@2001:8000:200c:0:c11d:117a:c167:16df) has joined #ceph
[1:13] * aldavud (~aldavud@217-162-119-191.dynamic.hispeed.ch) Quit (Ping timeout: 480 seconds)
[1:14] * nljmo_ (~nljmo@5ED6C263.cm-7-7d.dynamic.ziggo.nl) has joined #ceph
[1:14] * nljmo (~nljmo@5ED6C263.cm-7-7d.dynamic.ziggo.nl) Quit (Read error: Connection reset by peer)
[1:19] * oms101 (~oms101@p20030057EA43C800EEF4BBFFFE0F7062.dip0.t-ipconnect.de) Quit (Ping timeout: 480 seconds)
[1:20] * rendar (~I@host41-180-dynamic.56-79-r.retail.telecomitalia.it) Quit ()
[1:21] * zerick (~eocrospom@190.187.21.53) Quit (Read error: Connection reset by peer)
[1:22] * dmsimard_away is now known as dmsimard
[1:24] * Chandler (~Chandler@111.199.90.213) Quit (Ping timeout: 480 seconds)
[1:25] * darkfader (~floh@88.79.251.60) has joined #ceph
[1:26] * ghost1 (~pablodelg@c-174-61-25-255.hsd1.fl.comcast.net) has joined #ceph
[1:26] * jakku (~jakku@ad041065.dynamic.ppp.asahi-net.or.jp) has joined #ceph
[1:27] * oms101 (~oms101@p20030057EA4D3800EEF4BBFFFE0F7062.dip0.t-ipconnect.de) has joined #ceph
[1:34] * mancdaz (~mancdaz@2a00:1a48:7807:102:94f4:6b56:ff08:886c) Quit (Ping timeout: 480 seconds)
[1:37] * dmsimard is now known as dmsimard_away
[1:39] * alram (~alram@cpe-172-250-2-46.socal.res.rr.com) Quit (Quit: leaving)
[1:44] * AfC (~andrew@nat-gw2.syd4.anchor.net.au) has joined #ceph
[1:47] * jakku (~jakku@ad041065.dynamic.ppp.asahi-net.or.jp) Quit (Quit: Leaving...)
[2:05] * ghartz (~ghartz@ircad17.u-strasbg.fr) Quit (Ping timeout: 480 seconds)
[2:07] * fsimonce (~simon@host133-25-dynamic.250-95-r.retail.telecomitalia.it) Quit (Quit: Coyote finally caught me)
[2:11] * JustATroll (~asdf@0001ca68.user.oftc.net) has joined #ceph
[2:11] * FloodBotGen1 (~FloodBotG@atlantic393.serverprofi24.com) has joined #ceph
[2:11] <FloodBotGen1> HELLO.
[2:11] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:11] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:11] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] * xDDDDxD (~xDDDDxD@atlantic393.serverprofi24.com) has joined #ceph
[2:12] <xDDDDxD> HELLO.
[2:12] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:12] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] * JustATroll is now known as Amazing
[2:13] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] * LeaChim (~LeaChim@host86-161-89-237.range86-161.btcentralplus.com) has joined #ceph
[2:13] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <xDDDDxD> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] <FloodBotGen1> BITCHES LOL XDDD XD XDDDDDDDD FEEEEEEEEEEEEEEEL MYYYYYYYYYYYY HAAAAAAAAAAAARRRRRTRRRRREEEEEEEEEEEEEED
[2:13] * dmick sets mode +b FloodBotGen1!*@*
[2:13] * ghost1 (~pablodelg@c-174-61-25-255.hsd1.fl.comcast.net) Quit (Quit: ghost1)
[2:13] * FloodBotGen1 (~FloodBotG@659AAMSZ2.tor-irc.dnsbl.oftc.net) Quit (Remote host closed the connection)
[2:13] * xDDDDxD (~xDDDDxD@659AAMSZ3.tor-irc.dnsbl.oftc.net) Quit (Remote host closed the connection)
[2:14] <dmick> children
[2:14] * Tamil1 (~Adium@cpe-108-184-74-11.socal.res.rr.com) Quit (Quit: Leaving.)
[2:14] <lurbs> They're not even every inventive these days, are they?
[2:15] <houkouonchi-home> if they really wanted to piss people off they should have jumped in here and spammed like gluster ownz ceph or something :P
[2:15] * ghartz (~ghartz@ircad17.u-strasbg.fr) has joined #ceph
[2:15] <houkouonchi-home> would have been more inventive than 'bitches'
[2:16] <dmick> wth is a a hartreed anyway
[2:17] * danieagle (~Daniel@179.184.165.184.static.gvt.net.br) Quit (Quit: Obrigado por Tudo! :-) inte+ :-))
[2:18] <Amazing> he's at #virt
[2:19] <dmick> oh the bot is
[2:22] * rmoe (~quassel@12.164.168.117) Quit (Ping timeout: 480 seconds)
[2:22] * Tamil1 (~Adium@cpe-108-184-74-11.socal.res.rr.com) has joined #ceph
[2:25] * bandrus (~oddo@216.57.72.205) Quit (Quit: Leaving.)
[2:25] * Amazing is now known as DInosaurio
[2:25] * Tamil1 (~Adium@cpe-108-184-74-11.socal.res.rr.com) Quit (Read error: Connection reset by peer)
[2:25] * Tamil1 (~Adium@cpe-108-184-74-11.socal.res.rr.com) has joined #ceph
[2:27] * DInosaurio is now known as Dinosaurio
[2:27] * thomnico (~thomnico@sccc-66-78-236-243.smartcity.com) Quit (Ping timeout: 480 seconds)
[2:30] * rmoe (~quassel@173-228-89-134.dsl.static.sonic.net) has joined #ceph
[2:30] * lucas1 (~Thunderbi@218.76.25.66) has joined #ceph
[2:31] * thomnico (~thomnico@8.25.222.10) has joined #ceph
[2:32] * KevinPerks (~Adium@cpe-098-025-128-231.sc.res.rr.com) has joined #ceph
[2:35] * LeaChim (~LeaChim@host86-161-89-237.range86-161.btcentralplus.com) Quit (Read error: Operation timed out)
[2:37] * Chandler (~Chandler@123.115.79.17) has joined #ceph
[2:39] * sputnik13 (~sputnik13@207.8.121.241) Quit (Quit: My MacBook has gone to sleep. ZZZzzz???)
[2:42] * lucas1 (~Thunderbi@218.76.25.66) Quit (Quit: lucas1)
[2:42] * Tamil1 (~Adium@cpe-108-184-74-11.socal.res.rr.com) Quit (Quit: Leaving.)
[2:43] * thomnico_ (~thomnico@8.25.222.10) has joined #ceph
[2:44] * thomnico (~thomnico@8.25.222.10) Quit (Ping timeout: 480 seconds)
[2:44] * KaZeR (~kazer@64.201.252.132) Quit (Remote host closed the connection)
[2:46] * Dinosaurio (~asdf@0001ca68.user.oftc.net) Quit (autokilled: This host is in violation of network policy. Contact support@oftc.net with questions. (2014-07-31 00:45:01))
[2:47] * xarses (~andreww@12.164.168.117) Quit (Ping timeout: 480 seconds)
[2:55] * thomnico_ (~thomnico@8.25.222.10) Quit (Quit: Ex-Chat)
[2:58] * zerick (~eocrospom@190.187.21.53) has joined #ceph
[2:58] * verdurin (~adam@2001:8b0:281:78ec:e2cb:4eff:fe01:f767) Quit (Ping timeout: 480 seconds)
[2:58] * KevinPerks (~Adium@cpe-098-025-128-231.sc.res.rr.com) Quit (Quit: Leaving.)
[3:01] * sjusthm (~sam@24-205-54-233.dhcp.gldl.ca.charter.com) Quit (Quit: Leaving.)
[3:03] * KindOne (kindone@0001a7db.user.oftc.net) Quit (Quit: ZNC - http://znc.in)
[3:04] * joef (~Adium@c-24-130-254-66.hsd1.ca.comcast.net) has joined #ceph
[3:16] * Tamil1 (~Adium@cpe-108-184-74-11.socal.res.rr.com) has joined #ceph
[3:16] * Tamil1 (~Adium@cpe-108-184-74-11.socal.res.rr.com) Quit ()
[3:19] * lucas1 (~Thunderbi@218.76.25.66) has joined #ceph
[3:21] * verdurin (~adam@2001:8b0:281:78ec:e2cb:4eff:fe01:f767) has joined #ceph
[3:23] * KindOne (kindone@0001a7db.user.oftc.net) has joined #ceph
[3:28] * bitserker (~toni@63.pool85-52-240.static.orange.es) Quit (Ping timeout: 480 seconds)
[3:31] * bitserker (~toni@63.pool85-52-240.static.orange.es) has joined #ceph
[3:34] * xarses (~andreww@c-76-126-112-92.hsd1.ca.comcast.net) has joined #ceph
[3:35] * gregsfortytwo1 (~Adium@126-206-207-216.dsl.mi.winntel.net) has joined #ceph
[3:36] * ircolle (~Adium@2601:1:a580:145a:fd00:bf10:f53f:1a5d) Quit (Read error: Connection reset by peer)
[3:36] * adamcrume (~quassel@2601:9:6680:47:5c9a:4982:2be6:983b) Quit (Remote host closed the connection)
[3:37] * ircolle (~Adium@2601:1:a580:145a:fd00:bf10:f53f:1a5d) has joined #ceph
[3:38] * lucas1 (~Thunderbi@218.76.25.66) Quit (Quit: lucas1)
[3:57] * vbellur (~vijay@122.172.106.250) has joined #ceph
[4:03] * joef (~Adium@c-24-130-254-66.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[4:06] * diegows (~diegows@190.190.5.238) Quit (Read error: Operation timed out)
[4:14] * lupu (~lupu@86.107.101.214) has joined #ceph
[4:20] * drankis (~drankis__@89.111.13.198) Quit (Ping timeout: 480 seconds)
[4:24] * zhaochao (~zhaochao@124.207.139.23) has joined #ceph
[4:33] * haomaiwa_ (~haomaiwan@223.223.183.114) Quit (Remote host closed the connection)
[4:34] * haomaiwang (~haomaiwan@203.69.59.199) has joined #ceph
[4:37] * Cube1 (~Cube@66-87-130-154.pools.spcsdns.net) has joined #ceph
[4:37] * Cube (~Cube@66-87-130-154.pools.spcsdns.net) Quit (Read error: Connection reset by peer)
[4:38] * nigwil (~oftc-webi@101.172.127.231) has joined #ceph
[4:39] * bkopilov (~bkopilov@213.57.17.162) Quit (Ping timeout: 480 seconds)
[4:41] * zerick (~eocrospom@190.187.21.53) Quit (Ping timeout: 480 seconds)
[4:44] * Cube (~Cube@66-87-130-154.pools.spcsdns.net) has joined #ceph
[4:44] * Cube1 (~Cube@66-87-130-154.pools.spcsdns.net) Quit (Read error: Connection reset by peer)
[4:46] * hasues (~hazuez@108-236-232-243.lightspeed.knvltn.sbcglobal.net) has joined #ceph
[4:46] <longguang> what does leveldb do in ceph?
[4:50] * haomaiwa_ (~haomaiwan@223.223.183.114) has joined #ceph
[4:50] * Cube (~Cube@66-87-130-154.pools.spcsdns.net) Quit (Read error: No route to host)
[4:51] * shang (~ShangWu@175.41.48.77) has joined #ceph
[4:53] * Cube (~Cube@66.87.130.154) has joined #ceph
[4:56] * haomaiwang (~haomaiwan@203.69.59.199) Quit (Ping timeout: 480 seconds)
[4:57] * Cube1 (~Cube@66-87-130-154.pools.spcsdns.net) has joined #ceph
[4:57] * Cube (~Cube@66.87.130.154) Quit (Read error: No route to host)
[5:01] * haomaiwang (~haomaiwan@223.223.183.114) has joined #ceph
[5:01] * haomaiwa_ (~haomaiwan@223.223.183.114) Quit (Read error: Connection reset by peer)
[5:03] * lucas1 (~Thunderbi@222.240.148.154) has joined #ceph
[5:03] * haomaiwang (~haomaiwan@223.223.183.114) Quit (Read error: Connection reset by peer)
[5:03] * haomaiwa_ (~haomaiwan@223.223.183.114) has joined #ceph
[5:05] * lucas1 (~Thunderbi@222.240.148.154) Quit (Remote host closed the connection)
[5:11] * haomaiwa_ (~haomaiwan@223.223.183.114) Quit (Remote host closed the connection)
[5:12] * haomaiwang (~haomaiwan@203.69.59.199) has joined #ceph
[5:14] * DV (~veillard@2001:41d0:1:d478::1) has joined #ceph
[5:14] * haomaiwa_ (~haomaiwan@223.223.183.114) has joined #ceph
[5:15] <dmick> it's the key-value store for monitor data and osd xattrs at least
[5:17] * wschulze (~wschulze@cpe-69-206-251-158.nyc.res.rr.com) Quit (Quit: Leaving.)
[5:19] * haomaiwang (~haomaiwan@203.69.59.199) Quit (Read error: Operation timed out)
[5:22] <longguang> dmick:xattrs? i thought attr is stored in file's meta.
[5:22] <longguang> file itself.
[5:22] * lxo (~aoliva@lxo.user.oftc.net) Quit (Ping timeout: 480 seconds)
[5:22] * zack_dol_ (~textual@p8505b4.tokynt01.ap.so-net.ne.jp) Quit (Quit: My MacBook has gone to sleep. ZZZzzz???)
[5:23] <dmick> some are, but if they get too big they have to spill somewhere, and we've gone back and forth about that; I think maybe "attrs larger than a certain size" are also immediately put in leveldb. It's at least some in leveldb if not all.
[5:23] <longguang> i only know pg directory has attr related to ceph. but do not find object has those attrs.
[5:24] <dmick> http://tracker.ceph.com/issues/7059 for instance
[5:25] <longguang> you mean a large quantity attrs are needed by ceph? whant knid of attrs?
[5:26] <dmick> I mean you can put attrs on RADOS objects
[5:26] <dmick> those get stored somewhere
[5:26] * wschulze (~wschulze@cpe-69-206-251-158.nyc.res.rr.com) has joined #ceph
[5:29] <longguang> what is the flag which tells xattrs is stored in leveldb ?
[5:30] <dmick> if you mean "what is the config option that makes Ceph store them in leveldb", I think it's the default now
[5:31] * Vacum_ (~vovo@i59F7AAD5.versanet.de) has joined #ceph
[5:31] * theanalyst (~abhi@106.51.241.107) has joined #ceph
[5:33] <longguang> that feature is ongoing?
[5:36] * lalatenduM (~lalatendu@122.172.39.176) has joined #ceph
[5:36] <dmick> I don't understand the question
[5:37] <longguang> does this feature is finished?
[5:38] <longguang> :)
[5:38] <dmick> when I say "Ceph does this", I mean "it does it", not that it will do so in the future
[5:38] <dmick> so yes
[5:38] <dmick> it is working that way today
[5:38] * Vacum (~vovo@88.130.214.70) Quit (Ping timeout: 480 seconds)
[5:39] <dmick> I'm sure I haven't given you all the uses of leveldb, but I know those are at least two
[5:40] <longguang> ' avoid leveldb check for xattr when possible', how ceph does the avoid?
[5:40] <longguang> which two?
[5:40] <dmick> (08:15:09 PM) dmick: it's the key-value store for monitor data and osd xattrs at least
[5:41] * rweeks (~rweeks@c-24-6-118-113.hsd1.ca.comcast.net) has joined #ceph
[5:41] <longguang> 'osd xattrs' what is that? i only know pg and object attrs.
[5:42] <dmick> I don't know what you mean by pg and object attrs
[5:42] <dmick> but by osd xattrs I mean "stuff the OSD stores for objects' RADOS xattrs"
[5:42] <dmick> which is partially in the underlying filesystem xattrs
[5:42] <dmick> and partially in leveldb
[5:43] <longguang> unconditionally maintain an xattr that indicates whether we spill over into DBObjectMap
[5:43] <longguang> do you know the xattr?
[5:44] * haomaiwa_ (~haomaiwan@223.223.183.114) Quit (Remote host closed the connection)
[5:44] <dmick> not offhand, but you're getting pretty close to the point where I'd recommend reading the source
[5:44] <dmick> what is your real question? Why are you looking into this?
[5:44] * jobewan (~jobewan@75.147.174-93-BusName-smpls.memphis.hfc.comcastbusiness.net) has joined #ceph
[5:44] * haomaiwang (~haomaiwan@203.69.59.199) has joined #ceph
[5:45] <longguang> learn of every part of ceph code.
[5:46] * angdraug (~angdraug@12.164.168.117) Quit (Quit: Leaving)
[5:48] <longguang> if at the first beginning ceph use leveldb to store xattrs. now ceph first choose to use fs's xattr except osd xattrs is too large?
[5:48] * haomaiwang (~haomaiwan@203.69.59.199) Quit (Read error: Connection reset by peer)
[5:48] <longguang> am i right?
[5:49] <dmick> I don't know; first, I'm not certain what your question is; second, I don't know the code at that level. Sorry. I recommend you read the code for yourself. Look for LevelDBStore.
[5:49] * scuttlemonkey is now known as scuttle|afk
[5:50] <longguang> ok
[5:53] * kanagaraj (~kanagaraj@117.197.204.119) has joined #ceph
[5:53] * wschulze (~wschulze@cpe-69-206-251-158.nyc.res.rr.com) Quit (Quit: Leaving.)
[5:56] <longguang> what are the files in 'current/meta' directory?
[5:59] <longguang> dmick:do you know any developers?
[6:00] * rweeks (~rweeks@c-24-6-118-113.hsd1.ca.comcast.net) Quit (Quit: Leaving)
[6:02] <dmick> current/meta is part of the implementation of the OSD FileStore
[6:02] <dmick> I both know developers and I am one
[6:03] * zack_dolby (~textual@ai126194010036.1.tss.access-internet.ne.jp) has joined #ceph
[6:04] <dmick> you might find some of the stuff in doc/dev helpful
[6:04] <longguang> like mds? i know all objects are stored in pg directory. i do not know what is stored in meta?
[6:06] <longguang> http://ceph.com/docs/master/dev/
[6:06] * bkopilov (~bkopilov@nat-pool-tlv-t.redhat.com) has joined #ceph
[6:07] <dmick> yep
[6:08] <longguang> what is collections?
[6:09] * zack_dolby (~textual@ai126194010036.1.tss.access-internet.ne.jp) Quit (Quit: Textual IRC Client: www.textualapp.com)
[6:09] * lucas1 (~Thunderbi@222.240.148.154) has joined #ceph
[6:10] * Chandler (~Chandler@123.115.79.17) Quit (Read error: No route to host)
[6:11] <longguang> it seems that there are 2 kinds of function can operate xattrs. one is started with collection, another is started with omap. how to differentiate them?
[6:11] <longguang> lauch.... thanks dmick.
[6:33] * lucas1 (~Thunderbi@222.240.148.154) Quit (Quit: lucas1)
[6:37] * ghartz (~ghartz@ircad17.u-strasbg.fr) Quit (Ping timeout: 480 seconds)
[6:47] * joef (~Adium@c-24-130-254-66.hsd1.ca.comcast.net) has joined #ceph
[6:47] * joef (~Adium@c-24-130-254-66.hsd1.ca.comcast.net) Quit ()
[6:48] * ghartz (~ghartz@ircad17.u-strasbg.fr) has joined #ceph
[6:48] * MACscr (~Adium@c-50-158-183-38.hsd1.il.comcast.net) Quit (Quit: Leaving.)
[6:50] * benner (~benner@162.243.49.163) Quit (Read error: Operation timed out)
[6:51] * benner (~benner@162.243.49.163) has joined #ceph
[6:58] * KindOne (kindone@0001a7db.user.oftc.net) Quit (Ping timeout: 480 seconds)
[6:58] * lalatenduM (~lalatendu@122.172.39.176) Quit (Read error: Operation timed out)
[6:59] * rdas (~rdas@121.244.87.115) has joined #ceph
[7:00] * sputnik13 (~sputnik13@wsip-68-105-248-60.sd.sd.cox.net) has joined #ceph
[7:10] * lalatenduM (~lalatendu@122.171.95.253) has joined #ceph
[7:22] * Cube1 (~Cube@66-87-130-154.pools.spcsdns.net) Quit (Quit: Leaving.)
[7:24] * Nats_ (~Nats@2001:8000:200c:0:f4b4:821:1f5a:23a8) has joined #ceph
[7:26] * theanalyst (~abhi@106.51.241.107) Quit (Ping timeout: 480 seconds)
[7:26] * cury (~cury@123.115.79.17) has joined #ceph
[7:29] * hasues (~hazuez@108-236-232-243.lightspeed.knvltn.sbcglobal.net) Quit (Quit: Leaving.)
[7:31] * Nats__ (~Nats@2001:8000:200c:0:75c2:4c57:c60f:148f) Quit (Ping timeout: 480 seconds)
[7:33] * lucas1 (~Thunderbi@222.240.148.130) has joined #ceph
[7:35] * hasues (~hazuez@108-236-232-243.lightspeed.knvltn.sbcglobal.net) has joined #ceph
[7:35] * hasues (~hazuez@108-236-232-243.lightspeed.knvltn.sbcglobal.net) Quit ()
[7:36] * lucas1 (~Thunderbi@222.240.148.130) Quit (Remote host closed the connection)
[7:40] * lupu (~lupu@86.107.101.214) Quit (Quit: Leaving.)
[7:40] * lupu (~lupu@86.107.101.214) has joined #ceph
[7:43] * nigwil (~oftc-webi@101.172.127.231) Quit (Quit: Page closed)
[7:45] * lupu (~lupu@86.107.101.214) Quit (Read error: Operation timed out)
[7:58] * lalatenduM (~lalatendu@122.171.95.253) Quit (Quit: Leaving)
[7:59] * lalatenduM (~lalatendu@122.171.95.253) has joined #ceph
[8:03] * drankis (~drankis__@91.188.43.210) has joined #ceph
[8:08] * thb (~me@2a02:2028:2d3:dab0:45e9:a354:3145:9258) has joined #ceph
[8:09] * tank100 (~tank@84.200.17.138) Quit (Read error: Operation timed out)
[8:11] * tank100 (~tank@84.200.17.138) has joined #ceph
[8:12] * drankis (~drankis__@91.188.43.210) Quit (Ping timeout: 480 seconds)
[8:15] * chowmeined (~chow@c-24-19-66-251.hsd1.wa.comcast.net) has joined #ceph
[8:16] * Nacer (~Nacer@203-206-190-109.dsl.ovh.fr) has joined #ceph
[8:17] <chowmeined> Hello, I am experimenting with a single-node test ceph configuration. I have the cluster running and its working roughly how I'd expect. The cluster is healthy. Now I'm trying to create an SSD tier. I've created a customized crush map with a different root for SSD OSDs and then I've created a pool and specified the ruleset using this ssd root. However, the issue I'm running into now is the cluster is "stuck unclean" for this newly creat
[8:17] <chowmeined> ed pool. How can I diagnose this further?
[8:28] * kanagaraj_ (~kanagaraj@117.216.106.38) has joined #ceph
[8:32] * kanagaraj (~kanagaraj@117.197.204.119) Quit (Read error: Operation timed out)
[8:37] * MACscr (~Adium@c-50-158-183-38.hsd1.il.comcast.net) has joined #ceph
[8:41] <chowmeined> okay, figured it out. Only had 2 SSDs and the pool replication size was 3 :)
[8:43] * kanagaraj (~kanagaraj@61.3.145.128) has joined #ceph
[8:44] * kanagaraj_ (~kanagaraj@117.216.106.38) Quit (Read error: Operation timed out)
[8:48] * cookednoodles (~eoin@eoin.clanslots.com) has joined #ceph
[8:50] <Clabbe> Any way to copy a pg from an old not in cluster node?
[8:51] <Clabbe> Something went wrong removing an osd :(
[8:51] * madkiss (~madkiss@chello084112124211.20.11.vie.surfer.at) Quit (Ping timeout: 480 seconds)
[8:51] * michalefty (~micha@p20030071CE596925891354B9A104B6D1.dip0.t-ipconnect.de) has joined #ceph
[8:55] * lxo (~aoliva@lxo.user.oftc.net) has joined #ceph
[8:58] * jksM (~jks@3e6b5724.rev.stofanet.dk) Quit (Ping timeout: 480 seconds)
[8:58] * jksM_ (~jks@3e6b5724.rev.stofanet.dk) Quit (Ping timeout: 480 seconds)
[9:01] * hyperbaba (~hyperbaba@private.neobee.net) has joined #ceph
[9:03] * michalefty (~micha@p20030071CE596925891354B9A104B6D1.dip0.t-ipconnect.de) has left #ceph
[9:05] * mtl1 (~Adium@c-67-174-109-212.hsd1.co.comcast.net) has joined #ceph
[9:05] * mtl2 (~Adium@c-67-174-109-212.hsd1.co.comcast.net) Quit (Read error: Connection reset by peer)
[9:05] * steki (~steki@91.195.39.5) has joined #ceph
[9:05] * Nacer (~Nacer@203-206-190-109.dsl.ovh.fr) Quit (Remote host closed the connection)
[9:08] * i_m (~ivan.miro@nat-5-carp.hcn-strela.ru) has joined #ceph
[9:11] * AfC (~andrew@nat-gw2.syd4.anchor.net.au) Quit (Quit: Leaving.)
[9:12] * sputnik13 (~sputnik13@wsip-68-105-248-60.sd.sd.cox.net) Quit (Quit: My MacBook has gone to sleep. ZZZzzz???)
[9:17] * astellwag|afk is now known as astellwag
[9:18] * Teduardo (~Teduardo@57.0.be.static.xlhost.com) Quit (Read error: Connection reset by peer)
[9:18] * Teduardo (~Teduardo@57.0.be.static.xlhost.com) has joined #ceph
[9:27] <chowmeined> The rbd driver in the kernel I'm on (3.13) doesn't appear to support cache tiers. I'm trying to test with rbd-fuse but when I 'ls' the mountpoint it hard locks for about 5 minutes and then dismounts. I can't find any logs anywhere, are they in a special location?
[9:29] * b0e (~aledermue@juniper1.netways.de) has joined #ceph
[9:29] * fsimonce (~simon@host133-25-dynamic.250-95-r.retail.telecomitalia.it) has joined #ceph
[9:29] * theanalyst (~abhi@106.51.241.107) has joined #ceph
[9:30] * jtang_ (~jtang@80.111.83.231) has joined #ceph
[9:31] * Nacer (~Nacer@252-87-190-213.intermediasud.com) has joined #ceph
[9:37] * mancdaz (~mancdaz@2a00:1a48:7807:102:94f4:6b56:ff08:886c) has joined #ceph
[9:37] <Clabbe> phew got it reintroduced, pg data is back
[9:37] <Clabbe> now back to the "old" issue
[9:37] <Clabbe> What is causing the slow requests
[9:37] <Clabbe> Any idea on how to debug?
[9:39] * jobewan (~jobewan@75.147.174-93-BusName-smpls.memphis.hfc.comcastbusiness.net) Quit (Quit: Leaving)
[9:40] * jordanP (~jordan@2a04:2500:0:b00:704c:7dbf:7cc2:3ad8) has joined #ceph
[9:46] * atmark (~atmark@50708355.static.ziggozakelijk.nl) has joined #ceph
[9:48] * lcavassa (~lcavassa@89.184.114.246) has joined #ceph
[9:49] <chowmeined> Clabbe, is this rbd?
[9:49] <Clabbe> yes also
[9:49] <Clabbe> running with openstack
[9:49] <atmark> Hello, is there a community version of ceph binary package for RHEL 7?
[9:49] <chowmeined> im using rbd, but not with openstack
[9:50] <chowmeined> so are you seeing high latency on your block devices?
[9:50] <chowmeined> on my tests im seeing latency averaging 75ms/iop
[9:50] * sverrest_ (~sverrest@cm-84.208.166.184.getinternet.no) has joined #ceph
[9:50] <Clabbe> chowmeined: I see it with the syncronization
[9:50] <Clabbe> no clients connected atm
[9:51] <chowmeined> atm I can't get rbd-fuse to work, it deadlocks
[9:51] <chowmeined> what kind of network do you have? do you have a single network or a private cluster network?
[9:52] * sverrest (~sverrest@cm-84.208.166.184.getinternet.no) Quit (Ping timeout: 480 seconds)
[9:52] <Clabbe> chowmeined: private cluster network
[9:52] <chowmeined> atmark, do the centos 6 packages not work on rhel7?
[9:53] <Clabbe> 10GB
[9:53] <atmark> chowmeined: I haven't tried it yet. You think will it work?
[9:53] <Clabbe> a lot of pgs are stuck in peering
[9:54] <chowmeined> atmark, hey look, I found it by changing the URL, i guess its not listed: http://ceph.com/rpm/rhel7/x86_64/
[9:54] * sputnik13 (~sputnik13@wsip-68-105-248-60.sd.sd.cox.net) has joined #ceph
[9:55] <atmark> chowmeined: hey thanks!
[9:56] <chowmeined> Clabbe, did you try these steps already? http://ceph.com/docs/master/rados/troubleshooting/troubleshooting-pg/#placement-group-down-peering-failure
[9:56] <chowmeined> whats the comment for the pg stuck in peering?
[9:58] <Clabbe> hmm having issues querying it :|
[9:58] <Clabbe> Ill try another
[9:59] * kalleh (~kalleh@37-46-175-162.customers.ownit.se) has joined #ceph
[9:59] * jordanP (~jordan@2a04:2500:0:b00:704c:7dbf:7cc2:3ad8) Quit (Ping timeout: 480 seconds)
[9:59] <Clabbe> "Error EINTR: problem getting command descriptions from pg.4.4"
[10:02] <absynth> su -
[10:02] <absynth> err
[10:02] * ghartz (~ghartz@ircad17.u-strasbg.fr) Quit (Ping timeout: 480 seconds)
[10:07] * rotbeard (~redbeard@b2b-94-79-138-170.unitymedia.biz) has joined #ceph
[10:08] * jordanP (~jordan@185.23.92.11) has joined #ceph
[10:13] * ghartz (~ghartz@ircad17.u-strasbg.fr) has joined #ceph
[10:14] * BManojlovic (~steki@cable-94-189-160-74.dynamic.sbb.rs) Quit (Remote host closed the connection)
[10:15] * BManojlovic (~steki@cable-94-189-160-74.dynamic.sbb.rs) has joined #ceph
[10:19] <chowmeined> Clabbe, odd
[10:20] <Clabbe> ceph pg dump is ok
[10:20] <chowmeined> Clabbe, wait, are your monitors healthy?
[10:20] <chowmeined> hm
[10:20] <Clabbe> {"name":"ceph-mon3","rank":2,"state":"peon","election_epoch":862,"quorum":[0,1,2],"outside_quorum":[],"extra_probe_peers":[],"sync_provider":[],"monmap":{"epoch":2,"fsid":"037a2145-39f5-4c97-9f5b-e7cd16dea37b","modified":"2014-04-17 07:50:21.093565","created":"0.000000","mons":[{"rank":0,"name":"ceph-mon1","addr":"192.168.150.1:6789\/0"},{"rank":1,"name":"ceph-mon2","addr":"192.168.150.2:6789\/0"},{"rank":2,"name":"ceph-mon3","addr":"192.168.150
[10:20] <chowmeined> are the OSDs listed as up?
[10:20] <Clabbe> 4up 4in
[10:21] <chowmeined> are the osd volumes mounted on the osd boxes?
[10:21] * atmark (~atmark@50708355.static.ziggozakelijk.nl) Quit (Quit: leaving)
[10:21] <chowmeined> i just had an issue where udev didnt correctly remount them on reboot
[10:21] * xarses (~andreww@c-76-126-112-92.hsd1.ca.comcast.net) Quit (Read error: Operation timed out)
[10:21] <chowmeined> but they still were listed as 'in'
[10:21] <Clabbe> chowmeined: how did that look like?
[10:21] * rendar (~I@host101-178-dynamic.26-79-r.retail.telecomitalia.it) has joined #ceph
[10:22] <chowmeined> osdmap e116: 8 osds: 2 up, 8 in
[10:22] * danieljh_ (~daniel@HSI-KBW-046-005-197-128.hsi8.kabel-badenwuerttemberg.de) has joined #ceph
[10:22] <chowmeined> yet all 8 were not mounted
[10:22] <Clabbe> '/dev/sdb1 1.9T 805G 1.1T 44% /var/lib/ceph/osd/ceph-4
[10:23] <chowmeined> any clues in /var/log/ceph/ceph-osd*.log files?
[10:23] <Clabbe> .fault, initiating reconnect
[10:23] * zhaochao_ (~zhaochao@124.207.139.23) has joined #ceph
[10:23] <Clabbe> .accept connect_seq 138 vs existing 138 state connecting
[10:24] * danieljh (~daniel@0001b4e9.user.oftc.net) Quit (Ping timeout: 480 seconds)
[10:24] * cok (~chk@2a02:2350:18:1012:fda5:7ecc:78a2:65b4) has joined #ceph
[10:24] <Clabbe> chowmeined: slow request 61446.168784 seconds old, received at 2014-07-30 17:08:10.293071: osd_op(client.281093.0:25292 rbd_data.172a46f45e8d1.0000000000000380 [write 4136960~4096] 3.25891ad3 RETRY=7 snapc 26=[26] e3068) v4 currently reached pg
[10:24] * zhaochao (~zhaochao@124.207.139.23) Quit (Read error: Operation timed out)
[10:24] * zhaochao_ is now known as zhaochao
[10:24] <chowmeined> firewalls, can you ping between all OSDs to all OSDs and mons?
[10:24] <chowmeined> are you using jumbo frames? link aggregation? any network errors on ifconfig? any network related errors in dmesg?
[10:25] <Clabbe> jumbo frames
[10:25] <Clabbe> are being used
[10:25] <Clabbe> in sync network
[10:26] <Clabbe> maybe i should change back
[10:26] <chowmeined> can you 'ping -M do -s 8000 <node_ip>' between all nodes?
[10:27] <chowmeined> to ensure no frames are getting dropped because something isnt set for jumbo frames
[10:27] * astellwag is now known as astellwag|afk
[10:27] <chowmeined> if you get 'fragmentation needed' something is dropping jumbo frames
[10:28] * astellwag|afk is now known as astellwag
[10:28] <Clabbe> i dont get frag,
[10:28] <Clabbe> but i cant ping :O
[10:28] <Clabbe> :|
[10:28] <chowmeined> message too long?
[10:28] <chowmeined> are all switches configured for jumbo frames as well?
[10:29] <chowmeined> on all regular interfaces and trunks if in use
[10:29] <Clabbe> its a 10g unmanaged switch
[10:29] <chowmeined> oh
[10:30] <chowmeined> jumbo frames are usually not on by default
[10:30] <chowmeined> if its unmanaged it may not be an option
[10:31] <Clabbe> Ill make a try disabling jumbo frames
[10:32] * kalleeh (~kalleh@37-46-175-162.customers.ownit.se) has joined #ceph
[10:33] * i_m (~ivan.miro@nat-5-carp.hcn-strela.ru) Quit (Quit: Leaving.)
[10:35] * lxo (~aoliva@lxo.user.oftc.net) Quit (Ping timeout: 480 seconds)
[10:35] <Clabbe> chowmeined: maybe not unmanaged :O
[10:35] <chowmeined> model?
[10:36] * kalleh (~kalleh@37-46-175-162.customers.ownit.se) Quit (Read error: Operation timed out)
[10:36] * xarses (~andreww@c-76-126-112-92.hsd1.ca.comcast.net) has joined #ceph
[10:37] <Clabbe> o by the way chowmeined changing mtu size made a difference ':):) ty ty ty
[10:38] <chowmeined> ah, glad to hear it
[10:38] * TMM (~hp@sams-office-nat.tomtomgroup.com) has joined #ceph
[10:38] <chowmeined> well if the switch happens to be managed, you should be able to get it setup for jumbo frames
[10:38] <chowmeined> and then you can get the performance you seek
[10:39] <chowmeined> until you decide you want infiniband :)
[10:40] * darkling (~hrm@2001:630:53:b74:227:10ff:fe02:5c8) has joined #ceph
[10:53] * zhangdongmao (~zhangdong@203.192.156.9) has joined #ceph
[11:02] * madkiss (~madkiss@194.112.182.214) has joined #ceph
[11:03] * i_m (~ivan.miro@gbibp9ph1--blueice2n1.emea.ibm.com) has joined #ceph
[11:06] * yuriw (~Adium@c-76-126-35-111.hsd1.ca.comcast.net) Quit (Ping timeout: 480 seconds)
[11:08] * madkiss1 (~madkiss@213162068015.public.t-mobile.at) has joined #ceph
[11:08] * madkiss1 (~madkiss@213162068015.public.t-mobile.at) Quit ()
[11:09] * jksM (~jks@4810ds1-ns.2.fullrate.dk) has joined #ceph
[11:09] * jksM_ (~jks@4810ds1-ns.2.fullrate.dk) has joined #ceph
[11:10] * madkiss (~madkiss@194.112.182.214) Quit (Ping timeout: 480 seconds)
[11:13] * JC (~JC@AMontpellier-651-1-340-70.w92-145.abo.wanadoo.fr) Quit (Ping timeout: 480 seconds)
[11:17] * Cybertinus (~Cybertinu@cybertinus.customer.cloud.nl) has joined #ceph
[11:19] <Cybertinus> hi
[11:19] <Cybertinus> I'm trying to setup my first Ceph cluster, so I'm messing arround with some VirtualBox VMs, but I get stuck during the install :(
[11:19] <Cybertinus> This is the output I'm getting: https://paste.cybertinus.nl/p/ExYy7o8wCv
[11:20] <Cybertinus> and I don't know why the pgs are stuck, for some reason. When I check with ceph pg dump it states that it is creating all the pgs
[11:20] <Cybertinus> I'm following http://ceph.com/docs/master/start/quick-ceph-deploy/ by the way
[11:21] <classicsnail> Cybertinus: ceph osd get data size returns?
[11:21] <Cybertinus> Atm I'm using 4 VMs for my Ceph cluster: a Ceph admin node which I use to run the Ceph commands on, a monitor node and 2 data nodes with a dedicated disk for the osd
[11:21] <Cybertinus> everything is running CentOS 6.5
[11:22] <classicsnail> default, you have 3 replicas
[11:22] <classicsnail> you have two osds you can write to
[11:23] <Cybertinus> true classicsnail, but in the howto on ceph.com you turn that back to 2, so I should get a healthy state with only 2 OSDs
[11:23] <classicsnail> so you've turned it back to two?
[11:23] <Cybertinus> yes
[11:23] <classicsnail> other thing I've found is restarting the osds can help
[11:23] <classicsnail> have you tried that?
[11:24] * yuriw (~Adium@c-76-126-35-111.hsd1.ca.comcast.net) has joined #ceph
[11:24] * KindOne (kindone@0001a7db.user.oftc.net) has joined #ceph
[11:24] <Cybertinus> yeah, I tried. I can't start the osd deamons, for some reason:
[11:24] <Cybertinus> https://paste.cybertinus.nl/p/z3K3epsnks
[11:25] <Cybertinus> but when you look at the osd tree from my first paste, osd.0 is defined and should be available on ceph-data1...
[11:26] <classicsnail> odd, what's your ceph.conf read?
[11:26] <topro> Cybertinus: your first post states that both your osds are down
[11:26] * amospalla (~amospalla@0001a39c.user.oftc.net) Quit (Quit: WeeChat 0.4.3)
[11:26] <Cybertinus> topro: https://paste.cybertinus.nl/p/sA4YSADDUO
[11:27] <topro> dis you deploy using ceph-deploy?
[11:27] <Cybertinus> yes
[11:27] <topro> s/dis/did/
[11:27] <kraken> topro meant to say: did you deploy using ceph-deploy?
[11:28] <topro> Cybertinus: ^^
[11:29] * Midnightmyth (~quassel@93-167-84-102-static.dk.customer.tdc.net) has joined #ceph
[11:29] <Cybertinus> topro: yes, I deployed via ceph-deploy
[11:30] <topro> ok, then I'm of no further help as I still stick to manual configuration. all I can say is that 'ceph osd tree' should not show 'down' as status to your osds.
[11:31] <classicsnail> what does the ceph osd log say, from /var/log/ceph?
[11:32] <Cybertinus> classicsnail: That I created them yesterday. But nothing is mentioned about starting them today
[11:33] <Cybertinus> classicsnail: https://paste.cybertinus.nl/p/XEjD92n48i <-- the log
[11:35] <classicsnail> if you mount the osd, is there anything in the data directory?
[11:35] <classicsnail> the changing fsids and the missing keyrings are interesting
[11:38] <Cybertinus> well, I think I can explain that
[11:38] <Cybertinus> first I was creating the OSD with prepare and activate. This didn't work. Then I switched over to a new disk and used create. And that worked
[11:39] <Cybertinus> or at least: that command didn't return an error
[11:39] * amospalla (~amospalla@0001a39c.user.oftc.net) has joined #ceph
[11:43] <classicsnail> does the keyring on the osd data directory match the keyring present when you look at the same osd in the tput form ceph auth list?
[11:45] * lucas1 (~Thunderbi@218.76.25.66) has joined #ceph
[11:47] <Cybertinus> classicsnail: https://paste.cybertinus.nl/p/kx5dXaUBva <-- I would say they are the same on both admin and data1
[11:49] * AfC (~andrew@2001:44b8:31cb:d400:6e88:14ff:fe33:2a9c) has joined #ceph
[11:52] * dis is now known as Guest4383
[11:52] * dis (~dis@109.110.67.1) has joined #ceph
[11:53] <Cybertinus> classicsnail: https://paste.cybertinus.nl/p/FNWlIqBMgu <-- that is different...
[11:54] * Guest4383 (~dis@109.110.67.20) Quit (Ping timeout: 480 seconds)
[12:04] <trond> Crushtool segfaults when we check for bad mappings. And uploading an unmodified crushmap to the cluster crashes the monitor. This is ceph 0.80.5
[12:06] * cury (~cury@123.115.79.17) Quit (Ping timeout: 480 seconds)
[12:12] * coreping (~xuser@hugin.coreping.org) Quit (Quit: WeeChat 0.3.7)
[12:15] * sputnik13 (~sputnik13@wsip-68-105-248-60.sd.sd.cox.net) Quit (Quit: My MacBook has gone to sleep. ZZZzzz???)
[12:16] * marrusl (~mark@209-150-43-182.c3-0.wsd-ubr2.qens-wsd.ny.cable.rcn.com) has joined #ceph
[12:16] * kalleeh (~kalleh@37-46-175-162.customers.ownit.se) Quit (Ping timeout: 480 seconds)
[12:19] * AfC (~andrew@2001:44b8:31cb:d400:6e88:14ff:fe33:2a9c) Quit (Quit: Leaving.)
[12:21] <Cybertinus> classicsnail: https://paste.cybertinus.nl/p/VfVpoxDxsS <-- matches the output of ceph auth list
[12:26] * narurien (~henning@trh.betradar.com) has joined #ceph
[12:26] * steki (~steki@91.195.39.5) Quit (Quit: Ja odoh a vi sta 'ocete...)
[12:27] * steki (~steki@91.195.39.5) has joined #ceph
[12:37] * shang (~ShangWu@175.41.48.77) Quit (Ping timeout: 480 seconds)
[12:45] * cok (~chk@2a02:2350:18:1012:fda5:7ecc:78a2:65b4) Quit (Quit: Leaving.)
[12:48] * lucas1 (~Thunderbi@218.76.25.66) Quit (Quit: lucas1)
[12:50] * steki (~steki@91.195.39.5) Quit (Quit: Ja odoh a vi sta 'ocete...)
[12:56] * diegows (~diegows@190.190.5.238) has joined #ceph
[12:57] * steki (~steki@91.195.39.5) has joined #ceph
[13:06] * Vacum_ (~vovo@i59F7AAD5.versanet.de) Quit (Remote host closed the connection)
[13:12] * Vacum (~vovo@i59F7AAD5.versanet.de) has joined #ceph
[13:15] * dmsimard_away is now known as dmsimard
[13:25] * zhaochao (~zhaochao@124.207.139.23) has left #ceph
[13:32] * johnfoo (~johnfoo@ip-133.net-89-3-152.rev.numericable.fr) Quit (Quit: leaving)
[13:32] * kalleh (~kalleh@37-46-175-162.customers.ownit.se) has joined #ceph
[13:37] <joao> trond, we are aware that may happen when an improperly built crushmap is passed
[13:38] <joao> see issue 6800
[13:38] <kraken> joao might be talking about http://tracker.ceph.com/issues/6800 [rbd/qemu-iotests.sh Failing on Arm.]
[13:38] <joao> oops
[13:38] <joao> issue 8600
[13:38] <kraken> joao might be talking about http://tracker.ceph.com/issues/8600 [MON crashes on new crushmap injection]
[13:38] <joao> there we go
[13:38] * rotbeard (~redbeard@b2b-94-79-138-170.unitymedia.biz) Quit (Quit: Leaving)
[13:41] * JC (~JC@AMontpellier-651-1-340-70.w92-145.abo.wanadoo.fr) has joined #ceph
[13:44] * DV (~veillard@2001:41d0:1:d478::1) Quit (Ping timeout: 480 seconds)
[13:46] <djh-work> Huh, it seems like using ceph-deploy install on localhost as one of the hosts still requires an ssh login to the local machine? I'm user root, but ceph-deploy wants to login at root@localhost, thus requiring the root password.
[13:46] * Jakey (uid1475@id-1475.uxbridge.irccloud.com) Quit (Quit: Connection closed for inactivity)
[13:46] <djh-work> Why does ceph-deploy do it this way (ssh to localhost) instead of just installing the packages?
[13:49] <Gugge-47527> localhost is just a hostname like any other
[13:49] * kapil (~ksharma@2620:113:80c0:5::2222) Quit (Quit: Leaving)
[13:50] * kanagaraj (~kanagaraj@61.3.145.128) Quit (Quit: Leaving)
[13:51] * kanagaraj (~kanagaraj@61.3.145.128) has joined #ceph
[13:51] * sz0 (~sz0@94.55.197.185) Quit ()
[13:53] <djh-work> Gugge-47527: yes, and I got it to do what I want after all, but I thought ceph-deploy recognizes the opportunity of taking a shortcut here..
[13:53] * kapil (~ksharma@2620:113:80c0:5::2222) has joined #ceph
[13:54] * cok (~chk@2a02:2350:18:1012:84f2:ed83:76a9:f6a3) has joined #ceph
[13:57] <djh-work> Reading this issue: http://tracker.ceph.com/issues/4367 it should take the shortcut, or am I wrong here?
[14:02] <Gugge-47527> yes, for the local hostname
[14:02] <Gugge-47527> not for "localhost"
[14:02] <Gugge-47527> as far as i can tell from the tests i just did :)
[14:05] * darkling (~hrm@2001:630:53:b74:227:10ff:fe02:5c8) Quit (Ping timeout: 480 seconds)
[14:07] * Sysadmin88 (~IceChat77@054287fa.skybroadband.com) Quit (Read error: Connection reset by peer)
[14:13] * DV (~veillard@veillard.com) has joined #ceph
[14:14] * sz0 (~sz0@94.55.197.185) has joined #ceph
[14:20] * ghost1 (~pablodelg@38.104.90.194) has joined #ceph
[14:30] * hyperbaba (~hyperbaba@private.neobee.net) Quit (Ping timeout: 480 seconds)
[14:31] <alfredodeza> djh-work: that is not supported, you should not use localhost
[14:31] * sz0 (~sz0@94.55.197.185) Quit (Quit: My iMac has gone to sleep. ZZZzzz???)
[14:32] * ghartz (~ghartz@ircad17.u-strasbg.fr) Quit (Ping timeout: 480 seconds)
[14:32] <alfredodeza> djh-work: something is not right from your description, ceph-deploy will not ssh to the current host if it doesn't need to
[14:39] * JC (~JC@AMontpellier-651-1-340-70.w92-145.abo.wanadoo.fr) Quit (Quit: Leaving.)
[14:40] * int0x21 (~int0x21@mail.planit.se) has joined #ceph
[14:42] <int0x21> Hello. I had a little question. I been going to and from with a storage solution and after a couple issues with freebsd and its ctld and solaris lack off vaai im considering going linux and ceph instead
[14:43] <int0x21> the server i have for this project is a dual cpu 3x lsi 9300 hba directly attached to 24 sata ssd drives
[14:43] * ghartz (~ghartz@ircad17.u-strasbg.fr) has joined #ceph
[14:43] <int0x21> Is this viable for ceph or do i need to zfs it first and do ceph ontop of zfs ?
[14:44] <int0x21> My end result is to use this as a vmware datastore for ouer vdi
[14:44] * wschulze (~wschulze@cpe-69-206-251-158.nyc.res.rr.com) has joined #ceph
[14:44] * markbby (~Adium@168.94.245.4) has joined #ceph
[14:46] * lofejndif (~lsqavnbok@176.10.100.229) has joined #ceph
[14:50] * Kdecherf (~kdecherf@shaolan.kdecherf.com) Quit (Ping timeout: 480 seconds)
[14:52] * ganders (~root@200-127-158-54.net.prima.net.ar) has joined #ceph
[14:54] * lxo (~aoliva@lxo.user.oftc.net) has joined #ceph
[15:02] * lofejndif (~lsqavnbok@7DKAACO5L.tor-irc.dnsbl.oftc.net) Quit (Quit: gone)
[15:04] * kalleeh (~kalleh@37-46-175-162.customers.ownit.se) has joined #ceph
[15:05] * rwheeler (~rwheeler@nat-pool-bos-t.redhat.com) has joined #ceph
[15:06] * bkopilov (~bkopilov@nat-pool-tlv-t.redhat.com) Quit (Ping timeout: 480 seconds)
[15:10] * kalleh (~kalleh@37-46-175-162.customers.ownit.se) Quit (Ping timeout: 480 seconds)
[15:17] * amospalla (~amospalla@0001a39c.user.oftc.net) Quit (Quit: WeeChat 0.4.3)
[15:18] * amospalla (~amospalla@0001a39c.user.oftc.net) has joined #ceph
[15:19] * kalleh (~kalleh@37-46-175-162.customers.ownit.se) has joined #ceph
[15:22] * darkling (~hrm@2001:630:53:b74:227:10ff:fe02:5c8) has joined #ceph
[15:23] * gregmark (~Adium@cet-nat-254.ndceast.pa.bo.comcast.net) has joined #ceph
[15:24] * kalleeh (~kalleh@37-46-175-162.customers.ownit.se) Quit (Ping timeout: 480 seconds)
[15:30] * i_m (~ivan.miro@gbibp9ph1--blueice2n1.emea.ibm.com) Quit (Read error: Connection reset by peer)
[15:34] * jksM (~jks@4810ds1-ns.2.fullrate.dk) Quit (Remote host closed the connection)
[15:34] * jksM_ (~jks@4810ds1-ns.2.fullrate.dk) Quit (Remote host closed the connection)
[15:38] * brad_mssw (~brad@shop.monetra.com) has joined #ceph
[15:40] * i_m (~ivan.miro@gbibp9ph1--blueice4n1.emea.ibm.com) has joined #ceph
[15:41] * kanagaraj (~kanagaraj@61.3.145.128) Quit (Read error: Operation timed out)
[15:41] * i_m (~ivan.miro@gbibp9ph1--blueice4n1.emea.ibm.com) Quit ()
[15:41] * i_m (~ivan.miro@gbibp9ph1--blueice4n1.emea.ibm.com) has joined #ceph
[15:42] * cury (~cury@114.248.224.218) has joined #ceph
[15:42] * lupu (~lupu@86.107.101.214) has joined #ceph
[15:51] * JC (~JC@AMontpellier-651-1-317-192.w92-133.abo.wanadoo.fr) has joined #ceph
[15:51] * michalefty (~micha@p20030071CE596953891354B9A104B6D1.dip0.t-ipconnect.de) has joined #ceph
[15:54] * madkiss (~madkiss@212095007004.public.telering.at) has joined #ceph
[15:55] * DV (~veillard@veillard.com) Quit (Ping timeout: 480 seconds)
[16:02] * rmoe (~quassel@173-228-89-134.dsl.static.sonic.net) Quit (Remote host closed the connection)
[16:06] * JC1 (~JC@AMontpellier-651-1-317-192.w92-133.abo.wanadoo.fr) has joined #ceph
[16:06] * Sysadmin88 (~IceChat77@054287fa.skybroadband.com) has joined #ceph
[16:08] * JC (~JC@AMontpellier-651-1-317-192.w92-133.abo.wanadoo.fr) Quit (Ping timeout: 480 seconds)
[16:08] * scuttle|afk is now known as scuttlemonkey
[16:09] * PerlStalker (~PerlStalk@2620:d3:8000:192::70) has joined #ceph
[16:11] * KevinPerks (~Adium@cpe-098-025-128-231.sc.res.rr.com) has joined #ceph
[16:12] * brunoleon_ (~quassel@ARennes-658-1-86-216.w90-32.abo.wanadoo.fr) has joined #ceph
[16:13] * DV (~veillard@2001:41d0:1:d478::1) has joined #ceph
[16:14] * KevinPerks1 (~Adium@cpe-098-025-128-231.sc.res.rr.com) has joined #ceph
[16:14] * KevinPerks (~Adium@cpe-098-025-128-231.sc.res.rr.com) Quit (Read error: Connection reset by peer)
[16:16] * brunoleon (~quassel@ARennes-658-1-175-12.w92-139.abo.wanadoo.fr) Quit (Ping timeout: 480 seconds)
[16:18] * hasues (~hasues@kwfw01.scrippsnetworksinteractive.com) has joined #ceph
[16:22] * amospalla (~amospalla@0001a39c.user.oftc.net) Quit (Quit: WeeChat 0.4.3)
[16:23] * amospalla (~amospalla@0001a39c.user.oftc.net) has joined #ceph
[16:32] * amospalla (~amospalla@0001a39c.user.oftc.net) Quit (Quit: WeeChat 0.4.3)
[16:33] * amospalla (~amospalla@0001a39c.user.oftc.net) has joined #ceph
[16:35] * allsystemsarego (~allsystem@79.115.170.35) has joined #ceph
[16:42] * lalatenduM (~lalatendu@122.171.95.253) Quit (Read error: Connection reset by peer)
[16:47] * bkopilov (~bkopilov@213.57.17.135) has joined #ceph
[16:48] * sz0 (~sz0@94.55.197.185) has joined #ceph
[16:51] * rdas (~rdas@121.244.87.115) Quit (Quit: Leaving)
[16:51] <burley> anyone hitting RCU stall warnings on 0.80.1 on ubuntu 14.04 clients?
[16:53] <jiffe> so I created a second mds node and it seems to work in standby, I shut off the primary mds and it switched to the standby
[16:53] * thomnico (~thomnico@8.25.222.10) has joined #ceph
[16:57] * angdraug (~angdraug@c-67-169-181-128.hsd1.ca.comcast.net) has joined #ceph
[16:58] * cok (~chk@2a02:2350:18:1012:84f2:ed83:76a9:f6a3) has left #ceph
[16:58] <jiffe> there any guidelines to running mds nodes in active/standby?
[17:02] * int0x21 (~int0x21@mail.planit.se) Quit (Ping timeout: 480 seconds)
[17:04] * madkiss (~madkiss@212095007004.public.telering.at) Quit (Ping timeout: 480 seconds)
[17:05] * KaZeR (~kazer@64.201.252.132) has joined #ceph
[17:05] * madkiss (~madkiss@178.188.60.118) has joined #ceph
[17:06] * astellwag is now known as astellwag|afk
[17:07] * astellwag|afk is now known as astellwag
[17:08] * baylight (~tbayly@204.15.85.169) has joined #ceph
[17:08] * kevinc (~kevinc__@client65-44.sdsc.edu) has joined #ceph
[17:10] * KevinPerks1 (~Adium@cpe-098-025-128-231.sc.res.rr.com) Quit (Quit: Leaving.)
[17:11] * bandrus (~oddo@216.57.72.205) has joined #ceph
[17:11] * astellwag is now known as astellwag|afk
[17:13] <trond> joao: Thanks, we figured as much, when we started working through the crushmap. :)
[17:13] <yuriw> loicd: ping are you there?
[17:13] * madkiss (~madkiss@178.188.60.118) Quit (Ping timeout: 480 seconds)
[17:13] <loicd> yuriw: yes
[17:14] <yuriw> oh hello! back already?
[17:14] <joao> trond, can you please share what your issue was?
[17:14] * michalefty (~micha@p20030071CE596953891354B9A104B6D1.dip0.t-ipconnect.de) Quit (Quit: Leaving.)
[17:14] <joao> just to make sure it aligns with what we've seen
[17:14] <loicd> yuriw: almost ;-) What can I do for you ?
[17:15] <yuriw> loicd: I am looking at ## 8736, 8737, 8740 and results I got today teuthology-2014-07-30_13:09:01-upgrade:firefly-x:stress-split-next---basic-plana
[17:15] <yuriw> loicd: trying to understand if they are still present?
[17:16] * loicd looking
[17:16] <yuriw> *mercy*
[17:17] <loicd> http://pulpito.ceph.com/teuthology-2014-07-30_13:09:01-upgrade:firefly-x:stress-split-next---basic-plana/
[17:17] <yuriw> yes
[17:18] <loicd> there are much less tests than when I ran it earlier this months, why is that ? http://pulpito.ceph.com/?suite=upgrade%3Afirefly-x%3Astress-split
[17:18] <loicd> yuriw: ^
[17:18] * loicd looking at the latest changes
[17:18] <yuriw> i ran on plana
[17:18] <yuriw> no changes in tests
[17:19] * steki (~steki@91.195.39.5) Quit (Ping timeout: 480 seconds)
[17:19] <loicd> that's odd
[17:19] <narurien> joao: I can answer for trond here. Basically we saw the same symptoms as in #8600 (choose vs. chooseleaf). In addition it crashes if you have a rule like "choose firstn 2 type rack" but min_size=1
[17:19] <yuriw> what?
[17:21] <joao> narurien, mind adding that to the ticket?
[17:21] <narurien> can do
[17:21] <joao> ty
[17:21] <loicd> yuriw: I find strange that http://pulpito.ceph.com/teuthology-2014-07-30_13:09:01-upgrade:firefly-x:stress-split-next---basic-plana/ nas a total of 20 tests while http://pulpito.ceph.com/loic-2014-07-02_23:05:05-upgrade:firefly-x:stress-split-firefly-testing-basic-vps/ has a total of 60 for the exact same suite. Could it be that pulpito only got some of them and missed the others ?
[17:22] <yuriw> I think it can be because when run on plana machines - distros are not called
[17:22] <yuriw> i can re-run on vps
[17:23] * i_m (~ivan.miro@gbibp9ph1--blueice4n1.emea.ibm.com) Quit (Ping timeout: 480 seconds)
[17:23] <loicd> oh
[17:23] <loicd> that explains it :-)
[17:24] <loicd> indeed just ubuntu
[17:24] <loicd> yuriw: is there a way to run a suite and only focus on one distro while using vps ?
[17:26] <loicd> yuriw: http://pulpito.ceph.com/loic-2014-07-02_23:05:05-upgrade:firefly-x:stress-split-firefly-testing-basic-vps/ shows only two fails for ubuntu. Now looking if the fails are the same.
[17:27] <yuriw> loicd: we have no such option in teuthology-suite yet (see http://tracker.ceph.com/issues/8954)
[17:27] <yuriw> but you can run with ???suite-dir option and modify test code a little to point/use only one distro
[17:28] <yuriw> loicd: no but yes :)
[17:28] <loicd> :-)
[17:28] <loicd> thanks for the tip
[17:28] <yuriw> sure
[17:28] * kalleh (~kalleh@37-46-175-162.customers.ownit.se) Quit (Ping timeout: 480 seconds)
[17:29] <loicd> comparing takes a little time, internet is not fast in Kathmandu ;-)
[17:30] <yuriw> i feel guilty, sorry for pining you on vacation
[17:30] <yuriw> you know let me re-run on vps and we can check back then, a?
[17:30] <yuriw> and I can ask about rgw failures
[17:30] <loicd> hum
[17:31] * jordanP (~jordan@185.23.92.11) Quit (Remote host closed the connection)
[17:31] <loicd> the errors seem different
[17:31] <loicd> earlier this months I had one fail and one dead, now you have two fail with the same error "Command failed on plana61 with status 1: "SWIFT_TEST_CONFIG_FILE=/home/ubuntu/cephtest/archive/testswift.client.0.conf /home/ubuntu/cephtest/swift/virtualenv/bin/nosetests -w /home/ubuntu/cephtest/swift/test/functional -v -a '!fails_on_rgw'" "
[17:32] <loicd> yuriw: it's actually nice to dive into this now. It feels a little strange after a few weeks off ;-)
[17:33] <yuriw> loic: after my France trip I could not remeber half of what I did before :)
[17:33] <loicd> yuriw: at first glance I don't see anything that would indicate problems are resolved. Sorry for the disapointment ;-)
[17:33] <loicd> ahaha
[17:33] <yuriw> no prob
[17:33] <yuriw> when are you back?
[17:33] <loicd> next tuesday
[17:33] * dgbaley27 (~matt@c-98-245-167-2.hsd1.co.comcast.net) has joined #ceph
[17:33] <loicd> august 3rd
[17:33] <yuriw> cool, thx
[17:33] <yuriw> I will ping you with new results then
[17:33] <loicd> yw :-)
[17:34] <yuriw> have fun !
[17:34] <yuriw> :)
[17:34] * brunoleon_ (~quassel@ARennes-658-1-86-216.w90-32.abo.wanadoo.fr) Quit (Read error: No route to host)
[17:37] * scuttlemonkey is now known as scuttle|afk
[17:37] * danieagle (~Daniel@179.184.165.184.static.gvt.net.br) has joined #ceph
[17:38] * v2_ (~venky@ov42.x.rootbsd.net) Quit (Quit: leaving)
[17:38] * scuttle|afk is now known as scuttlemonkey
[17:38] * Tamil1 (~Adium@cpe-108-184-74-11.socal.res.rr.com) has joined #ceph
[17:46] * i_m (~ivan.miro@nat-5-carp.hcn-strela.ru) has joined #ceph
[17:46] * adamcrume (~quassel@50.247.81.99) has joined #ceph
[17:50] * rmoe (~quassel@173-228-89-134.dsl.static.sonic.net) has joined #ceph
[17:52] * jordanP (~jordan@185.23.92.11) has joined #ceph
[17:53] * KevinPerks (~Adium@cpe-098-025-128-231.sc.res.rr.com) has joined #ceph
[17:56] * bandrus1 (~Adium@216.57.72.205) has joined #ceph
[17:57] * ghartz (~ghartz@ircad17.u-strasbg.fr) Quit (Ping timeout: 480 seconds)
[17:58] * xarses (~andreww@c-76-126-112-92.hsd1.ca.comcast.net) Quit (Read error: Operation timed out)
[17:59] <yuriw> loicd: running 60 now - http://pulpito.front.sepia.ceph.com/teuthology-2014-07-31_08:55:02-upgrade:firefly-x:stress-split-next---basic-vps/
[17:59] * sz0 (~sz0@94.55.197.185) Quit (Quit: My iMac has gone to sleep. ZZZzzz???)
[17:59] * dignus (~jkooijman@t-x.dignus.nl) Quit (Read error: Operation timed out)
[17:59] * bandrus1 (~Adium@216.57.72.205) Quit ()
[18:00] * wschulze (~wschulze@cpe-69-206-251-158.nyc.res.rr.com) Quit (Quit: Leaving.)
[18:01] * KevinPerks (~Adium@cpe-098-025-128-231.sc.res.rr.com) Quit (Quit: Leaving.)
[18:02] * jordanP (~jordan@185.23.92.11) Quit (Quit: Leaving)
[18:08] * theanalyst (~abhi@106.51.241.107) Quit (Ping timeout: 480 seconds)
[18:08] * ghartz (~ghartz@ircad17.u-strasbg.fr) has joined #ceph
[18:09] * b0e (~aledermue@juniper1.netways.de) Quit (Quit: Leaving.)
[18:09] * wschulze (~wschulze@cpe-69-206-251-158.nyc.res.rr.com) has joined #ceph
[18:12] * Tamil1 (~Adium@cpe-108-184-74-11.socal.res.rr.com) Quit (Quit: Leaving.)
[18:12] * thomnico (~thomnico@8.25.222.10) Quit (Read error: Operation timed out)
[18:13] * fghaas (~florian@88.128.80.42) has joined #ceph
[18:15] * rweeks (~rweeks@pat.hitachigst.com) has joined #ceph
[18:17] * thomnico (~thomnico@sccc-66-78-236-243.smartcity.com) has joined #ceph
[18:22] * thb (~me@0001bd58.user.oftc.net) Quit (Ping timeout: 480 seconds)
[18:24] * darkling (~hrm@2001:630:53:b74:227:10ff:fe02:5c8) Quit (Ping timeout: 480 seconds)
[18:29] * shang (~ShangWu@220-135-203-169.HINET-IP.hinet.net) has joined #ceph
[18:29] * TMM (~hp@sams-office-nat.tomtomgroup.com) Quit (Quit: Ex-Chat)
[18:32] * vbellur (~vijay@122.172.106.250) Quit (Quit: Leaving.)
[18:33] * xarses (~andreww@12.164.168.117) has joined #ceph
[18:37] <KaZeR> some of my OSDs are down, and i've found messages like "claims to be 192.168.20.39:6801/299018791 not 192.168.20.39:6801/19018791 - wrong node!"
[18:37] <KaZeR> what means the value after the / ?
[18:38] * sigsegv (~sigsegv@188.25.123.201) has joined #ceph
[18:39] * sigsegv (~sigsegv@188.25.123.201) has left #ceph
[18:39] * dignus (~jkooijman@t-x.dignus.nl) has joined #ceph
[18:45] * lxo (~aoliva@lxo.user.oftc.net) Quit (Remote host closed the connection)
[18:46] * lxo (~aoliva@lxo.user.oftc.net) has joined #ceph
[18:46] * Tamil1 (~Adium@cpe-108-184-74-11.socal.res.rr.com) has joined #ceph
[18:47] * madkiss (~madkiss@178.188.60.118) has joined #ceph
[18:52] * reed (~reed@75-101-54-131.dsl.static.sonic.net) has joined #ceph
[18:52] * jeff-YF (~jeffyf@67.23.117.122) has joined #ceph
[18:54] * sjusthm (~sam@24-205-54-233.dhcp.gldl.ca.charter.com) has joined #ceph
[18:55] * terje (~joey@63.228.91.225) Quit (Ping timeout: 480 seconds)
[18:55] * sputnik13 (~sputnik13@wsip-68-105-248-60.sd.sd.cox.net) has joined #ceph
[18:56] * lcavassa (~lcavassa@89.184.114.246) Quit (Quit: Leaving)
[19:01] * zerick (~eocrospom@190.187.21.53) has joined #ceph
[19:02] * adamcrume (~quassel@50.247.81.99) Quit (Remote host closed the connection)
[19:02] * ghost1 (~pablodelg@38.104.90.194) Quit (Quit: ghost1)
[19:03] * lcavassa (~lcavassa@89.184.114.246) has joined #ceph
[19:04] * ghost1 (~pablodelg@38.104.90.194) has joined #ceph
[19:08] * rmoe (~quassel@173-228-89-134.dsl.static.sonic.net) Quit (Read error: Operation timed out)
[19:11] * lcavassa (~lcavassa@89.184.114.246) Quit (Quit: Leaving)
[19:14] * rmoe (~quassel@12.164.168.117) has joined #ceph
[19:20] * JuanEpstein (~rweeks@pat.hitachigst.com) has joined #ceph
[19:20] * JuanEpstein (~rweeks@pat.hitachigst.com) Quit ()
[19:25] * adamcrume (~quassel@c-71-204-162-10.hsd1.ca.comcast.net) has joined #ceph
[19:27] * adamcrume_ (~quassel@c-71-204-162-10.hsd1.ca.comcast.net) has joined #ceph
[19:27] * adamcrume (~quassel@c-71-204-162-10.hsd1.ca.comcast.net) Quit (Remote host closed the connection)
[19:27] * adamcrume_ (~quassel@c-71-204-162-10.hsd1.ca.comcast.net) Quit (Remote host closed the connection)
[19:28] * Cube (~Cube@12.248.40.138) has joined #ceph
[19:28] * adamcrume (~quassel@c-71-204-162-10.hsd1.ca.comcast.net) has joined #ceph
[19:28] * madkiss (~madkiss@178.188.60.118) Quit (Quit: Leaving.)
[19:31] * dgbaley27 (~matt@c-98-245-167-2.hsd1.co.comcast.net) Quit (Remote host closed the connection)
[19:37] * kevinc (~kevinc__@client65-44.sdsc.edu) Quit (Quit: This computer has gone to sleep)
[19:43] * LeaChim (~LeaChim@host86-161-89-237.range86-161.btcentralplus.com) has joined #ceph
[19:44] * reistin (~reistin@5.141.195.170) has joined #ceph
[19:44] <reistin> hi all!
[19:44] * thomnico (~thomnico@sccc-66-78-236-243.smartcity.com) Quit (Ping timeout: 480 seconds)
[19:44] * Nacer (~Nacer@252-87-190-213.intermediasud.com) Quit (Ping timeout: 480 seconds)
[19:45] * kevinc (~kevinc__@client65-44.sdsc.edu) has joined #ceph
[19:46] <reistin> Why on some clusters command 'ceph osd create host:disk' do all the work(prepare disk, add it to crush map) and on some it only prepare disk, and you have to activate, mount it manually?
[19:58] * angdraug (~angdraug@c-67-169-181-128.hsd1.ca.comcast.net) Quit (Quit: Leaving)
[20:05] * garphy`aw is now known as garphy
[20:07] * thb (~me@port-53857.pppoe.wtnet.de) has joined #ceph
[20:07] * shang (~ShangWu@220-135-203-169.HINET-IP.hinet.net) Quit (Ping timeout: 480 seconds)
[20:08] * thomnico (~thomnico@sccc-66-78-236-243.smartcity.com) has joined #ceph
[20:08] * lupu (~lupu@86.107.101.214) Quit (Read error: Operation timed out)
[20:09] * terje__ (~joey@63.228.91.225) has joined #ceph
[20:12] * Nacer (~Nacer@2001:41d0:fe82:7200:31ff:86e4:4722:8dc6) has joined #ceph
[20:14] * theanalyst (~abhi@106.51.233.171) has joined #ceph
[20:18] <bens> I dunno how to find out what the actifvity of rados looks like
[20:19] <bens> is there anyway to log changes to the pool
[20:19] <bens> at an object level
[20:21] * fghaas (~florian@88.128.80.42) Quit (Quit: Leaving.)
[20:25] * garphy is now known as garphy`aw
[20:27] * angdraug (~angdraug@12.164.168.117) has joined #ceph
[20:27] * garphy`aw is now known as garphy
[20:33] * sputnik13 (~sputnik13@wsip-68-105-248-60.sd.sd.cox.net) Quit (Quit: Textual IRC Client: www.textualapp.com)
[20:34] * kevinc (~kevinc__@client65-44.sdsc.edu) Quit (Quit: This computer has gone to sleep)
[20:35] * rwheeler (~rwheeler@nat-pool-bos-t.redhat.com) Quit (Quit: Leaving)
[20:38] * sputnik13 (~sputnik13@wsip-68-105-248-60.sd.sd.cox.net) has joined #ceph
[20:40] * sz0 (~sz0@94.55.197.185) has joined #ceph
[20:42] * ikrstic (~ikrstic@109-93-184-39.dynamic.isp.telekom.rs) has joined #ceph
[20:49] * rendar (~I@host101-178-dynamic.26-79-r.retail.telecomitalia.it) Quit (Ping timeout: 480 seconds)
[20:51] * rendar (~I@host101-178-dynamic.26-79-r.retail.telecomitalia.it) has joined #ceph
[20:59] * sputnik13 (~sputnik13@wsip-68-105-248-60.sd.sd.cox.net) Quit (Quit: My MacBook has gone to sleep. ZZZzzz???)
[21:03] * garphy is now known as garphy`aw
[21:04] * cury (~cury@114.248.224.218) Quit (Ping timeout: 480 seconds)
[21:05] * hijacker (~hijacker@bgva.sonic.taxback.ess.ie) Quit (Ping timeout: 480 seconds)
[21:05] * hijacker (~hijacker@213.91.163.5) has joined #ceph
[21:07] * hasues (~hasues@kwfw01.scrippsnetworksinteractive.com) Quit (Quit: Leaving.)
[21:07] * hasues (~hasues@kwfw01.scrippsnetworksinteractive.com) has joined #ceph
[21:09] * reistin (~reistin@5.141.195.170) Quit ()
[21:16] * Midnightmyth (~quassel@93-167-84-102-static.dk.customer.tdc.net) Quit (Ping timeout: 480 seconds)
[21:20] * theanalyst (~abhi@106.51.233.171) Quit (Remote host closed the connection)
[21:31] * dis is now known as Guest4415
[21:31] * dis (~dis@109.110.67.234) has joined #ceph
[21:33] * Guest4415 (~dis@109.110.67.1) Quit (Ping timeout: 480 seconds)
[21:35] * ghartz (~ghartz@ircad17.u-strasbg.fr) Quit (Ping timeout: 480 seconds)
[21:36] * hasues (~hasues@kwfw01.scrippsnetworksinteractive.com) Quit (Quit: Leaving.)
[21:36] * kevinc (~kevinc__@client65-44.sdsc.edu) has joined #ceph
[21:36] * thomnico (~thomnico@sccc-66-78-236-243.smartcity.com) Quit (Ping timeout: 480 seconds)
[21:37] * hasues (~hasues@kwfw01.scrippsnetworksinteractive.com) has joined #ceph
[21:39] * jksM (~jks@3e6b5724.rev.stofanet.dk) has joined #ceph
[21:40] <chowmeined> What does it often mean when 'rbd ls' hangs? I can't map any of the devices anymore but the cluster health is okay
[21:42] * cury (~cury@114.248.224.218) has joined #ceph
[21:45] * ghartz (~ghartz@ircad17.u-strasbg.fr) has joined #ceph
[21:47] * ircolle (~Adium@2601:1:a580:145a:fd00:bf10:f53f:1a5d) Quit (Read error: Connection reset by peer)
[21:47] * ircolle (~Adium@c-67-172-132-222.hsd1.co.comcast.net) has joined #ceph
[21:57] * tcos (~will@s11-241.rb.gh.centurytel.net) has joined #ceph
[21:58] * kevinc (~kevinc__@client65-44.sdsc.edu) Quit (Quit: This computer has gone to sleep)
[21:58] * cok1 (~chk@46.30.211.29) has joined #ceph
[22:02] * ghost1 (~pablodelg@38.104.90.194) Quit (Quit: ghost1)
[22:07] * ircolle is now known as ircolle-afk
[22:11] * sz0 (~sz0@94.55.197.185) Quit (Quit: My iMac has gone to sleep. ZZZzzz???)
[22:14] * redcavalier (~redcavali@office-mtl1-nat-146-218-70-69.gtcomm.net) has joined #ceph
[22:16] * andreask (~andreask@h081217017238.dyn.cm.kabsi.at) has joined #ceph
[22:16] * ChanServ sets mode +v andreask
[22:17] * ganders (~root@200-127-158-54.net.prima.net.ar) Quit (Quit: WeeChat 0.4.1)
[22:17] <redcavalier> Hi, got a few hardware requirements questions/suggestions requests. I've been looking at inktank's Hardware configuration guide and on their recomended setups, they tend to put several OSDs on the same physical nodes. Isn't that a terrible idea?
[22:18] <Sysadmin88> no, Ceph will work with what you give it
[22:19] * sputnik13 (~sputnik13@207.8.121.241) has joined #ceph
[22:19] <redcavalier> Let me just finish explaining my reasoning
[22:19] <gchristensen> redcavalier: the recommendation is one OSD per disk
[22:19] <Sysadmin88> the only problem with lots of OSDs on a single node is there is more failure when that node goes down... but if you have lots of other big nodes it doesnt matter so much
[22:20] <redcavalier> Let's say I have 3 physical nodes, 10 OSDs each with an object replication of 3. Now, one physical node goes down. Is it possible that the 3 copies of a single object end up all among those 10 OSDs that just wnet down?
[22:20] <redcavalier> or does ceph account for that?
[22:21] <Sysadmin88> then manybe you need more nodes for that setup
[22:21] <Sysadmin88> iirc ceph will stop before compromising data
[22:21] * andreask1 (~andreask@h081217017238.dyn.cm.kabsi.at) has joined #ceph
[22:21] * ChanServ sets mode +v andreask1
[22:21] * andreask is now known as Guest4429
[22:21] * andreask1 is now known as andreask
[22:21] * andreask (~andreask@h081217017238.dyn.cm.kabsi.at) Quit ()
[22:21] * andreask (~andreask@h081217017238.dyn.cm.kabsi.at) has joined #ceph
[22:21] * ChanServ sets mode +v andreask
[22:22] <Sysadmin88> if you need replication of 3, then only having 3 nodes is a bit mad
[22:23] <redcavalier> I can understand. But even with let's say, 7 physical nodes, the odds of losing files does scare me a bit.
[22:23] <Sysadmin88> in your crushmap you can specify where the failure domains is and make sure the replicas are not in the same failure domain
[22:23] <Sysadmin88> so you would set it up so the replicas are all on different nodes, maybe even different racks
[22:23] <redcavalier> I see
[22:23] <redcavalier> That makes a lot of sense
[22:26] * andreask (~andreask@h081217017238.dyn.cm.kabsi.at) has left #ceph
[22:27] * Guest4429 (~andreask@h081217017238.dyn.cm.kabsi.at) Quit (Ping timeout: 480 seconds)
[22:27] * sz0 (~sz0@94.55.197.185) has joined #ceph
[22:27] * ikrstic (~ikrstic@109-93-184-39.dynamic.isp.telekom.rs) Quit (Quit: Konversation terminated!)
[22:31] * KevinPerks (~Adium@cpe-098-025-128-231.sc.res.rr.com) has joined #ceph
[22:31] * kevinc (~kevinc__@client65-44.sdsc.edu) has joined #ceph
[22:32] * i_m (~ivan.miro@nat-5-carp.hcn-strela.ru) Quit (Quit: Leaving.)
[22:33] * markbby (~Adium@168.94.245.4) Quit (Remote host closed the connection)
[22:37] * markbby (~Adium@168.94.245.4) has joined #ceph
[22:39] * cok1 (~chk@46.30.211.29) has left #ceph
[22:44] * lupu (~lupu@86.107.101.214) has joined #ceph
[22:45] * andreask (~andreask@h081217017238.dyn.cm.kabsi.at) has joined #ceph
[22:45] * ChanServ sets mode +v andreask
[22:46] * andreask is now known as Guest4433
[22:47] * andreask (~andreask@h081217017238.dyn.cm.kabsi.at) has joined #ceph
[22:47] * ChanServ sets mode +v andreask
[22:48] * t0rn1 (~ssullivan@2607:fad0:32:a02:d227:88ff:fe02:9896) Quit (Quit: Leaving.)
[22:49] * andreask (~andreask@h081217017238.dyn.cm.kabsi.at) has left #ceph
[22:53] * Guest4433 (~andreask@h081217017238.dyn.cm.kabsi.at) Quit (Ping timeout: 480 seconds)
[22:54] <blynch> is there a procedure for migrating an existing rgw bucket from a replicated placement pool to a EC pool?
[22:56] * stewiem20001 (~stewiem20@195.10.250.233) Quit (Ping timeout: 480 seconds)
[22:56] * markbby (~Adium@168.94.245.4) Quit (Quit: Leaving.)
[22:56] * TiCPU (~jeromepou@190-130.cgocable.ca) Quit (Quit: Ex-Chat)
[22:57] * markbby (~Adium@168.94.245.4) has joined #ceph
[22:57] * rweeks (~rweeks@pat.hitachigst.com) Quit (Read error: Operation timed out)
[22:59] * ircolle-afk is now known as ircolle
[23:00] * KevinPerks (~Adium@cpe-098-025-128-231.sc.res.rr.com) Quit (Quit: Leaving.)
[23:00] * brad_mssw (~brad@shop.monetra.com) Quit (Quit: Leaving)
[23:14] * markbby (~Adium@168.94.245.4) Quit (Quit: Leaving.)
[23:15] <Cybertinus> redcavalier: and you can specify those groups yourself. Say you have a bit weird datacentre in which you have 10 racks. 5 of those racks are running on powerfeed A and the other 5 on powerfeed B. Then you don't care in which rack the data is, as long as you have one replica available on each powerfeed. Then you can add the powerfeed group :)
[23:17] * jeff-YF (~jeffyf@67.23.117.122) Quit (Quit: jeff-YF)
[23:21] <redcavalier> Well, my main concern was mostly with making no two copies of objects end up on the same node. We have several solutions for power redundancy and such. I've read the doc since I asked the question. Setting up the failure domains seems a bit complicated, but I guess I'll get the hang of it as I test it.
[23:21] <Cybertinus> ah, yes, for sure
[23:22] <Cybertinus> the default configuration of Ceph is (as far as I can interpreted the output of Ceph commands already) is that it will not put on 2 replica's on the same host
[23:22] * kevinc (~kevinc__@client65-44.sdsc.edu) Quit (Quit: This computer has gone to sleep)
[23:23] <redcavalier> ok, so it does consider this issue by default
[23:23] <Cybertinus> (I'm fighting with my first Ceph test install since yesterday, but I'm following the project for a while now, so I know about the theoretical possibilities ;) )
[23:23] <Cybertinus> yes
[23:23] <Cybertinus> by default you have 10 different levels of seperation or so :)
[23:23] <Cybertinus> don't know them from the top of my head
[23:23] <redcavalier> Right now I'm actually considering hardware options for a beta openstack cloud
[23:23] <Cybertinus> ok
[23:24] <redcavalier> we've been running a proof of concept for a while and it runs very well
[23:24] <Cybertinus> my employee is gonna run it as the storage platform behind his cloudplatform as well
[23:24] * kevinc (~kevinc__@client65-44.sdsc.edu) has joined #ceph
[23:24] <Cybertinus> we are gonna start with 6 data nodes with 4 OSD each. And then 3 monitor nodes, so 9 boxes in total
[23:25] <redcavalier> Sounds good. You can scale up anyway if it's not enough.
[23:25] <Cybertinus> jup, indeed
[23:25] <Cybertinus> which is something we will do in the pretty near future
[23:26] <Cybertinus> not because it will be absolutely neccaserry, but because we have all the hardware to spare. We can put it on the selve gathering dust, or speed up the Ceph cluster :)
[23:26] <Cybertinus> not a hard choice :p
[23:27] <redcavalier> hahaha, indeed
[23:27] <Cybertinus> and then we will start playing with more OSDs in 1 physical machine. Like 8 or 12 OSDs in one machine. But then we also add 3 of those machines
[23:28] <Cybertinus> so that prevents that we will loose data, when a physical machine goes down.
[23:28] * cookednoodles (~eoin@eoin.clanslots.com) Quit (Quit: Ex-Chat)
[23:28] <redcavalier> right now we're thinking of building it to satisfy a maximum of 25k iops and 15TB of space at the very least. Still haven't decided exactly how many nodes we're gonna split the cluster into, but I'd like to keep the number of OSDs per machine fairly low.
[23:29] <Cybertinus> yeah, I think (at least for starters) that 4 or 8 OSDs per node is a nice startingpoint
[23:29] <Sysadmin88> SSDs?
[23:29] <Cybertinus> just for the data distributions
[23:30] <redcavalier> Sysadmin88 : inktank recommends either SSDs or 7200 rpms drives
[23:30] <Sysadmin88> 15TB of space in the cluster or 15tb usable space?
[23:30] <redcavalier> I'm following that for now, unless you have suggestions
[23:31] * allsystemsarego (~allsystem@79.115.170.35) Quit (Quit: Leaving)
[23:31] <redcavalier> 15 tb in the cluster. Right now we only need 5tb usable.
[23:31] <Sysadmin88> 15tb is doable for SSDs... but if your replicating 3 times 45TB is a chunk more...
[23:31] <redcavalier> I was calculating 5 tb usable space replicated 3 times
[23:32] <Sysadmin88> doable with pure SSDs
[23:32] <redcavalier> that'S 100 vm at 50 GB . This is for beta purpose though.
[23:32] <Sysadmin88> just get good networking so your not limited that way
[23:34] * gregmark (~Adium@cet-nat-254.ndceast.pa.bo.comcast.net) Quit (Quit: Leaving.)
[23:34] <redcavalier> yea, I'm not exactly sure what the network admin has in mind yet, but I'll make sure to specify that.
[23:34] <Cybertinus> we use 10 gbit switches for the networking. Each cephnode is connected to both 10 gbe switches, via a reduntant setup :)
[23:34] <Cybertinus> lacp iirc
[23:35] <lurbs> Must be some kind of state sync between the switches too, if it's active/active bonds across the switches.
[23:35] <Cybertinus> yeah, the two switches are configured in a virtual chassis, so that sync is there
[23:36] * kevinc (~kevinc__@client65-44.sdsc.edu) Quit (Quit: This computer has gone to sleep)
[23:38] <jiffe> so if I have osds split between 2 datacenters and the pool size = 3, how can I make sure that at least one replicated object will end up in each datacenter?
[23:39] <Sysadmin88> careful with multi datacenter
[23:39] <Cybertinus> you need to add the correct rules to your crushmap (yes, that is pretty vage, but I don't know the exact details of it too)
[23:41] <redcavalier> Do you have dedicated pipes between your datacenters? (transfering ceph data through the regular network sounds very risky.)
[23:41] <jiffe> we have multiple lightly used 10G links between them
[23:41] <Sysadmin88> problem may be that ceph writes ALL nodes before ack'ing
[23:42] <Vacum> jiffe: why only 1 copy in one of the two DCs? if the other goes down you only have 1 copy left
[23:42] <lurbs> jiffe: http://ceph.com/docs/master/rados/operations/crush-map/ would be a good starting point.
[23:45] <lurbs> Also, if you have a split brain situation then (at least) the data centre that has fewer monitors in it (and therefore out of quorum) would be read only, regardless of if it had any replicas.
[23:47] * Tamil1 (~Adium@cpe-108-184-74-11.socal.res.rr.com) Quit (Quit: Leaving.)
[23:47] <lurbs> Actually, probably not even readable.
[23:48] * jakes (~oftc-webi@128-107-239-235.cisco.com) has joined #ceph
[23:49] <Vacum> lurbs: correct, not readable. the osds not being able to see a quorum will stop working and they will also be marked down the other osds that can still talk to the leader mon
[23:49] <jakes> I want to know if pools can be formed whose OSDs are hosted by different Servers.
[23:49] <Vacum> jakes: yes
[23:49] <lurbs> jakes: You can have a different CRUSH map per pool, so the placement logic can be separate.
[23:51] * kevinc (~kevinc__@client65-44.sdsc.edu) has joined #ceph
[23:51] <jakes> ok. Thanks. I have been reading more ceph placement on http://www.sebastien-han.fr/blog/2012/12/07/ceph-2-speed-storage-with-crush/. But the author said no to "We have 3 servers and each server has 2 SSD , so OSD.0 , OSD.1 ( server 1 ) OSD.2 and 3 ( server 2 ) , OSD.4 and 5 ( server 3 ) . Please note there are only SSD. Quest : Can we create a pool with OSD.0 OSD.2 and OSD.4 and another pool as OSD.1 , OSD.3 and OSD.5 ??". Hence i got confused
[23:52] * Tamil1 (~Adium@cpe-108-184-74-11.socal.res.rr.com) has joined #ceph
[23:53] <jakes> In the above case, I would be able to create two pools. correct ?
[23:54] <Vacum> jakes: you can create more than one pool on the same OSDs
[23:55] <Vacum> jakes: Sebastien tries something different there. he wants to create two pools that use OSDs on the same hosts differently. ie SSDs and HDDs in the same hosts, but used by different pools
[23:56] <Vacum> I'm not sure if that is a good idea. and I doubt it is easily manageable.
[23:57] <Vacum> jakes: ceph's default crush rule since some versions is to put replicas of PGs to different hosts.
[23:57] <Vacum> jakes: so if you want to create a pool with replica size 3, you need at least 3 hosts (well, better make that 4 hosts for several reasons)
[23:58] * dmsimard is now known as dmsimard_away
[23:59] <jakes> Vacum: I am sorry . I am still confused. In my example, osds of each pool are from different hosts.

These logs were automatically created by CephLogBot on irc.oftc.net using the Java IRC LogBot.