#ceph IRC Log

Index

IRC Log for 2014-06-26

Timestamps are in GMT/BST.

[0:01] * gregmark (~Adium@68.87.42.115) Quit (Quit: Leaving.)
[0:01] * hufman (~hufman@cpe-184-58-235-28.wi.res.rr.com) Quit (Quit: leaving)
[0:03] * bandrus (~oddo@162.223.167.195) Quit (Read error: Connection reset by peer)
[0:04] * bandrus (~oddo@162.223.167.195) has joined #ceph
[0:15] * dmsimard is now known as dmsimard_away
[0:17] * bandrus (~oddo@162.223.167.195) Quit (Read error: Connection reset by peer)
[0:17] * Cube (~Cube@12.248.40.138) has joined #ceph
[0:18] * bandrus (~oddo@162.223.167.195) has joined #ceph
[0:22] * AfC (~andrew@2001:44b8:31cb:d400:6e88:14ff:fe33:2a9c) has joined #ceph
[0:22] * Cube (~Cube@12.248.40.138) Quit (Read error: Operation timed out)
[0:23] * JuanEpstein (~rweeks@50.141.85.7) has joined #ceph
[0:23] * rweeks is now known as Guest348
[0:23] * JuanEpstein is now known as rweeks
[0:23] * aldavud (~aldavud@217-162-119-191.dynamic.hispeed.ch) has joined #ceph
[0:24] * markbby (~Adium@168.94.245.2) Quit (Quit: Leaving.)
[0:24] * Cube1 (~Cube@12.248.40.138) Quit (Ping timeout: 480 seconds)
[0:27] * Guest348 (~rweeks@50.141.85.7) Quit (Read error: Operation timed out)
[0:27] * rweeks (~rweeks@50.141.85.7) Quit ()
[0:30] <scuttlemonkey> ** 30m until CDS day 2 begins **
[0:31] * ChanServ sets mode +o scuttlemonkey
[0:31] <athrift> got it sorted, our developer was not waiting for long enough since starting the scrub to see the status change
[0:32] * joao|lap (~JL@2607:f298:a:607:ad2c:eade:9348:185b) Quit (Quit: Leaving)
[0:33] * joao|lap (~JL@2607:f298:a:607:3095:db69:97f6:1b54) has joined #ceph
[0:33] * ChanServ sets mode +o joao|lap
[0:33] * KevinPerks (~Adium@CPE602ad091f0ad-CM602ad091f0aa.cpe.net.cable.rogers.com) Quit (Quit: Leaving.)
[0:33] <scuttlemonkey> vilobhmm: it's joshd, if you are looking for Josh Durgin
[0:34] * sarob (~sarob@nat-dip27-wl-a.cfw-a-gci.corp.yahoo.com) Quit (Remote host closed the connection)
[0:37] * sjm (~sjm@cloudgate.cs.utsa.edu) has left #ceph
[0:38] * Nacer (~Nacer@c2s31-2-83-152-89-219.fbx.proxad.net) Quit (Read error: Operation timed out)
[0:41] * PerlStalker (~PerlStalk@72.166.192.70) Quit (Quit: ...)
[0:41] * rudolfsteiner (~federicon@181.21.148.181) has joined #ceph
[0:42] * PureNZ (~paul@122-62-45-132.jetstream.xtra.co.nz) Quit (Ping timeout: 480 seconds)
[0:49] <vilobhmm> hi scuttlemonkey
[0:50] * aldavud (~aldavud@217-162-119-191.dynamic.hispeed.ch) Quit (Ping timeout: 480 seconds)
[0:50] * rudolfsteiner (~federicon@181.21.148.181) Quit (Quit: rudolfsteiner)
[0:52] * Cube (~Cube@66.87.131.209) has joined #ceph
[0:53] <scuttlemonkey> vilobhmm: howdy
[0:54] * zidarsk8 (~zidar@89-212-142-10.dynamic.t-2.net) has joined #ceph
[0:54] * tdasilva (~quassel@dhcp-0-26-5a-b5-f8-68.cpe.townisp.com) has joined #ceph
[0:55] <scuttlemonkey> ** 5m until CDS Day 2 begins **
[0:55] <vilobhmm> looks like cinder fails to create volume from image stored on ceph
[0:56] * rturk|afk is now known as rturk
[0:56] <vilobhmm> in havana bug : https://bugs.launchpad.net/cinder/+bug/1247998
[0:56] <vilobhmm> Mike perez asked me to talk to you
[0:56] <vilobhmm> i tired converting the image to raw format
[0:57] <vilobhmm> and then try again by setting the volume_tmp_dir in cinder.conf
[0:57] <vilobhmm> to /tmp
[0:57] <vilobhmm> so that imge to volume conversion happens but ran into https://bugs.launchpad.net/cinder/+bug/1183283
[1:00] <vilobhmm> scuttlemonkey : ping
[1:01] <scuttlemonkey> vilobhmm: we're just starting day 2 of our ceph developer summit
[1:01] * grepory (uid29799@uxbridge.irccloud.com) Quit (Ping timeout: 480 seconds)
[1:01] <scuttlemonkey> we can catch up tomorrow if you have a chance though
[1:01] <vilobhmm> scuttlemonkey : sure
[1:01] <scuttlemonkey> although I'm probably not the guy to help...might be easier to send it to the lists
[1:02] <vilobhmm> scuttlemonkey : sure will do that ; any one specific you know who can be helpful here
[1:03] <scuttlemonkey> vilobhmm: several folks potentially...Joshd is probably the right guy eventually though
[1:03] <vilobhmm> ok
[1:05] * primechuck (~primechuc@69.170.148.179) Quit (Ping timeout: 480 seconds)
[1:05] * kevinc (~kevinc__@client65-78.sdsc.edu) Quit (Quit: This computer has gone to sleep)
[1:06] * kevinc (~kevinc__@client65-78.sdsc.edu) has joined #ceph
[1:06] * via (~via@smtp2.matthewvia.info) Quit (Ping timeout: 480 seconds)
[1:06] * saaby_ (~as@mail.saaby.com) Quit (Ping timeout: 480 seconds)
[1:08] * zack_dolby (~textual@pdf8519e7.tokynt01.ap.so-net.ne.jp) Quit (Quit: My MacBook has gone to sleep. ZZZzzz???)
[1:10] * via (~via@smtp2.matthewvia.info) has joined #ceph
[1:11] * saaby (~as@mail.saaby.com) has joined #ceph
[1:12] * wschulze (~wschulze@cpe-69-206-251-158.nyc.res.rr.com) Quit (Quit: Leaving.)
[1:12] * grepory (uid29799@id-29799.uxbridge.irccloud.com) has joined #ceph
[1:13] * mdjp (~mdjp@2001:41d0:52:100::343) Quit (Quit: mdjp has quit)
[1:14] * diegows (~diegows@190.190.5.238) Quit (Ping timeout: 480 seconds)
[1:14] * zidarsk8 (~zidar@89-212-142-10.dynamic.t-2.net) has left #ceph
[1:16] * Nats_ (~Nats@2001:8000:200c:0:ccb0:7f7e:aca8:ad13) has joined #ceph
[1:16] * ssejourne (~ssejourne@37.187.216.206) Quit (Remote host closed the connection)
[1:17] * jlogan (~Thunderbi@72.5.59.176) Quit (Read error: Operation timed out)
[1:17] * lpabon (~quassel@66-189-8-115.dhcp.oxfr.ma.charter.com) has joined #ceph
[1:17] * jlogan (~Thunderbi@72.5.59.176) has joined #ceph
[1:17] * aldavud (~aldavud@217-162-119-191.dynamic.hispeed.ch) has joined #ceph
[1:18] * lpabon (~quassel@66-189-8-115.dhcp.oxfr.ma.charter.com) Quit (Remote host closed the connection)
[1:18] * primechuck (~primechuc@69.170.148.179) has joined #ceph
[1:19] * lpabon (~quassel@66-189-8-115.dhcp.oxfr.ma.charter.com) has joined #ceph
[1:20] * joef (~Adium@138-72-131-163.pixar.com) Quit (Remote host closed the connection)
[1:21] * rendar (~I@87.19.182.180) Quit ()
[1:23] * Nats (~Nats@2001:8000:200c:0:f007:38b3:d5f8:4d9c) Quit (Ping timeout: 480 seconds)
[1:26] <sage> CDS day 2 is kicking off! https://bluejeans.com/379649356/browser
[1:28] * dlan (~dennis@116.228.88.131) Quit (Ping timeout: 480 seconds)
[1:28] * aldavud (~aldavud@217-162-119-191.dynamic.hispeed.ch) Quit (Read error: Operation timed out)
[1:29] * bandrus1 (~oddo@162.223.167.195) has joined #ceph
[1:29] * bandrus (~oddo@162.223.167.195) Quit (Read error: Connection reset by peer)
[1:31] * rudolfsteiner (~federicon@181.21.148.181) has joined #ceph
[1:32] * dlan (~dennis@116.228.88.131) has joined #ceph
[1:34] * rweeks (~rweeks@c-24-6-118-113.hsd1.ca.comcast.net) has joined #ceph
[1:37] * sjm (~sjm@ip-64-134-57-152.public.wayport.net) has joined #ceph
[1:39] * tdasilva (~quassel@dhcp-0-26-5a-b5-f8-68.cpe.townisp.com) Quit (Ping timeout: 480 seconds)
[1:39] * rudolfsteiner (~federicon@181.21.148.181) Quit (Ping timeout: 480 seconds)
[1:47] * dpippenger (~Adium@66-192-9-78.static.twtelecom.net) has joined #ceph
[1:50] * mikedawson (~chatzilla@c-98-220-189-67.hsd1.in.comcast.net) has joined #ceph
[1:50] * rweeks is now known as rweeks|away
[1:51] * sputnik13 (~sputnik13@207.8.121.241) Quit (Quit: My MacBook has gone to sleep. ZZZzzz???)
[1:51] * JayJ (~jayj@157.130.21.226) Quit (Quit: Computer has gone to sleep.)
[1:58] * nwat (~textual@207.194.238.3) has joined #ceph
[2:01] * kevinc (~kevinc__@client65-78.sdsc.edu) Quit (Quit: This computer has gone to sleep)
[2:06] * zack_dolby (~textual@e0109-114-22-12-137.uqwimax.jp) has joined #ceph
[2:08] * dpippenger (~Adium@66-192-9-78.static.twtelecom.net) has left #ceph
[2:09] * runfromnowhere (~runfromno@0001c1e7.user.oftc.net) has joined #ceph
[2:09] * guerby (~guerby@ip165-ipv6.tetaneutral.net) Quit (Ping timeout: 480 seconds)
[2:09] * sileht (~sileht@gizmo.sileht.net) Quit (Ping timeout: 480 seconds)
[2:10] * huangjun (~kvirc@111.173.81.152) has joined #ceph
[2:11] * xarses (~andreww@12.164.168.117) Quit (Read error: Operation timed out)
[2:14] * AfC (~andrew@2001:44b8:31cb:d400:6e88:14ff:fe33:2a9c) Quit (Quit: Leaving.)
[2:18] * nwat (~textual@207.194.238.3) Quit (Quit: My MacBook has gone to sleep. ZZZzzz???)
[2:19] * chrisjones (~chrisjone@12.237.137.162) has joined #ceph
[2:19] * primechuck (~primechuc@69.170.148.179) Quit (Remote host closed the connection)
[2:19] * lpabon (~quassel@66-189-8-115.dhcp.oxfr.ma.charter.com) Quit (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.)
[2:19] * lpabon (~quassel@66-189-8-115.dhcp.oxfr.ma.charter.com) has joined #ceph
[2:19] * lpabon (~quassel@66-189-8-115.dhcp.oxfr.ma.charter.com) Quit ()
[2:22] <chrisjones> Anyone have links on benchmarks in general and benchmarks with ceph vs glusterfs?
[2:24] * rmoe (~quassel@12.164.168.117) Quit (Ping timeout: 480 seconds)
[2:27] * sjm (~sjm@ip-64-134-57-152.public.wayport.net) has left #ceph
[2:29] * koleosfuscus (~koleosfus@adsl-84-226-68-69.adslplus.ch) Quit (Quit: koleosfuscus)
[2:37] * bandrus1 (~oddo@162.223.167.195) Quit (Quit: Leaving.)
[2:37] * beardo_ (~sma310@208-58-255-215.c3-0.drf-ubr1.atw-drf.pa.cable.rcn.com) Quit (Ping timeout: 480 seconds)
[2:37] * rmoe (~quassel@173-228-89-134.dsl.static.sonic.net) has joined #ceph
[2:44] * beardo_ (~sma310@216-164-126-13.c3-0.drf-ubr1.atw-drf.pa.cable.rcn.com) has joined #ceph
[2:49] * hasues (~hazuez@adsl-74-178-236-65.jax.bellsouth.net) has joined #ceph
[2:51] * joao|lap (~JL@2607:f298:a:607:3095:db69:97f6:1b54) Quit (Ping timeout: 480 seconds)
[2:56] * KevinPerks (~Adium@CPE602ad091f0ad-CM602ad091f0aa.cpe.net.cable.rogers.com) has joined #ceph
[2:57] * prudhvi (~p@tau.supr.io) has joined #ceph
[2:58] <prudhvi> Hi, is the a way to run osd's directly on a block files instead of on a filesystem.
[2:58] * prudhvi (~p@tau.supr.io) has left #ceph
[2:58] * prudhvi (~p@tau.supr.io) has joined #ceph
[3:02] <dmick> prudhvi: no. the journal can use a block device, but the osd data needs a filesystem
[3:05] * longguang (~chatzilla@123.126.33.253) has joined #ceph
[3:09] * markbby (~Adium@168.94.245.1) has joined #ceph
[3:09] * narb (~Jeff@38.99.52.10) Quit (Quit: narb)
[3:11] * AfC (~andrew@jim1020952.lnk.telstra.net) has joined #ceph
[3:14] * markbby1 (~Adium@168.94.245.2) has joined #ceph
[3:14] * markbby (~Adium@168.94.245.1) Quit (Remote host closed the connection)
[3:14] * rturk is now known as rturk|afk
[3:15] * yanzheng (~zhyan@jfdmzpr03-ext.jf.intel.com) has joined #ceph
[3:17] <runfromnowhere> Has anyone had any experience trying to run CephFS with large directories (like 7,000,000+ files)? Or is this perhaps something that Should Never Be Done?
[3:17] * adamcrume (~quassel@c-71-204-162-10.hsd1.ca.comcast.net) Quit (Ping timeout: 480 seconds)
[3:17] * lucas1 (~Thunderbi@222.240.148.130) has joined #ceph
[3:18] * adamcrume (~quassel@c-71-204-162-10.hsd1.ca.comcast.net) has joined #ceph
[3:19] <yanzheng> runfromnowhere, I have tried 1M files
[3:19] * zerick (~eocrospom@190.187.21.53) Quit (Read error: Operation timed out)
[3:20] <yanzheng> MDS will use lots of memory in this case
[3:20] * adamcrume (~quassel@c-71-204-162-10.hsd1.ca.comcast.net) Quit (Read error: Operation timed out)
[3:24] <runfromnowhere> yanzheng: Right now I'm running one MDS and was able to get about 2 million files into it before it stopped being able to handle it reasonably. Right now it's busy deleting the directory at a very slow pace
[3:24] <runfromnowhere> I'm wondering whether multiple MDS instances will help
[3:24] <yanzheng> mds bal_frag = 1
[3:24] <yanzheng> mds bal_split_bits = 2
[3:24] <yanzheng> mds bal_split_size = 8192
[3:24] <yanzheng> mds bal_merge_size = 1024
[3:25] <yanzheng> add these lines to your ceph.conf
[3:25] <runfromnowhere> I've seen this - will it improve performance with only a single MDS?
[3:25] <runfromnowhere> Or will it only help in a multi-MDS environment (as there's more than one MDS to fragment between)
[3:25] <yanzheng> no, but i will make MDS handle large directory better
[3:26] <yanzheng> also works for single MDS
[3:26] <gregsfortytwo1> it will help for a single MDS as well, though less than if it had multiple MDSes
[3:26] * runfromnowhere nods
[3:26] <runfromnowhere> I'll give it a try
[3:27] <runfromnowhere> I was going to try adding another MDS but I was seeing a weird flapping issue and figured it might be due to having this giant directory clogging things up :)
[3:27] <gregsfortytwo1> the size of the "directory frag" is one of the basic units that the MDS works in; it always reads an entire frag out of RADOS at a time, etc
[3:27] <gregsfortytwo1> you'll also want to turn up the size of the MDS cache if you can
[3:28] <gregsfortytwo1> ("mds cache size", defaults to 100000 inodes)
[3:28] * Nacer (~Nacer@c2s31-2-83-152-89-219.fbx.proxad.net) has joined #ceph
[3:28] * adamcrume (~quassel@c-71-204-162-10.hsd1.ca.comcast.net) has joined #ceph
[3:28] <runfromnowhere> Let's see what it does if I try it right now!
[3:28] <runfromnowhere> Fortunately this setup isn't in production :)
[3:31] * rturk|afk is now known as rturk
[3:31] * xarses (~andreww@c-24-23-183-44.hsd1.ca.comcast.net) has joined #ceph
[3:34] <runfromnowhere> Wow that's a huge improvement
[3:36] * xarses (~andreww@c-24-23-183-44.hsd1.ca.comcast.net) Quit (Read error: Operation timed out)
[3:37] * LeaChim (~LeaChim@host86-174-77-240.range86-174.btcentralplus.com) Quit (Ping timeout: 480 seconds)
[3:37] <runfromnowhere> LOL still 90 seconds to do 'ls -f | wc -l' on 1.3 million files but that's still infinitely better than "it doesn't ever finish"
[3:41] * adamcrume (~quassel@c-71-204-162-10.hsd1.ca.comcast.net) Quit (Remote host closed the connection)
[3:41] * KaZeR (~kazer@64.201.252.132) Quit (Remote host closed the connection)
[3:49] * tiger (~textual@58.213.102.114) has joined #ceph
[3:59] * angdraug (~angdraug@12.164.168.117) Quit (Quit: Leaving)
[4:01] * joao|lap (~JL@209-252-179-57.ip.mcleodusa.net) has joined #ceph
[4:01] * ChanServ sets mode +o joao|lap
[4:09] * jcsp_ (~john@ip-64-134-103-250.public.wayport.net) has joined #ceph
[4:10] * jcsp_ (~john@ip-64-134-103-250.public.wayport.net) Quit ()
[4:10] * jcsp1 (~jcsp@ip-64-134-103-250.public.wayport.net) has joined #ceph
[4:12] * longguang (~chatzilla@123.126.33.253) Quit (Ping timeout: 480 seconds)
[4:16] * markbby1 (~Adium@168.94.245.2) Quit (Ping timeout: 480 seconds)
[4:21] * tiger (~textual@58.213.102.114) Quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz???)
[4:23] * prudhvi (~p@tau.supr.io) Quit (Quit: leaving)
[4:23] * tiger (~textual@58.213.102.114) has joined #ceph
[4:28] * bkopilov (~bkopilov@213.57.16.185) Quit (Ping timeout: 480 seconds)
[4:28] * zack_dolby (~textual@e0109-114-22-12-137.uqwimax.jp) Quit (Read error: Connection reset by peer)
[4:29] * zack_dolby (~textual@e0109-114-22-12-137.uqwimax.jp) has joined #ceph
[4:34] * dmsimard_away is now known as dmsimard
[4:35] * danieagle (~Daniel@179.184.165.184.static.gvt.net.br) Quit (Quit: Obrigado por Tudo! :-) inte+ :-))
[4:37] * chrisjones (~chrisjone@12.237.137.162) Quit (Quit: chrisjones)
[4:42] * rturk is now known as rturk|afk
[4:46] * AfC (~andrew@jim1020952.lnk.telstra.net) Quit (Quit: Leaving.)
[4:46] * aarcane_ (~aarcane@99-42-64-118.lightspeed.irvnca.sbcglobal.net) Quit (Quit: Leaving)
[4:47] * shang (~ShangWu@175.41.48.77) has joined #ceph
[4:48] * longguang (~chatzilla@123.126.33.253) has joined #ceph
[4:50] * rturk|afk is now known as rturk
[4:52] * sjusthm (~sam@66-214-251-229.dhcp.gldl.ca.charter.com) Quit (Quit: Leaving.)
[4:53] * rudolfsteiner (~federicon@181.21.134.151) has joined #ceph
[4:54] * shang (~ShangWu@175.41.48.77) Quit (Read error: Operation timed out)
[4:57] * jcsp1 (~jcsp@ip-64-134-103-250.public.wayport.net) Quit (Ping timeout: 480 seconds)
[5:00] * shang (~ShangWu@103.17.197.129) has joined #ceph
[5:01] * madkiss (~madkiss@217.194.72.154) Quit (Read error: Connection reset by peer)
[5:04] * jianingy (~jianingy@211.151.238.51) Quit (Quit: Lost terminal)
[5:06] * alexxy (~alexxy@2001:470:1f14:106::2) Quit (Read error: Connection reset by peer)
[5:07] * alexxy (~alexxy@2001:470:1f14:106::2) has joined #ceph
[5:10] * longguang (~chatzilla@123.126.33.253) Quit (Read error: Connection reset by peer)
[5:10] * longguang (~chatzilla@123.126.33.253) has joined #ceph
[5:11] * Vacum_ (~vovo@88.130.217.220) has joined #ceph
[5:12] * alexxy (~alexxy@2001:470:1f14:106::2) Quit (Remote host closed the connection)
[5:15] * rturk is now known as rturk|afk
[5:15] * rturk|afk is now known as rturk
[5:16] * alexxy (~alexxy@79.173.81.171) has joined #ceph
[5:18] * Vacum (~vovo@88.130.221.161) Quit (Ping timeout: 480 seconds)
[5:27] * KevinPerks (~Adium@CPE602ad091f0ad-CM602ad091f0aa.cpe.net.cable.rogers.com) Quit (Quit: Leaving.)
[5:28] * tiger (~textual@58.213.102.114) Quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz???)
[5:28] * madkiss (~madkiss@217.194.72.154) has joined #ceph
[5:30] * rturk is now known as rturk|afk
[5:36] * codice_ (~toodles@97-94-175-73.static.mtpk.ca.charter.com) has joined #ceph
[5:37] * codice (~toodles@97-94-175-73.static.mtpk.ca.charter.com) Quit (Read error: Operation timed out)
[5:41] * xarses (~andreww@c-24-23-183-44.hsd1.ca.comcast.net) has joined #ceph
[5:42] <longguang> if ceph compress data during interaction between client and osd?
[5:44] * rweeks|away (~rweeks@c-24-6-118-113.hsd1.ca.comcast.net) Quit (Quit: Leaving)
[5:47] * dmsimard is now known as dmsimard_away
[5:56] * lucas1 (~Thunderbi@222.240.148.130) Quit (Quit: lucas1)
[5:59] * bkopilov (~bkopilov@nat-pool-tlv-t.redhat.com) has joined #ceph
[6:06] <sherry> how to bring the OSD which is down and out after interval (5 minutes)?
[6:07] <sherry> how to bring the OSD up which is down and out after interval (5 minutes)?
[6:07] * shang (~ShangWu@103.17.197.129) Quit (Ping timeout: 480 seconds)
[6:11] * athrift (~nz_monkey@203.86.205.13) Quit (Quit: No Ping reply in 180 seconds.)
[6:11] * athrift (~nz_monkey@203.86.205.13) has joined #ceph
[6:11] * xarses (~andreww@c-24-23-183-44.hsd1.ca.comcast.net) Quit (Read error: Connection reset by peer)
[6:15] * shang (~ShangWu@175.41.48.77) has joined #ceph
[6:17] * Muhlemmer (~kvirc@cable-90-50.zeelandnet.nl) has joined #ceph
[6:19] * jcsp (~Adium@0001bf3a.user.oftc.net) Quit (Read error: Connection reset by peer)
[6:20] * lalatenduM (~lalatendu@121.244.87.117) has joined #ceph
[6:23] * jcsp (~Adium@82-71-55-202.dsl.in-addr.zen.co.uk) has joined #ceph
[6:30] * wschulze (~wschulze@cpe-69-206-251-158.nyc.res.rr.com) has joined #ceph
[6:38] * Muhlemmer (~kvirc@cable-90-50.zeelandnet.nl) Quit (Read error: Operation timed out)
[6:41] * mjeanson (~mjeanson@00012705.user.oftc.net) Quit (Remote host closed the connection)
[6:42] * mjeanson (~mjeanson@bell.multivax.ca) has joined #ceph
[6:47] * longguang (~chatzilla@123.126.33.253) Quit (Ping timeout: 480 seconds)
[6:50] * tiger (~textual@58.213.102.114) has joined #ceph
[7:02] * pmatulis (~peter@ec2-23-23-42-7.compute-1.amazonaws.com) Quit (Remote host closed the connection)
[7:03] * lucas1 (~Thunderbi@218.76.25.66) has joined #ceph
[7:04] * ircolle (~Adium@2601:1:8380:2d9:14fe:84b:6e9d:df3d) Quit (Quit: Leaving.)
[7:05] * rdas (~rdas@121.244.87.115) has joined #ceph
[7:07] * saurabh (~saurabh@121.244.87.117) has joined #ceph
[7:08] * AfC (~andrew@2001:44b8:31cb:d400:6e88:14ff:fe33:2a9c) has joined #ceph
[7:11] * runfromnowhere (~runfromno@0001c1e7.user.oftc.net) Quit (Quit: Changing server)
[7:11] * guerby (~guerby@ip165-ipv6.tetaneutral.net) has joined #ceph
[7:15] * runfromnowhere (~runfromno@pool-108-29-25-203.nycmny.fios.verizon.net) has joined #ceph
[7:20] * DavidThunder (~Thunderbi@ip68-4-173-198.oc.oc.cox.net) Quit (Quit: DavidThunder)
[7:20] * vbellur (~vijay@122.167.109.216) Quit (Ping timeout: 480 seconds)
[7:26] * wschulze (~wschulze@cpe-69-206-251-158.nyc.res.rr.com) Quit (Quit: Leaving.)
[7:26] * michalefty (~micha@p20030071CF5304002DC9E382D2950321.dip0.t-ipconnect.de) has joined #ceph
[7:28] * houkouonchi-home (~linux@pool-71-189-160-82.lsanca.fios.verizon.net) Quit (Remote host closed the connection)
[7:29] * rudolfsteiner (~federicon@181.21.134.151) Quit (Quit: rudolfsteiner)
[7:30] * longguang (~chatzilla@123.126.33.253) has joined #ceph
[7:30] * lucas1 (~Thunderbi@218.76.25.66) Quit (Quit: lucas1)
[7:30] * pmatulis (~peter@ec2-23-23-42-7.compute-1.amazonaws.com) has joined #ceph
[7:32] * madkiss (~madkiss@217.194.72.154) Quit (Quit: Leaving.)
[7:42] * vbellur (~vijay@209.132.188.8) has joined #ceph
[7:47] * tiger (~textual@58.213.102.114) Quit (Quit: Textual IRC Client: www.textualapp.com)
[7:51] * pingu (~christian@mail.ponies.io) Quit (Quit: leaving)
[7:54] * lucas1 (~Thunderbi@222.247.57.50) has joined #ceph
[7:57] * i_m (~ivan.miro@nat-5-carp.hcn-strela.ru) has joined #ceph
[8:00] * ikrstic (~ikrstic@178-222-94-242.dynamic.isp.telekom.rs) has joined #ceph
[8:03] * davidzlap (~Adium@ip68-4-173-198.oc.oc.cox.net) Quit (Quit: Leaving.)
[8:13] * Cube (~Cube@66.87.131.209) Quit (Ping timeout: 480 seconds)
[8:28] * hasues (~hazuez@adsl-74-178-236-65.jax.bellsouth.net) Quit (Ping timeout: 480 seconds)
[8:30] * vilobhmm (~vilobhmm@nat-dip27-wl-a.cfw-a-gci.corp.yahoo.com) Quit (Quit: vilobhmm)
[8:33] * vbellur (~vijay@209.132.188.8) Quit (Read error: Operation timed out)
[8:33] * houkouonchi-home (~linux@pool-71-189-160-82.lsanca.fios.verizon.net) has joined #ceph
[8:35] * Sysadmin88 (~IceChat77@94.4.20.0) Quit (Quit: For Sale: Parachute. Only used once, never opened, small stain.)
[8:36] * aldavud (~aldavud@213.55.176.144) has joined #ceph
[8:36] * madkiss (~madkiss@nat.nue.novell.com) has joined #ceph
[8:37] <longguang> when create a rbd, if can we specify a format like qcow2?
[8:41] * Nacer (~Nacer@c2s31-2-83-152-89-219.fbx.proxad.net) Quit (Remote host closed the connection)
[8:43] * Nacer (~Nacer@c2s31-2-83-152-89-219.fbx.proxad.net) has joined #ceph
[8:43] * sileht (~sileht@gizmo.sileht.net) has joined #ceph
[8:44] * rendar (~I@host117-179-dynamic.56-79-r.retail.telecomitalia.it) has joined #ceph
[8:44] * vbellur (~vijay@121.244.87.117) has joined #ceph
[8:48] * stephan (~stephan@62.217.45.26) Quit (Quit: Ex-Chat)
[8:48] * Nacer (~Nacer@c2s31-2-83-152-89-219.fbx.proxad.net) Quit (Remote host closed the connection)
[8:50] * lucas1 (~Thunderbi@222.247.57.50) Quit (Quit: lucas1)
[8:56] * lucas1 (~Thunderbi@222.247.57.50) has joined #ceph
[9:04] * haomaiwang (~haomaiwan@175.152.18.107) has joined #ceph
[9:06] * ScOut3R (~ScOut3R@catv-89-133-22-210.catv.broadband.hu) has joined #ceph
[9:07] * BManojlovic (~steki@91.195.39.5) has joined #ceph
[9:08] * haomaiwa_ (~haomaiwan@125-227-255-23.HINET-IP.hinet.net) has joined #ceph
[9:08] * zack_dolby (~textual@e0109-114-22-12-137.uqwimax.jp) Quit (Quit: My MacBook has gone to sleep. ZZZzzz???)
[9:12] * haomaiwang (~haomaiwan@175.152.18.107) Quit (Ping timeout: 480 seconds)
[9:15] * zack_dolby (~textual@e0109-114-22-12-137.uqwimax.jp) has joined #ceph
[9:22] * analbeard (~shw@support.memset.com) has joined #ceph
[9:36] <Hell_Fire__> longguang: rbds are just blocks of storage, they're all thinly/sparse provisioned anyway
[9:36] * Hell_Fire__ is now known as Hell_Fire
[9:36] * cookednoodles (~eoin@eoin.clanslots.com) has joined #ceph
[9:38] <longguang> Hell_Fire__: just only raw fromat?
[9:39] * aldavud (~aldavud@213.55.176.144) Quit (Ping timeout: 480 seconds)
[9:39] <Hell_Fire> well yeah, qcow2 is a file format, rbds are presented more like actual disks, you don't format actual disks with qcow2 :)
[9:40] * zidarsk8 (~zidar@prevod.fri1.uni-lj.si) has joined #ceph
[9:41] * Nacer (~Nacer@252-87-190-213.intermediasud.com) has joined #ceph
[9:41] * Nacer (~Nacer@252-87-190-213.intermediasud.com) Quit (Remote host closed the connection)
[9:42] * Nacer (~Nacer@252-87-190-213.intermediasud.com) has joined #ceph
[9:43] <longguang> (actually, we can. it is not important at the moment) you mean if use rbd , we can not use qcow2? and if rbd supports the advantage qcow2 has?
[9:44] * zidarsk8 (~zidar@prevod.fri1.uni-lj.si) has left #ceph
[9:44] * Infitialis (~infitiali@194.30.182.18) has joined #ceph
[9:44] <longguang> maybe the only advantage is snapshot???
[9:44] <Hell_Fire> yeah, rbds do snapshots
[9:45] <longguang> do you know how qcow2 takes snapshot? do the both have the same theory?
[9:46] <absynth> http://wiki.qemu.org/Features/Snapshots
[9:47] <longguang> i know qcow2's theory, rather than rbd
[9:48] <absynth> https://ceph.com/docs/master/rbd/rbd-snapshot/
[9:49] <absynth> since you can interoperate rbd snapshots and qcow snapshots on the API level (ie. qemu snapshotting works with both underlying storage layers), i guess the theory is more or less the same
[9:49] * ikrstic (~ikrstic@178-222-94-242.dynamic.isp.telekom.rs) Quit (Quit: Konversation terminated!)
[9:50] <longguang> you mean qemu know how to take snapshot based on backend storage?
[9:50] <cookednoodles> longguang, you already asked this question a few days ago, and I already answered you
[9:50] <cookednoodles> why not just .. test it ?
[9:51] <absynth> yeah
[9:51] <longguang> it should be. i do not have environment now.
[9:51] <absynth> no blah blah, just test it
[9:51] <longguang> i encounter some problems during setup a ceph cluster.
[9:52] * odyssey4me (~odyssey4m@165.233.71.2) has joined #ceph
[9:52] <Hell_Fire> get that going first :) It's not too tricky
[9:53] <cookednoodles> right, if you have setup issues, then ask about the setup :P
[9:53] <longguang> in my case , i only have one node. and i want to use local ext4. how to write a conf file?
[9:54] <cookednoodles> how many osds ?
[9:54] <absynth> if you have one node, ceph is completely idiotic
[9:54] * fsimonce (~simon@host29-76-dynamic.9-79-r.retail.telecomitalia.it) has joined #ceph
[9:54] <absynth> use local qcow2 images and be done with it
[9:54] <longguang> one
[9:54] <absynth> the smallest testbed setup for ceph consists of three nodes
[9:55] <absynth> with colocated OSDs and MONs, and ideally between 6 and 12 OSDs
[9:55] <cookednoodles> nah you can do 2 nodes
[9:55] <longguang> i want to study ceph.
[9:55] <Hell_Fire> if you don't have enough nodes for the testbed, virtualize it :)
[9:55] <Hell_Fire> my dummy/test setup is virtual :)
[9:55] <absynth> cookednoodles: only it doesn't make any sense
[9:55] <cookednoodles> yeah but for testing and learning how it works, its not bad
[9:55] <absynth> since you cannot test MON redundancy in a meaningful way
[9:55] <cookednoodles> sure you can
[9:55] <cookednoodles> stop an osd randomly
[9:56] <absynth> errm
[9:56] <absynth> mon redundancy, not osd redundancy
[9:56] * zack_dolby (~textual@e0109-114-22-12-137.uqwimax.jp) Quit (Ping timeout: 480 seconds)
[9:56] <absynth> i.e. what happens if you kill a mon
[9:56] <cookednoodles> ah yep
[9:56] <absynth> in a 2 node setup, your cluster is then broken
[9:56] <cookednoodles> but thats for another day :P
[9:57] <cookednoodles> to mess about, store stuff and learn all the syntax and settings
[9:57] <cookednoodles> you can do it on one machine, 2 osds
[9:57] <longguang> please. i do not intend to test high feature. just know how it work. how qemu work with rbd.
[9:57] <cookednoodles> INSTALL IT
[9:57] <longguang> yes.
[9:57] <cookednoodles> go
[9:58] <longguang> :(
[9:58] <cookednoodles> http://ceph.com/docs/master/start/
[10:00] * rotbeard (~redbeard@b2b-94-79-138-170.unitymedia.biz) has joined #ceph
[10:01] * sz0 (~sz0@94.54.193.66) Quit (Ping timeout: 480 seconds)
[10:02] * sz0 (~sz0@94.54.193.66) has joined #ceph
[10:04] * zack_dolby (~textual@e0109-114-22-12-137.uqwimax.jp) has joined #ceph
[10:05] <longguang> another matter , are you familar with the code?
[10:07] * jordanP (~jordan@185.23.92.11) has joined #ceph
[10:07] * Cube (~Cube@66.87.64.87) has joined #ceph
[10:07] * jordanP (~jordan@185.23.92.11) Quit ()
[10:08] * jordanP (~jordan@185.23.92.11) has joined #ceph
[10:08] * jordanP (~jordan@185.23.92.11) Quit ()
[10:09] * jordanP (~jordan@185.23.92.11) has joined #ceph
[10:11] * fdmanana (~fdmanana@bl10-142-244.dsl.telepac.pt) Quit (Quit: Leaving)
[10:13] <longguang> Error ENOENT: osd.0 does not exist. create it before updating the crush map
[10:19] * mikedawson (~chatzilla@c-98-220-189-67.hsd1.in.comcast.net) Quit (Read error: Connection reset by peer)
[10:19] * mikedawson (~chatzilla@c-98-220-189-67.hsd1.in.comcast.net) has joined #ceph
[10:19] * thomnico (~thomnico@2a01:e35:8b41:120:1c11:9e1d:44e4:a62c) has joined #ceph
[10:20] * thomnico (~thomnico@2a01:e35:8b41:120:1c11:9e1d:44e4:a62c) Quit ()
[10:28] * zhaochao (~zhaochao@220.181.73.105) has joined #ceph
[10:29] * mdjp (~mdjp@2001:41d0:52:100::343) has joined #ceph
[10:29] * sz0_ (~sz0@94.54.193.66) has joined #ceph
[10:30] * sz0 (~sz0@94.54.193.66) Quit (Ping timeout: 480 seconds)
[10:34] * thomnico (~thomnico@2a01:e35:8b41:120:1c11:9e1d:44e4:a62c) has joined #ceph
[10:37] * sz0_ (~sz0@94.54.193.66) Quit (Ping timeout: 480 seconds)
[10:41] * zhaochao (~zhaochao@220.181.73.105) Quit (Ping timeout: 480 seconds)
[10:42] * ade (~abradshaw@193.202.255.218) has joined #ceph
[10:49] * zhaochao (~zhaochao@123.151.134.231) has joined #ceph
[10:52] * Cube (~Cube@66.87.64.87) Quit (Read error: Connection reset by peer)
[10:52] * Cube (~Cube@66-87-64-87.pools.spcsdns.net) has joined #ceph
[10:56] * yanzheng (~zhyan@jfdmzpr03-ext.jf.intel.com) Quit (Remote host closed the connection)
[10:58] * ScOut3R (~ScOut3R@catv-89-133-22-210.catv.broadband.hu) Quit (Quit: Leaving...)
[10:59] * ScOut3R (~ScOut3R@catv-89-133-22-210.catv.broadband.hu) has joined #ceph
[11:02] * ikrstic (~ikrstic@c82-214-88-26.loc.akton.net) has joined #ceph
[11:06] * thomnico (~thomnico@2a01:e35:8b41:120:1c11:9e1d:44e4:a62c) Quit (Ping timeout: 480 seconds)
[11:12] * Cube (~Cube@66-87-64-87.pools.spcsdns.net) Quit (Quit: Leaving.)
[11:19] * Cube (~Cube@66-87-64-87.pools.spcsdns.net) has joined #ceph
[11:19] <odyssey4me> odd - my ceph-mon is reporting that one of the mons has a skew clock, but I've verified that the clocks are all the same... what could be the issue here?
[11:21] <absynth> high network latency between the mons?
[11:22] <Serbitar> you are running ntp right?
[11:22] <odyssey4me> absynth - nope, they're all on the same switch and communicating over bonded 10GB connections
[11:23] <odyssey4me> and yes, using ntp - they're all configured to use external time sources and are peered with each other
[11:24] <odyssey4me> well, the physical servers are peered - two mons are virtualised and don't participate in the peering, they just use the various time sources which all refer back to the same tier 2 ntp source
[11:24] <odyssey4me> oddly enough - it's one of the physical mons that's reported as having a skew clock
[11:25] <absynth> that may or may not be true
[11:25] <absynth> how many mons do you ahve all in all
[11:25] <absynth> ?
[11:25] <odyssey4me> 3 mons
[11:25] <odyssey4me> the offset reported by ntp is -4.704
[11:26] <odyssey4me> now -1.227
[11:26] * zack_dolby (~textual@e0109-114-22-12-137.uqwimax.jp) Quit (Ping timeout: 480 seconds)
[11:26] <odyssey4me> tiny
[11:26] * Cube (~Cube@66-87-64-87.pools.spcsdns.net) Quit (Quit: Leaving.)
[11:27] <absynth> so basically the two virtual mons could have the same skew, and the bare-metal mon could have an accurate time, right?
[11:27] <absynth> i'm asking because we were seeing weird clock skew issues on VMs quite regularly
[11:28] <odyssey4me> ntp on the virtuals is set with 'tinker panic 0' to ensure that regardless of how insane the time difference is, the vm must change it's time to the same as the source
[11:28] <kraken> http://i.imgur.com/WS4S2.gif
[11:28] <absynth> is that a trigger?
[11:28] <absynth> panic
[11:28] <kraken> http://i.imgur.com/fH9e2.gif
[11:29] <odyssey4me> the two virtual mons are on different physical hosts, so the skew between them and the physical mon is different... but very, very small
[11:29] <absynth> what the hell? panic
[11:29] <kraken> http://i.imgur.com/SNvM6CZ.gif
[11:29] <odyssey4me> lol, more and more panic ! ;)
[11:29] <kraken> http://i.imgur.com/WS4S2.gif
[11:29] <absynth> now he's repeating itself
[11:29] <absynth> s/he/it/
[11:29] <kraken> absynth meant to say: now it's repeating itself
[11:29] <absynth> who's bot is this?
[11:29] <odyssey4me> lol
[11:30] <odyssey4me> it appears to be a ceph bot : ~kraken@gw.sepia.ceph.com
[11:30] <absynth> kraken: help
[11:30] * kraken whispers to absynth
[11:31] * bitserker (~toni@63.pool85-52-240.static.orange.es) has joined #ceph
[11:32] * cronix1 (~cronix@5.199.139.166) Quit (Read error: Operation timed out)
[11:35] <longguang> what is 'erasure coded pools'?
[11:36] <absynth> dude, we're not a google frontend ;)
[11:36] <absynth> http://ceph.com/docs/firefly/dev/erasure-coded-pool/
[11:40] <odyssey4me> lol, if only kraken could be taught to do that :p
[11:40] <longguang> you are funny, i am newbies
[11:41] * beardo (~sma310@beardo.cc.lehigh.edu) Quit (Read error: Operation timed out)
[11:43] <cookednoodles> newbis google to find answers
[11:43] <odyssey4me> interesting, after waiting a while the clock skew went away
[11:43] <cookednoodles> well thats how ntp works :P
[11:44] <odyssey4me> it's possible that the clock skew was temporary after I rebooted the mon host, and the skew clock references were marked on some objects... which cleaned up over a little time
[11:44] <odyssey4me> cookednoodles - yeah :p ntp didn't think the clock was skew... only the mon did
[11:56] <Infitialis> kraken: help
[11:56] * kraken whispers to Infitialis
[11:57] <Infitialis> helga norris infitialis
[11:57] <Infitialis> awww
[11:57] <Infitialis> lol
[12:01] * kapil (~ksharma@nat.nue.novell.com) Quit (Ping timeout: 480 seconds)
[12:03] * yuki-chm (yuki-chm@111.173.81.152) has joined #ceph
[12:03] * kapil (~ksharma@2620:113:80c0:5::2222) has joined #ceph
[12:04] * ScOut3R (~ScOut3R@catv-89-133-22-210.catv.broadband.hu) Quit (Read error: Operation timed out)
[12:21] * zhaochao_ (~zhaochao@123.151.134.227) has joined #ceph
[12:23] * ScOut3R (~ScOut3R@catv-80-99-64-8.catv.broadband.hu) has joined #ceph
[12:26] * zhaochao (~zhaochao@123.151.134.231) Quit (Ping timeout: 480 seconds)
[12:26] * zhaochao_ is now known as zhaochao
[12:30] * lalatenduM (~lalatendu@121.244.87.117) Quit (Quit: Leaving)
[12:31] * mikedawson (~chatzilla@c-98-220-189-67.hsd1.in.comcast.net) Quit (Ping timeout: 480 seconds)
[12:34] * vbellur (~vijay@121.244.87.117) Quit (Read error: Operation timed out)
[12:36] * cronix1 (~cronix@5.199.139.166) has joined #ceph
[12:43] * ghartz (~ghartz@ircad17.u-strasbg.fr) has joined #ceph
[12:45] * cronix1 (~cronix@5.199.139.166) Quit (Ping timeout: 480 seconds)
[12:50] * beardo (~sma310@beardo.cc.lehigh.edu) has joined #ceph
[12:51] * odyssey4me (~odyssey4m@165.233.71.2) Quit (Ping timeout: 480 seconds)
[12:52] * vbellur (~vijay@209.132.188.8) has joined #ceph
[12:52] * lalatenduM (~lalatendu@209.132.188.8) has joined #ceph
[12:56] * yuki-chm (yuki-chm@111.173.81.152) Quit ()
[13:00] * cronix1 (~cronix@5.199.139.166) has joined #ceph
[13:01] <huangjun> we migrated vmware esxi vm (250GB) from ceph cluster to local disk but the mirgation rate is pretty slow, about 20MB/s for 8 OSDs
[13:03] <huangjun> we have two hosts OSD1 and OSD2, each have 1Gbs NIC
[13:04] * cronix1 (~cronix@5.199.139.166) Quit (Read error: Operation timed out)
[13:05] <lupu> how high is the disk utilization?
[13:05] * haomaiwa_ (~haomaiwan@125-227-255-23.HINET-IP.hinet.net) Quit (Remote host closed the connection)
[13:06] * haomaiwang (~haomaiwan@125-227-255-23.HINET-IP.hinet.net) has joined #ceph
[13:07] * koleosfuscus (~koleosfus@ws11-189.unine.ch) has joined #ceph
[13:08] * odyssey4me (~odyssey4m@165.233.71.2) has joined #ceph
[13:09] <huangjun> the disk is not busy
[13:10] <huangjun> i think it related to sparse file,
[13:11] <huangjun> we use kclient+NFS, and the NFS client read the sparse file randomly, read about 4K each request that will decrease the read performance
[13:11] * odyssey4me_ (~odyssey4m@165.233.71.2) has joined #ceph
[13:15] * saurabh (~saurabh@121.244.87.117) Quit (Quit: Leaving)
[13:16] * odyssey4me (~odyssey4m@165.233.71.2) Quit (Read error: Operation timed out)
[13:18] * odyssey4me_ is now known as odyssey4me
[13:19] * Infitialis (~infitiali@194.30.182.18) Quit ()
[13:21] * shang (~ShangWu@175.41.48.77) Quit (Read error: Operation timed out)
[13:22] * chrisjones (~chrisjone@12.237.137.162) has joined #ceph
[13:23] * mikedawson (~chatzilla@c-98-220-189-67.hsd1.in.comcast.net) has joined #ceph
[13:27] * odyssey4me (~odyssey4m@165.233.71.2) Quit (Ping timeout: 480 seconds)
[13:31] * zhaochao (~zhaochao@123.151.134.227) has left #ceph
[13:32] * longguang (~chatzilla@123.126.33.253) Quit (Ping timeout: 480 seconds)
[13:33] * thomnico (~thomnico@cro38-2-88-180-16-18.fbx.proxad.net) has joined #ceph
[13:34] * odyssey4me (~odyssey4m@165.233.71.2) has joined #ceph
[13:40] * shang (~ShangWu@175.41.48.77) has joined #ceph
[13:42] * yguang11 (~yguang11@2406:2000:ef96:e:8527:6f3f:98fe:1866) Quit ()
[13:44] * rudolfsteiner (~federicon@181.21.134.151) has joined #ceph
[13:45] * cronix1 (~cronix@5.199.139.166) has joined #ceph
[13:46] * mikedawson (~chatzilla@c-98-220-189-67.hsd1.in.comcast.net) Quit (Ping timeout: 480 seconds)
[13:46] * vbellur (~vijay@209.132.188.8) Quit (Ping timeout: 480 seconds)
[13:51] * Kioob (~kioob@sal69-4-78-192-172-15.fbxo.proxad.net) Quit (Ping timeout: 480 seconds)
[13:53] * huangjun (~kvirc@111.173.81.152) Quit (Ping timeout: 480 seconds)
[13:54] * cronix1 (~cronix@5.199.139.166) Quit (Ping timeout: 480 seconds)
[13:58] * vbellur (~vijay@121.244.87.117) has joined #ceph
[13:59] * longguang (~chatzilla@123.126.33.253) has joined #ceph
[14:00] <longguang> if ceph.ko is needed when mount cephfs?
[14:01] * haomaiwang (~haomaiwan@125-227-255-23.HINET-IP.hinet.net) Quit (Ping timeout: 480 seconds)
[14:02] * odyssey4me_ (~odyssey4m@165.233.71.2) has joined #ceph
[14:03] <joelio> hmm, my mds is stuck in replay
[14:03] <joelio> 2014-06-26 13:02:36.220118 7f414b068700 0 mds.0.cache creating system inode with ino:1
[14:03] <joelio> 0.80.1
[14:06] * diegows (~diegows@host-216-57-132-113.customer.veroxity.net) has joined #ceph
[14:07] * jordanP (~jordan@185.23.92.11) Quit (Ping timeout: 480 seconds)
[14:07] * odyssey4me (~odyssey4m@165.233.71.2) Quit (Read error: Operation timed out)
[14:13] <sherry> hi guys, I'd like to reiterate this bug, appreciate if u have any idea> http://tracker.ceph.com/issues/8641
[14:17] * t0rn (~ssullivan@2607:fad0:32:a02:d227:88ff:fe02:9896) has joined #ceph
[14:18] * dneary (~dneary@87-231-145-225.rev.numericable.fr) has joined #ceph
[14:29] * sjm (~sjm@174.47.140.2) has joined #ceph
[14:30] * cronix1 (~cronix@5.199.139.166) has joined #ceph
[14:32] * vbellur (~vijay@121.244.87.117) Quit (Ping timeout: 480 seconds)
[14:34] <longguang> hi
[14:34] * tdasilva (~quassel@nat-pool-bos-t.redhat.com) has joined #ceph
[14:38] * bkopilov (~bkopilov@nat-pool-tlv-t.redhat.com) Quit (Ping timeout: 480 seconds)
[14:38] * cronix1 (~cronix@5.199.139.166) Quit (Read error: Operation timed out)
[14:39] * mikedawson (~chatzilla@23-25-46-97-static.hfc.comcastbusiness.net) has joined #ceph
[14:39] * cronix1 (~cronix@5.199.139.166) has joined #ceph
[14:44] * thomnico (~thomnico@cro38-2-88-180-16-18.fbx.proxad.net) Quit (Quit: Ex-Chat)
[14:45] * thomnico (~thomnico@2a01:e35:8b41:120:1c11:9e1d:44e4:a62c) has joined #ceph
[14:46] * rwheeler (~rwheeler@nat-pool-bos-u.redhat.com) has joined #ceph
[14:46] * sz0 (~sz0@178.240.141.248) has joined #ceph
[14:47] * sroy (~sroy@2607:fad8:4:6:3e97:eff:feb5:1e2b) has joined #ceph
[14:47] * fretb (~fretb@drip.frederik.pw) Quit (Quit: leaving)
[14:48] * fretb (~fretb@drip.frederik.pw) has joined #ceph
[14:48] * imriz (~imriz@5.29.200.177) Quit (Ping timeout: 480 seconds)
[14:49] * wschulze (~wschulze@cpe-69-206-251-158.nyc.res.rr.com) has joined #ceph
[14:51] * pressureman (~pressurem@62.217.45.26) has joined #ceph
[14:51] * cronix1 (~cronix@5.199.139.166) Quit (Ping timeout: 480 seconds)
[14:51] <magicrobotmonkey> is there a way to clear out all my monitors but leave the osds intact?
[14:52] <magicrobotmonkey> I've made a huge mistake
[14:54] <cookednoodles> clear them out ?
[14:54] * KevinPerks (~Adium@CPE602ad091f0ad-CM602ad091f0aa.cpe.net.cable.rogers.com) has joined #ceph
[14:55] <magicrobotmonkey> yea i cant get them to start because paxos is freaking out
[14:55] <magicrobotmonkey> i've tried unintstalling one and bringing it up as a solo monitor
[14:56] <magicrobotmonkey> but im just getting tons of mon.george35@0(leader).paxos(paxos active c 1..11) is_readable now=2014-06-26 08:56:22.714565 lease_expire=0.000000 has v0 lc 11
[14:56] <magicrobotmonkey> in the log
[14:57] * rudolfsteiner (~federicon@181.21.134.151) Quit (Quit: rudolfsteiner)
[14:57] * mo- (~mo@2a01:4f8:141:3264::3) has joined #ceph
[15:01] * michalefty (~micha@p20030071CF5304002DC9E382D2950321.dip0.t-ipconnect.de) Quit (Ping timeout: 480 seconds)
[15:02] * huangjun (~kvirc@117.151.51.41) has joined #ceph
[15:02] <mo-> is there a way to use ceph-deploy prepare without it adding the osd to the cluster and starting the service?
[15:03] <alfredodeza> mo-: that is what activate does
[15:03] <alfredodeza> so, yes
[15:03] <mo-> I assumed as much
[15:03] <alfredodeza> you can 'prepare' and just not do 'activate'
[15:03] <mo-> yea thing is, it does activate it too
[15:03] <alfredodeza> what
[15:03] * alfredodeza checks
[15:05] <alfredodeza> from what I am reading it most certainly does not start the service
[15:06] <mo-> I used "ceph-deploy osd prepare locceph1c:sdd" and... well it doesn't actually look like its invoking the activation commands, but at the end it looks like this
[15:06] <alfredodeza> ah yes, show me the output!
[15:06] <mo-> [locceph1c][INFO ] checking OSD status...
[15:06] <mo-> [locceph1c][INFO ] Running command: ceph --cluster=ceph osd stat --format=json
[15:06] <mo-> [locceph1c][WARNIN] there is 1 OSD down
[15:06] <alfredodeza> aha
[15:06] <mo-> because it actually HAS added the osd to the cluster
[15:06] <alfredodeza> that is not starting the service
[15:07] <alfredodeza> that is just checking what is the status of the OSD
[15:07] <alfredodeza> wait, you said 'starting the service' not 'adding to the cluster'
[15:07] <mo-> I said both initially
[15:07] <mo-> or "either" wouldve been more precise
[15:08] <alfredodeza> ah yes you did! my mistake
[15:08] * alfredodeza read too fast
[15:08] <mo-> funny thing is
[15:08] <alfredodeza> so definitely nothing starts it
[15:08] <mo-> if I export the crushmap, osd.2 is not in there
[15:08] <mo-> but in "ceph osd tree"
[15:08] <mo-> it shows up under the last node, and even says "up"
[15:08] <alfredodeza> so this would be ceph-disk-prepare that is doing this cluster talk
[15:08] <alfredodeza> not ceph-deploy
[15:09] <alfredodeza> if you look at your ceph-deploy output there should be a line that says that is running ceph-disk-prepare on the remote end, yes?
[15:09] <mo-> suppose youre right
[15:09] * alfredodeza looks at ceph-disk-prepare
[15:09] <mo-> yes it did
[15:09] <mo-> [locceph1c][INFO ] Running command: ceph-disk-prepare --fs-type xfs --cluster ceph -- /dev/sdd
[15:09] <alfredodeza> there is nothing in ceph-deploy that I can see it adding it to the cluster
[15:10] <alfredodeza> so I am suspicious about ceph-disk-prepare
[15:10] <alfredodeza> one sec while I go read this
[15:10] <mo-> it doesnt output anything indicating ceph-deploy might do it
[15:10] <mo-> wait a sec
[15:10] <mo-> that disk actually had been in the cluster before (just experimenting here), but I properly used zap and rmdir'd its folder in var/lib/ceph/osd
[15:10] <mo-> maybe thats causing it to act up?
[15:10] <alfredodeza> oh
[15:11] <alfredodeza> that could totally be it
[15:11] <alfredodeza> that is not the way to remove an OSD by the way :)
[15:11] <mo-> its not? I used ceph osd rm and ceph osd crush rm and ceph auth del
[15:11] <mo-> then I zapped it and removed the dir
[15:11] <mo-> oh and umounted it
[15:12] <mo-> did I miss anything?
[15:12] <alfredodeza> ah so you did used ceph
[15:12] <mo-> yea
[15:12] <alfredodeza> I read that as just plain rm
[15:12] <mo-> nono, went through all that
[15:12] <alfredodeza> ok
[15:12] <alfredodeza> I am replicating this on a new box
[15:12] <alfredodeza> one sec
[15:12] <mo-> which begs the question, why is it doing that and where is it getting this "hey this used to be here" data
[15:13] <mo-> basically I am trying to get a cache pool to work
[15:13] * michalefty (~micha@p20030071CF513B002DC9E382D2950321.dip0.t-ipconnect.de) has joined #ceph
[15:13] <mo-> which is why I wanted to prepare new OSDs on the same hosts, and then manually add them to the crush map under a different root
[15:13] <mo-> but ceph adding these osds to the existing root kinda breaks that
[15:15] * AfC (~andrew@2001:44b8:31cb:d400:6e88:14ff:fe33:2a9c) Quit (Quit: Leaving.)
[15:16] * dlan (~dennis@116.228.88.131) Quit (Ping timeout: 480 seconds)
[15:21] <alfredodeza> hrmn
[15:21] <alfredodeza> I just replicated this
[15:21] <mo-> so youre seing the same behaviour?
[15:23] * JayJ (~jayj@157.130.21.226) has joined #ceph
[15:24] <alfredodeza> yes
[15:24] * odyssey4me_ (~odyssey4m@165.233.71.2) Quit (Ping timeout: 480 seconds)
[15:24] <mo-> at least its not me doing something wrong then
[15:26] <Vacum_> doesn't an osd always add itself to the host in the crushmap it is started at? (so you can even "move" an osd by just putting it to a different machine)
[15:26] <mo-> its doing something even weirder
[15:26] <mo-> like I said, it does show up in the osd tree
[15:26] <mo-> but it doesnt show up if you extract the crushmap
[15:26] <Vacum_> oh
[15:27] <alfredodeza> ceph-disk is (currently) a bit obscure on what is doing, let me try with a modified version that tells us exactly what is calling
[15:27] * rudolfsteiner (~federicon@181.21.148.181) has joined #ceph
[15:28] * sjm (~sjm@174.47.140.2) Quit (Read error: Operation timed out)
[15:30] <alfredodeza> I can't see what ceph-disk-prepare would be doing to add it to the cluster according to this output mo- http://fpaste.org/113451/37893841/
[15:30] * michalefty (~micha@p20030071CF513B002DC9E382D2950321.dip0.t-ipconnect.de) Quit (Quit: Leaving.)
[15:31] <alfredodeza> my last suspicion would be that osd tree looks in the /var/lib/osd dir for osds
[15:32] <mo-> would certainly explain the discrepancy between the tree and the crushmap
[15:32] <alfredodeza> but I don't feel authoritative on this, might want to ask on the mailing list
[15:33] <mo-> also hm
[15:33] <mo-> it cant be only the tree though
[15:33] <mo-> because the osd is actually running now
[15:34] <magicrobotmonkey> can anyone help me interpret this line: 7fede0763700 0 -- :/1040921 >> 172.16.17.55:6789/0 pipe(0x7feddc022470 sd=3 :0 s=1 pgs=0 cs=0 l=1 c=0x7feddc0226e0).fault
[15:34] * longguang (~chatzilla@123.126.33.253) Quit (Ping timeout: 480 seconds)
[15:34] <mo-> the machine youre on is unable to contact the monitor at 172.16.17.55
[15:35] <magicrobotmonkey> hmm
[15:35] <magicrobotmonkey> but then it comes up and the monmap includes it
[15:35] <mo-> alfredodeza: does the osd get started for you as well? maybe you can see something in its log, because Im kinda not
[15:37] <mo-> magicrobotmonkey: monitors sometimes take a little while when being started because they answer requests
[15:37] <magicrobotmonkey> ok thanks
[15:37] <mo-> err
[15:37] <mo-> *before
[15:37] <magicrobotmonkey> it seems to have cleared itself up
[15:40] <mo-> ima create the new root in the crushmap, see if that helps
[15:40] * brad_mssw (~brad@shop.monetra.com) has joined #ceph
[15:43] * lucas1 (~Thunderbi@222.247.57.50) Quit (Quit: lucas1)
[15:48] * vbellur (~vijay@122.172.247.85) has joined #ceph
[15:48] * sz0 (~sz0@178.240.141.248) Quit (Remote host closed the connection)
[15:49] * cronix1 (~cronix@5.199.139.166) has joined #ceph
[15:49] * markbby (~Adium@168.94.245.1) has joined #ceph
[15:50] * lucas1 (~Thunderbi@222.240.148.154) has joined #ceph
[15:50] * lucas1 (~Thunderbi@222.240.148.154) Quit (Remote host closed the connection)
[15:51] * rotbeard (~redbeard@b2b-94-79-138-170.unitymedia.biz) Quit (Quit: Leaving)
[15:52] * bitserker (~toni@63.pool85-52-240.static.orange.es) Quit (Read error: Operation timed out)
[15:55] <mo-> nope
[15:55] <kraken> http://i.imgur.com/xKYs9.gif
[15:58] * cronix1 (~cronix@5.199.139.166) Quit (Ping timeout: 480 seconds)
[15:58] * jrankin (~jrankin@d47-69-66-231.try.wideopenwest.com) has joined #ceph
[16:00] * dmsimard_away is now known as dmsimard
[16:01] * mwyatt (~mwyatt@137.152.127.16) has joined #ceph
[16:02] * dmsimard is now known as dmsimard_away
[16:02] * allig8r (~allig8r@128.135.219.116) Quit (Remote host closed the connection)
[16:03] * allig8r (~allig8r@128.135.219.116) has joined #ceph
[16:03] * ScOut3R (~ScOut3R@catv-80-99-64-8.catv.broadband.hu) Quit (Ping timeout: 480 seconds)
[16:03] * ScOut3R (~ScOut3R@catv-80-99-64-8.catv.broadband.hu) has joined #ceph
[16:08] * dmsimard_away (~dmsimard@198.72.123.142) Quit (Quit: Signed off)
[16:11] * dmsimard (~dmsimard@198.72.123.202) has joined #ceph
[16:16] <mwyatt> Morning all. New Ceph user here. I have a concept/best practice question about pg_num for each pool. Just built a 4 node test cluster. 3 are mon, all 4 osd. Each OSD node with 6 OSDs. Based on documentation, pg_num in my case should be (24osds*100)/3repcopies=800 for a single pool of objects. Round to 1024 assuming a single pool. My questions are: What about multiple pools in a cluster? Should
[16:16] <mwyatt> I set every pool to 1024 pgs or should I consider running fewer pgs in some pools and more in others so my sum total # of pgs is 1024 in this case? Is the only impact CPU/mem related on the OSD hosts? Thanks!
[16:20] <mo-> alfredodeza: this is weird as hell. but I think I got it to actually do what I wanted
[16:21] <mo-> I manually added the OSDs to the crushmap without them existing yet (in the different root), then ceph-deploy prepare
[16:21] <mo-> and that actually started up the OSDs correctly
[16:21] * jordanP (~jordan@185.23.92.11) has joined #ceph
[16:21] <mo-> whereas "activate" had moved the OSD back over to the other root which is totally not what I wanted
[16:24] * ikrstic (~ikrstic@c82-214-88-26.loc.akton.net) Quit (Quit: Konversation terminated!)
[16:24] * rudolfsteiner (~federicon@181.21.148.181) Quit (Quit: rudolfsteiner)
[16:31] <mo-> so something would definitely appear to be amiss, but despite the non-existant documentation I seem to have been able to put it together
[16:32] * jordanP (~jordan@185.23.92.11) Quit (Quit: Leaving)
[16:33] <brad_mssw> why, when restarting a node, does the OSD attempt to rewrite the crush map?
[16:34] <brad_mssw> it tries to reassign the osds it owns to another root or something
[16:34] <mo-> yea
[16:34] * benjamin (~bdstalcup@174.46.155.201) has joined #ceph
[16:34] <mo-> I guess. probably one of the quirks of having more than one 'host' on a physical machine
[16:35] <brad_mssw> I know you can add the osd_crush_update_on_start = false to the ceph.conf file, but it seems odd
[16:35] <brad_mssw> that you would need to
[16:35] <brad_mssw> mo-: are you saying if they weren't dual-purpose OSD + MON nodes, it wouldn't happen?
[16:35] <mo-> nono
[16:35] <mo-> those are within one host bucket
[16:36] <mo-> but if you have 2 buckets within the crushmap that are actually the same physical system ceph something seems to be amiss
[16:36] <brad_mssw> oh, I see what you're saying ... yes, true, I have 2 host buckets as I have 2 OSDs, one spindle one ssd
[16:36] <brad_mssw> per physical host
[16:36] <mo-> I guess youre currently experimenting with the same thing as I am
[16:37] <brad_mssw> hah, ok, I guess so ;)
[16:37] <mo-> so I havent dared to restart my setup yet. youre saying thatd relocate the OSDs?
[16:37] <mo-> i.e. I definitely need to set the ceph.conf option you just named
[16:37] <brad_mssw> yes, at least in my experience
[16:38] <brad_mssw> as soon as you restart one node, things go super wonky without that config
[16:38] <mo-> thats no great
[16:38] <brad_mssw> but maybe I was doing something else wrong, I don't know ...
[16:38] <mo-> well
[16:38] * jordanP (~jordan@185.23.92.11) has joined #ceph
[16:38] <mo-> I am sitting on a completely virtual test setup here. ima just save a crushmap and do the reboot, see what happens
[16:39] <brad_mssw> just rebooting one node will show it
[16:39] <mo-> lets see
[16:39] <brad_mssw> restoring the crush map after the reboot brings everything else back to normal
[16:39] * xarses (~andreww@c-24-23-183-44.hsd1.ca.comcast.net) has joined #ceph
[16:39] <pressureman> What is the recommended IO scheduler these days for running OSDs on XFS (HDD)? Debian still defaults to CFQ, but I think recent Ubuntu releases now default to deadline, even on HDDs
[16:39] <mo-> I suppose ceph is being weirded out because a linux system only has 1 hostname (despite being reachable under more names)
[16:40] * Pedras (~Adium@50.185.218.255) Quit (Quit: Leaving.)
[16:40] <mo-> wow brad_mssw youre right, it does relocate them
[16:40] * Pedras (~Adium@216.207.42.140) has joined #ceph
[16:41] <brad_mssw> pressureman: everything I've read appears to recommend deadline these days
[16:41] * ade (~abradshaw@193.202.255.218) Quit (Quit: Too sexy for his shirt)
[16:41] <brad_mssw> pressureman: and if you're using ssds, then noop is also an option I've seen recommended
[16:41] <mo-> whats worse is that apparently it instantly starts peering
[16:42] <brad_mssw> mo-: yeah, I wasn't monitoring my reboots and I rebooted 2 machines before I realized then my cluster was completely bricked
[16:42] <pressureman> brad_mssw, I'm running numerous VMs on RBD (qemu + librbd), so I expect the actual IO that the OSDs are seeing is quite to be quite random...
[16:42] <pressureman> my gut feeling was that i should take a long hard look at deadline scheduler
[16:43] <pressureman> i've heard that CFQ performance starts to suffer, once the number of queues goes beyond a certain number - whereas deadline can maintain more linear performance in such a situation
[16:44] <pressureman> (i wish were were running on SSDs, but this cluster hasn't been pimped out to that level so far)
[16:45] <brad_mssw> i'm still in testing myself so I'm only running on a 1Gbps network, so can't really benchmark yet
[16:45] <brad_mssw> should have 10Gbps gear in the next 30 days
[16:46] <mo-> I am also going to do some benchmarking next week on a 11GBit networking cluster
[16:46] <pressureman> this is also only a bonded 2 x 1Gbps setup... i will be setting up a cluster on QDR IB soon though (IPoIB initially, possibly test RDMA / Xio if it's ready by then)
[16:47] * Kioob`Taff (~plug-oliv@local.plusdinfo.com) Quit (Quit: Leaving.)
[16:47] <mo-> quick intermission: brad_mssw your ceph.conf thing does the trick, makes the setup reboot-safe
[16:47] <brad_mssw> mo-: cool, glad its not only me that needs that ;)
[16:47] <brad_mssw> makes me feel better that I'm not doing something wrong
[16:48] <mo-> still... I wouldnt feel great if I were to use that productively
[16:48] <mo-> it would seem like this cache pool thing isnt quite ready just yet
[16:49] <mo-> (unless youd have different nodes for both platter and ssd nodes)
[16:49] <mo-> maybe we should take that to the mailing list?
[16:49] * Pedras (~Adium@216.207.42.140) Quit (Read error: Connection reset by peer)
[16:49] * KaZeR (~kazer@64.201.252.132) has joined #ceph
[16:49] * primechuck (~primechuc@86.sub-70-197-192.myvzw.com) has joined #ceph
[16:51] * rdas (~rdas@121.244.87.115) Quit (Quit: Leaving)
[16:51] * Pedras (~Adium@216.207.42.140) has joined #ceph
[16:54] * bkopilov (~bkopilov@213.57.16.12) has joined #ceph
[16:55] * bitserker (~toni@238.56.79.188.dynamic.jazztel.es) has joined #ceph
[16:58] <brad_mssw> mo-: maybe, that said, I think I found that config option off the mailing list :/
[16:58] * sverrest (~sverrest@cm-84.208.166.184.getinternet.no) Quit (Read error: Connection reset by peer)
[16:58] <mo-> oh? well one should be able to find that back
[16:58] <brad_mssw> mo-: so it appears to be known, just don't have a clear explanation why it tries to rewrite it
[16:58] <mo-> well I would wager theres an underlying issue
[16:59] * sverrest (~sverrest@cm-84.208.166.184.getinternet.no) has joined #ceph
[16:59] <brad_mssw> mo-: i'd agree, I can't see how this would be intended behavior
[16:59] * gregmark (~Adium@cet-nat-254.ndceast.pa.bo.comcast.net) has joined #ceph
[16:59] <mo-> getting sidetracked by clock skew.. damn slow as molasses hypervisors
[17:00] * sjm (~sjm@cloudgate.cs.utsa.edu) has joined #ceph
[17:02] * BManojlovic (~steki@91.195.39.5) Quit (Ping timeout: 480 seconds)
[17:03] * madkiss (~madkiss@nat.nue.novell.com) Quit (Ping timeout: 480 seconds)
[17:05] * adamcrume (~quassel@c-71-204-162-10.hsd1.ca.comcast.net) has joined #ceph
[17:07] <mo-> ah brad_mssw found it: http://tracker.ceph.com/issues/6227
[17:07] <huangjun> we migrated vmware esxi vm (250GB) from ceph cluster to local disk but the mirgation rate is pretty slow, about 20MB/s for 8 OSDs
[17:08] <mo-> so I think this is still being listed as an untreated feature request
[17:08] * ScOut3R (~ScOut3R@catv-80-99-64-8.catv.broadband.hu) Quit (Read error: Operation timed out)
[17:10] <mo-> additionally, I assume you would probably want to set "noout" before adding SSDs to your setup to avoid ceph messing things up before you set the crush map right
[17:11] <brad_mssw> noout?
[17:11] <mo-> so the cluster doesnt rebalance data
[17:12] <mo-> when it temporarily gets new OSDs
[17:12] * kevinc (~kevinc__@client65-78.sdsc.edu) has joined #ceph
[17:12] <brad_mssw> mo-: oh, were's that config?
[17:12] <mo-> you set that at run time
[17:12] <mo-> "ceph osd set noout"
[17:12] <mo-> although hm wait a second
[17:13] <mo-> that only doesnt mark OSDs as out, not sure how itd react to NEW osds in that case
[17:17] * thomnico (~thomnico@2a01:e35:8b41:120:1c11:9e1d:44e4:a62c) Quit (Ping timeout: 480 seconds)
[17:17] * thomnico (~thomnico@cro38-2-88-180-16-18.fbx.proxad.net) has joined #ceph
[17:18] <mo-> do I feel adventurous enough to go and try that too?
[17:20] * bjornar (~bjornar@ti0099a430-0730.bb.online.no) has joined #ceph
[17:22] <magicrobotmonkey> can anyone help me get multipart uploads via radogw working with an erasure coded pool?
[17:22] <magicrobotmonkey> I've got the extra bucket set up
[17:23] <magicrobotmonkey> but im getting 500s
[17:23] * b0e (~aledermue@x2f336a2.dyn.telefonica.de) has joined #ceph
[17:24] <tnt_> Is there a release date for the next firefly maintenance release ? There seem to be quite a few fix waiting backport
[17:25] * lalatenduM (~lalatendu@209.132.188.8) Quit (Quit: Leaving)
[17:26] * AbyssOne is now known as a1-away
[17:27] <mo-> fun fact: apparently the VM deadlocked because of vmware tools after trying to map an rbd in
[17:27] * fsimonce (~simon@host29-76-dynamic.9-79-r.retail.telecomitalia.it) Quit (Quit: Coyote finally caught me)
[17:28] * xarses (~andreww@c-24-23-183-44.hsd1.ca.comcast.net) Quit (Read error: Operation timed out)
[17:30] * narb (~Jeff@38.99.52.10) has joined #ceph
[17:34] * dneary (~dneary@87-231-145-225.rev.numericable.fr) Quit (Read error: Operation timed out)
[17:39] * analbeard (~shw@support.memset.com) Quit (Quit: Leaving.)
[17:45] * dneary (~dneary@87-231-145-225.rev.numericable.fr) has joined #ceph
[17:49] * lpabon (~quassel@66-189-8-115.dhcp.oxfr.ma.charter.com) has joined #ceph
[17:49] <johnfoo> yeah so. the cluster is back being busy permanently repairing itself
[17:49] <johnfoo> apparently it was just a temporary fix
[17:49] * lpabon (~quassel@66-189-8-115.dhcp.oxfr.ma.charter.com) Quit ()
[17:50] * gregsfortytwo1 (~Adium@cpe-107-184-64-126.socal.res.rr.com) Quit (Quit: Leaving.)
[17:51] * b0e (~aledermue@x2f336a2.dyn.telefonica.de) Quit (Quit: Leaving.)
[17:53] * lpabon (~lpabon@nat-pool-bos-t.redhat.com) has joined #ceph
[17:53] <Pauline> mo-: 'ceph osd set noin"
[17:54] * xarses (~andreww@12.164.168.117) has joined #ceph
[17:55] <mo-> perfect thanks a lot
[17:56] * davidzlap (~Adium@ip68-4-173-198.oc.oc.cox.net) has joined #ceph
[17:59] * koleosfuscus (~koleosfus@ws11-189.unine.ch) has left #ceph
[18:00] * angdraug (~angdraug@12.164.168.117) has joined #ceph
[18:00] * joao|lap (~JL@209-252-179-57.ip.mcleodusa.net) Quit (Read error: Operation timed out)
[18:01] * hasues (~hazuez@adsl-74-178-236-65.jax.bellsouth.net) has joined #ceph
[18:02] <mo-> libceph: mon0 192.168.178.41:6789 feature set mismatch, my 4a042a42 < server's 84a042a42, missing 800000000
[18:02] <mo-> anybody know what feature bit that is?
[18:04] * sherlock (sherlock@d.clients.kiwiirc.com) has joined #ceph
[18:06] <gleam> CEPH_FEATURE_OSD_CACHEPOOL
[18:06] <gleam> http://cephnotes.ksperis.com/blog/2014/01/21/feature-set-mismatch-error-on-ceph-kernel-client
[18:07] <mo-> so Id need 3.14+ to use cache pools?
[18:08] <mo-> wish Id known that earlier, means its very much pointless atm :/
[18:08] * reed (~reed@75-101-54-131.dsl.static.sonic.net) has joined #ceph
[18:08] <gleam> yeah if you need to use krbd
[18:08] <mo-> oh. right suppose I actually dont on prodution use
[18:08] <mo-> just wanted to use it for testing
[18:08] <gleam> i think most people are using librbd with it w/radosgw or libvirt
[18:10] <mo-> yea qemu uses librbd
[18:10] <mo-> didnt think about that, my bad
[18:11] <mo-> but, I guess cache pools / tiering isnt as transparent to the client as I thought it would be
[18:13] * mikedawson (~chatzilla@23-25-46-97-static.hfc.comcastbusiness.net) Quit (Read error: Connection reset by peer)
[18:14] * joef (~Adium@2620:79:0:131:1163:c0b3:ed53:725b) has joined #ceph
[18:14] * rwheeler (~rwheeler@nat-pool-bos-u.redhat.com) Quit (Quit: Leaving)
[18:14] <gleam> i think the issue is that it sends the client a redirect rather than doing hte redirect internally
[18:16] <mo-> since clients are supposed to do their IO directly, I suppose that makes sense
[18:16] * alram (~alram@cpe-76-167-62-129.socal.res.rr.com) has joined #ceph
[18:17] <sherlock> Hello. I am new to CEPH and reading about it. We are trying to use Ceph with OpenStack. Can we have Ceph block and object storage on the same cluster?
[18:20] <gleam> yes, sherlock
[18:20] <gleam> they're just different pools
[18:20] <gleam> so for instance you could make the block storage pool all-ssd and the object storage pool erasure coded on spinners w/an ssd cache tier in front
[18:20] <gleam> or whatever.
[18:21] <gleam> but all in the same cluster
[18:22] <sherlock> Oh ok...So I need different monitors for each pool of such OSDs?
[18:22] <gleam> nope
[18:22] <kraken> http://i.minus.com/iUgVCKwjISSke.gif
[18:23] <janos_> monitors are cluster-wide
[18:23] <gleam> the pools are logical, multiple pools can coexist on the same group of osds under the same mon
[18:24] <sherlock> Ah got it.. Thanks.
[18:25] * sputnik13 (~sputnik13@207.8.121.241) has joined #ceph
[18:26] * diegows (~diegows@host-216-57-132-113.customer.veroxity.net) Quit (Ping timeout: 480 seconds)
[18:27] * sputnik13 (~sputnik13@207.8.121.241) Quit ()
[18:28] * sputnik13 (~sputnik13@207.8.121.241) has joined #ceph
[18:31] * kevinc (~kevinc__@client65-78.sdsc.edu) Quit (Quit: This computer has gone to sleep)
[18:31] * kevinc (~kevinc__@client65-78.sdsc.edu) has joined #ceph
[18:32] * hasues (~hazuez@adsl-74-178-236-65.jax.bellsouth.net) Quit (Ping timeout: 480 seconds)
[18:34] * sjm (~sjm@cloudgate.cs.utsa.edu) Quit (Ping timeout: 480 seconds)
[18:38] * reistin (~reistin@5.141.192.250) has joined #ceph
[18:38] <reistin> Hello all! How we can say to ceph-deploy to use upstart system instead of sysvinit?
[18:39] * diegows (~diegows@190.190.5.238) has joined #ceph
[18:39] <alfredodeza> reistin: that is automatically detected
[18:39] <alfredodeza> you cannot override it
[18:39] <alfredodeza> is it detecting it wrong?
[18:41] <reistin> <alfredodeza> hi)
[18:43] <reistin> we have a cluster with custom name, and then we want add new osd (ceph-deploy osd create) ceph-deploy don't want launch /usr/bin/ceph-osd program
[18:43] <alfredodeza> reistin: to add an osd you should not use create
[18:43] <alfredodeza> you should use `ceph-deploy osd add`
[18:43] * Kioob (~kioob@2a01:e34:ec0a:c0f0:21e:8cff:fe07:45b6) has joined #ceph
[18:44] <alfredodeza> err
[18:44] <alfredodeza> hrmn
[18:44] <brad_mssw> hmm, I tried removing 2 osds and removed the crush map entries too .... but I'm showing 128 active+degraded now, doesn't seem to be redistributing, etc
[18:44] <alfredodeza> nevermind
[18:44] <alfredodeza> got it wrong
[18:44] <brad_mssw> any idea where to check why ?
[18:44] <reistin> try wih add?
[18:45] <alfredodeza> no no, sorry I got confused with monitors :)
[18:45] <reistin> )
[18:45] <alfredodeza> so back to your issue is that you don't want to launch the osd?
[18:45] <alfredodeza> you might want to just prepare it then
[18:45] <reistin> now a try to add manual
[18:47] <reistin> we have so many problems with custom names)
[18:51] <brad_mssw> ugh, actually, I think it is because the default 'data' and 'metadata' pools are replica3, nevermind
[18:52] <brad_mssw> even though I'm not using them
[18:52] <reistin> If we make OSD manualy, we have to add /dev/disk to fstab manualy too?
[18:57] * koleosfuscus (~koleosfus@adsl-84-226-68-69.adslplus.ch) has joined #ceph
[18:58] * clfh (uid26720@id-26720.uxbridge.irccloud.com) Quit (Quit: Connection closed for inactivity)
[19:00] * cronix1 (~cronix@5.199.139.166) has joined #ceph
[19:03] <mo-> to add to that, how does ceph even know which osd corresponds to which UUID (disk)
[19:04] * jordanP (~jordan@185.23.92.11) Quit (Remote host closed the connection)
[19:04] <mo-> since its not doing fstab entries anymore
[19:04] * lpabon (~lpabon@nat-pool-bos-t.redhat.com) Quit (Quit: ZNC - http://znc.in)
[19:04] * mwyatt1 (~mwyatt@ip70-162-100-189.ph.ph.cox.net) has joined #ceph
[19:06] * lpabon (~lpabon@nat-pool-bos-t.redhat.com) has joined #ceph
[19:07] <reistin> if we create OSD manual and mount it manual(mount -o user_xattr /dev/{hdd} /var/lib/ceph/osd/ceph-{osd-number}) how it can mount automatic after reboot?
[19:07] <johnfoo> use /etc/fstab ?
[19:08] * cronix1 (~cronix@5.199.139.166) Quit (Ping timeout: 480 seconds)
[19:08] <reistin> so we have add to fstab?
[19:08] <johnfoo> sure
[19:08] * alexbligh1 (~alexbligh@89-16-176-215.no-reverse-dns-set.bytemark.co.uk) Quit (Quit: Terminated with extreme prejudice - dircproxy 1.0.5)
[19:08] * thomnico (~thomnico@cro38-2-88-180-16-18.fbx.proxad.net) Quit (Quit: Ex-Chat)
[19:09] * alexbligh1 (~alexbligh@89-16-176-215.no-reverse-dns-set.bytemark.co.uk) has joined #ceph
[19:09] <reistin> johnfoo, thank you
[19:09] <johnfoo> np
[19:10] * WhIteSidE (~chatzilla@31.6.14.157) has joined #ceph
[19:11] * joao|lap (~JL@2607:f298:a:607:cce3:2c52:47cd:a80e) has joined #ceph
[19:11] * ChanServ sets mode +o joao|lap
[19:11] * rturk|afk is now known as rturk
[19:13] * sjusthm (~sam@66-214-251-229.dhcp.gldl.ca.charter.com) has joined #ceph
[19:14] <WhIteSidE> Hello Cephers: I have a somewhat stuck cluster right now. I wrote a lot of data to rgw, to the exent that I have a full OSD. I have a rgw bucket that I can delete to free a lot of space, but whenever I try to run a rgw command, I get an error message.
[19:14] <WhIteSidE> 2014-06-26 13:07:53.244839 7f48634a7780 0 client.5475.objecter FULL, paused modify 0x11d7a80 tid 8
[19:14] <WhIteSidE> Any pointers on how I can free up some space by dropping the offending bucket?
[19:14] <mo-> when an OSD is full, ceph will block all writes
[19:15] <mo-> easiest way to fix that I found to temporarily increase the osd full ratio to .97
[19:15] <WhIteSidE> And then I should be able to drop the offending bucket?
[19:15] <mo-> just makes sure theres no blocked writes that would fill it right back up
[19:15] <mo-> yea
[19:15] <WhIteSidE> @mo-: How would I check the write queue?
[19:15] <cephalobot> WhIteSidE: Error: "mo-:" is not a valid command.
[19:15] <WhIteSidE> How would I check the write queue?
[19:15] <WhIteSidE> (Also, thanks for responding)
[19:16] <mo-> the f was is that bot trying to match there, haha
[19:16] <johnfoo> the @
[19:16] <WhIteSidE> Yep
[19:16] <mo-> cant help you with write queues im afraid, never touched rgw
[19:16] <WhIteSidE> Hmm, okay
[19:16] <WhIteSidE> I have no critical data in rgw (I was just testing it)
[19:17] <WhIteSidE> Is it safe(ish) to simply stop rgw and remove the pools directly from ceph?
[19:17] <WhIteSidE> (the rgw pools, that is)
[19:17] <WhIteSidE> I realize I'll lose all of my rgw buckets
[19:17] <WhIteSidE> But there's nothing there I cannot afford to lose.
[19:18] <johnfoo> WhIteSidE: you can kill rgw and purge the pools if you want to get rid of the data
[19:18] <johnfoo> it'll work alright
[19:18] <johnfoo> data will be lost but eh
[19:18] <johnfoo> data is overrated anyway
[19:18] <WhIteSidE> @johnfoo: But only the data in the rgw pools I remove, correct?
[19:18] <cephalobot> WhIteSidE: Error: "johnfoo:" is not a valid command.
[19:18] <johnfoo> yes
[19:19] <WhIteSidE> Okay, no more at symbol for me
[19:19] <johnfoo> you delete a pool, the data in that pool is gone
[19:19] <WhIteSidE> Thanks a lot
[19:19] <mo-> irc style to ping people is just the name, possibly a :, just ommit the @
[19:20] <WhIteSidE> mo-: Thanks
[19:20] <WhIteSidE> Hmm, I swear on my desktop it lets me use an at symbol
[19:21] <WhIteSidE> But of course this happens while I'm traveling on a laptop without an IRC client :)
[19:21] <mo-> its not that you cant use it, its just that the bot gets triggered by it ;)
[19:24] * mwyatt2 (~mwyatt@2600:1011:b12d:5d9d:0:9:e6f3:4401) has joined #ceph
[19:26] * ScOut3R (~ScOut3R@5401C4E7.dsl.pool.telekom.hu) has joined #ceph
[19:26] <WhIteSidE> Thanks very much for the help guys
[19:26] <WhIteSidE> I'm out of the woods and now I can add more nodes
[19:27] * rweeks (~rweeks@192.169.20.75.static.etheric.net) has joined #ceph
[19:28] * mwyatt1 (~mwyatt@ip70-162-100-189.ph.ph.cox.net) Quit (Ping timeout: 480 seconds)
[19:32] * kevinc (~kevinc__@client65-78.sdsc.edu) Quit (Quit: This computer has gone to sleep)
[19:32] * ScOut3R (~ScOut3R@5401C4E7.dsl.pool.telekom.hu) Quit (Remote host closed the connection)
[19:33] * ScOut3R (~ScOut3R@5401C4E7.dsl.pool.telekom.hu) has joined #ceph
[19:39] * kevinc (~kevinc__@client65-78.sdsc.edu) has joined #ceph
[19:41] * ScOut3R (~ScOut3R@5401C4E7.dsl.pool.telekom.hu) Quit (Ping timeout: 480 seconds)
[19:43] * brytown (~brytown@2620:79:0:8204:2867:3b82:3bf6:2c99) has joined #ceph
[19:44] * brytown (~brytown@2620:79:0:8204:2867:3b82:3bf6:2c99) has left #ceph
[19:45] * koleosfuscus (~koleosfus@adsl-84-226-68-69.adslplus.ch) Quit (Quit: koleosfuscus)
[19:52] * sjm (~sjm@cloudgate.cs.utsa.edu) has joined #ceph
[19:53] * ScOut3R (~ScOut3R@5401C4E7.dsl.pool.telekom.hu) has joined #ceph
[20:09] <magicrobotmonkey> can someone help me parse a monitor fault line?
[20:10] * danieagle (~Daniel@179.184.165.184.static.gvt.net.br) has joined #ceph
[20:10] <magicrobotmonkey> im getting intermittent faults on 2/5 monitors
[20:10] <magicrobotmonkey> and I'm having a hard time hunting down why
[20:11] * KaZeR (~kazer@64.201.252.132) Quit (Ping timeout: 480 seconds)
[20:11] * Nacer_ (~Nacer@252-87-190-213.intermediasud.com) has joined #ceph
[20:11] * Nacer_ (~Nacer@252-87-190-213.intermediasud.com) Quit (Remote host closed the connection)
[20:12] * koleosfuscus (~koleosfus@adsl-84-226-68-69.adslplus.ch) has joined #ceph
[20:14] * LeaChim (~LeaChim@host86-174-77-240.range86-174.btcentralplus.com) has joined #ceph
[20:15] * sjm (~sjm@cloudgate.cs.utsa.edu) Quit (Read error: Operation timed out)
[20:15] * KaZeR (~kazer@64.201.252.132) has joined #ceph
[20:17] * Nacer (~Nacer@252-87-190-213.intermediasud.com) Quit (Read error: Operation timed out)
[20:20] <mo-> anonymize the IPs (if theyre routed), possibly hostnames and put it on pastebin or something like that
[20:20] * tracphil (~tracphil@130.14.71.217) has joined #ceph
[20:20] * ikrstic (~ikrstic@178-222-94-242.dynamic.isp.telekom.rs) has joined #ceph
[20:21] * Cube (~Cube@66-87-130-18.pools.spcsdns.net) has joined #ceph
[20:21] * hasues (~hazuez@adsl-74-178-236-65.jax.bellsouth.net) has joined #ceph
[20:22] * WhIteSidE (~chatzilla@31.6.14.157) Quit (Ping timeout: 480 seconds)
[20:24] * nljmo (~nljmo@64.125.103.162) has joined #ceph
[20:27] * vilobhmm (~vilobhmm@nat-dip27-wl-a.cfw-a-gci.corp.yahoo.com) has joined #ceph
[20:31] * reistin (~reistin@5.141.192.250) Quit ()
[20:32] * sjm (~sjm@cloudgate.cs.utsa.edu) has joined #ceph
[20:32] * sarob (~sarob@2001:4998:effd:600:1d66:df75:6caf:32b) has joined #ceph
[20:32] * rendar (~I@host117-179-dynamic.56-79-r.retail.telecomitalia.it) Quit (Ping timeout: 480 seconds)
[20:34] * mwyatt2 (~mwyatt@2600:1011:b12d:5d9d:0:9:e6f3:4401) Quit (Quit: Bye)
[20:34] * mwyatt1 (~mwyatt@2600:1011:b12d:5d9d:0:9:e6f3:4401) has joined #ceph
[20:35] * rendar (~I@host117-179-dynamic.56-79-r.retail.telecomitalia.it) has joined #ceph
[20:36] * cronix1 (~cronix@5.199.139.166) has joined #ceph
[20:39] * analbeard (~shw@5.153.255.234) has joined #ceph
[20:40] * sjm (~sjm@cloudgate.cs.utsa.edu) Quit (Read error: Operation timed out)
[20:41] * rturk is now known as rturk|afk
[20:41] <vilobhmm> joshd : ping
[20:42] <vilobhmm> jdurgin : ping
[20:45] * cronix1 (~cronix@5.199.139.166) Quit (Ping timeout: 480 seconds)
[20:47] * mwyatt2 (~mwyatt@ip70-162-100-189.ph.ph.cox.net) has joined #ceph
[20:51] * ircolle (~Adium@2601:1:8380:2d9:64a6:ec92:ed02:408d) has joined #ceph
[20:51] * sjm (~sjm@cloudgate.cs.utsa.edu) has joined #ceph
[20:53] * sarob (~sarob@2001:4998:effd:600:1d66:df75:6caf:32b) Quit (Quit: Leaving...)
[20:54] * kevinc (~kevinc__@client65-78.sdsc.edu) Quit (Quit: This computer has gone to sleep)
[20:54] * mwyatt1 (~mwyatt@2600:1011:b12d:5d9d:0:9:e6f3:4401) Quit (Ping timeout: 480 seconds)
[21:00] * cronix1 (~cronix@5.199.139.166) has joined #ceph
[21:01] * Nacer (~Nacer@c2s31-2-83-152-89-219.fbx.proxad.net) has joined #ceph
[21:01] * i_m (~ivan.miro@nat-5-carp.hcn-strela.ru) Quit (Quit: Leaving.)
[21:02] * i_m (~ivan.miro@nat-5-carp.hcn-strela.ru) has joined #ceph
[21:02] * ScOut3R (~ScOut3R@5401C4E7.dsl.pool.telekom.hu) Quit (Read error: Connection reset by peer)
[21:03] * ScOut3R (~ScOut3R@5401C4E7.dsl.pool.telekom.hu) has joined #ceph
[21:06] * hasues (~hazuez@adsl-74-178-236-65.jax.bellsouth.net) Quit (Ping timeout: 480 seconds)
[21:10] * Cube1 (~Cube@66-87-65-226.pools.spcsdns.net) has joined #ceph
[21:10] * WhIteSidE (~chatzilla@92.86.84.202) has joined #ceph
[21:13] * Nacer (~Nacer@c2s31-2-83-152-89-219.fbx.proxad.net) Quit (Remote host closed the connection)
[21:14] * Nacer (~Nacer@c2s31-2-83-152-89-219.fbx.proxad.net) has joined #ceph
[21:14] * WhIteSidE_ (~chatzilla@31.6.14.79) has joined #ceph
[21:15] * codice (~toodles@97-94-175-73.static.mtpk.ca.charter.com) has joined #ceph
[21:16] * cronix1 (~cronix@5.199.139.166) Quit (Ping timeout: 480 seconds)
[21:16] * mwyatt2 (~mwyatt@ip70-162-100-189.ph.ph.cox.net) Quit (Quit: Bye)
[21:17] * codice_ (~toodles@97-94-175-73.static.mtpk.ca.charter.com) Quit (Ping timeout: 480 seconds)
[21:17] * Cube (~Cube@66-87-130-18.pools.spcsdns.net) Quit (Ping timeout: 480 seconds)
[21:18] * WhIteSidE (~chatzilla@92.86.84.202) Quit (Ping timeout: 480 seconds)
[21:18] * WhIteSidE_ is now known as WhIteSidE
[21:19] * Cube1 (~Cube@66-87-65-226.pools.spcsdns.net) Quit (Quit: Leaving.)
[21:24] * kevinc (~kevinc__@client65-78.sdsc.edu) has joined #ceph
[21:24] * codice_ (~toodles@97-94-175-73.static.mtpk.ca.charter.com) has joined #ceph
[21:25] * rweeks (~rweeks@192.169.20.75.static.etheric.net) Quit (Quit: Leaving)
[21:27] * codice (~toodles@97-94-175-73.static.mtpk.ca.charter.com) Quit (Read error: Operation timed out)
[21:29] * rturk|afk is now known as rturk
[21:30] * cronix1 (~cronix@5.199.139.166) has joined #ceph
[21:32] * bitserker (~toni@238.56.79.188.dynamic.jazztel.es) Quit (Ping timeout: 480 seconds)
[21:34] * sjm (~sjm@cloudgate.cs.utsa.edu) Quit (Ping timeout: 480 seconds)
[21:35] * WhIteSidE (~chatzilla@31.6.14.79) Quit (Ping timeout: 480 seconds)
[21:36] * primechuck (~primechuc@86.sub-70-197-192.myvzw.com) Quit (Remote host closed the connection)
[21:37] * kevinc (~kevinc__@client65-78.sdsc.edu) Quit (Quit: This computer has gone to sleep)
[21:38] * primechuck (~primechuc@86.sub-70-197-192.myvzw.com) has joined #ceph
[21:40] <magicrobotmonkey> is it normal for cache tier flushing to show up as client io in ceph status?
[21:40] * cronix1 (~cronix@5.199.139.166) Quit (Read error: Operation timed out)
[21:43] * jksM_ (~jks@3e6b5724.rev.stofanet.dk) has joined #ceph
[21:43] <Gugge-47527> magicrobotmonkey: ceph status show client io?
[21:43] * jksM (~jks@3e6b5724.rev.stofanet.dk) Quit (Read error: Connection reset by peer)
[21:43] <magicrobotmonkey> also i keep having osds go down under load because the symlink to the co-located journal disappears from /dev/disk/by-partuuid
[21:43] * janos (~messy@static-71-176-211-4.rcmdva.fios.verizon.net) has joined #ceph
[21:43] <magicrobotmonkey> has anyone seen that
[21:44] <Gugge-47527> ive never seen a /dev/disk/by-xx link disapper without some error logged
[21:45] * aldavud (~aldavud@217-162-119-191.dynamic.hispeed.ch) has joined #ceph
[21:45] <Gugge-47527> something is wrong if it just disappears :)
[21:46] * mwyatt (~mwyatt@137.152.127.16) Quit (Ping timeout: 480 seconds)
[21:46] * mwyatt (~mwyatt@137.152.127.16) has joined #ceph
[21:49] * rendar (~I@host117-179-dynamic.56-79-r.retail.telecomitalia.it) Quit (Ping timeout: 480 seconds)
[21:49] <ponyofdeath> hi, can anyone help me see why i am only getting ~45MB/s out of my 7 ssd cache tier with rados bench? http://bpaste.net/show/AIZD1Bds3Wu6J46MmcEg/ is my crush map
[21:49] * janos_ (~messy@static-71-176-211-4.rcmdva.fios.verizon.net) Quit (Read error: Operation timed out)
[21:49] <magicrobotmonkey> i cant wait to get on HBA instead of RAID cards
[21:49] * jrankin (~jrankin@d47-69-66-231.try.wideopenwest.com) Quit (Quit: Leaving)
[21:52] * sjm (~sjm@cloudgate.cs.utsa.edu) has joined #ceph
[21:52] * kevinc (~kevinc__@client65-78.sdsc.edu) has joined #ceph
[21:52] * rendar (~I@host117-179-dynamic.56-79-r.retail.telecomitalia.it) has joined #ceph
[21:53] * analbeard1 (~shw@5.153.255.234) has joined #ceph
[21:53] * analbeard (~shw@5.153.255.234) Quit (Read error: Connection reset by peer)
[22:01] * analbeard1 (~shw@5.153.255.234) Quit (Ping timeout: 480 seconds)
[22:02] * cronix1 (~cronix@5.199.139.166) has joined #ceph
[22:03] * sjm (~sjm@cloudgate.cs.utsa.edu) Quit (Ping timeout: 480 seconds)
[22:06] * cronix1 (~cronix@5.199.139.166) Quit (Read error: Operation timed out)
[22:08] * adamcrume (~quassel@c-71-204-162-10.hsd1.ca.comcast.net) Quit (Read error: Operation timed out)
[22:16] * Sysadmin88 (~IceChat77@94.4.20.0) has joined #ceph
[22:16] * adamcrume (~quassel@2601:9:6680:47:cd8d:a517:8aa0:3e3) has joined #ceph
[22:17] * t0rn (~ssullivan@2607:fad0:32:a02:d227:88ff:fe02:9896) Quit (Quit: Leaving.)
[22:19] * houkouonchi-home (~linux@pool-71-189-160-82.lsanca.fios.verizon.net) Quit (Ping timeout: 480 seconds)
[22:22] * primechuck (~primechuc@86.sub-70-197-192.myvzw.com) Quit (Remote host closed the connection)
[22:26] * sroy (~sroy@2607:fad8:4:6:3e97:eff:feb5:1e2b) Quit (Quit: Quitte)
[22:27] * sarob (~sarob@2001:4998:effd:600:1539:ffa7:4bba:f23a) has joined #ceph
[22:28] * tdasilva (~quassel@nat-pool-bos-t.redhat.com) Quit (Remote host closed the connection)
[22:29] * cronix1 (~cronix@5.199.139.166) has joined #ceph
[22:30] * pressureman (~pressurem@62.217.45.26) Quit (Quit: Ex-Chat)
[22:32] * neurodrone (~neurodron@static-108-30-171-7.nycmny.fios.verizon.net) Quit (Quit: neurodrone)
[22:34] <singler> what would be proper procedure to recreate a pg? I have some incomplete/corrupted pgs, and I would like to just recreate them
[22:37] * cronix1 (~cronix@5.199.139.166) Quit (Read error: Operation timed out)
[22:38] * diegows (~diegows@190.190.5.238) Quit (Ping timeout: 480 seconds)
[22:42] * houkouonchi-home (~linux@houkouonchi-1-pt.tunnel.tserv15.lax1.ipv6.he.net) has joined #ceph
[22:43] * adamcrume_ (~quassel@c-71-204-162-10.hsd1.ca.comcast.net) has joined #ceph
[22:44] * vbellur (~vijay@122.172.247.85) Quit (Read error: Operation timed out)
[22:46] * adamcrume (~quassel@2601:9:6680:47:cd8d:a517:8aa0:3e3) Quit (Ping timeout: 480 seconds)
[22:47] * cronix1 (~cronix@5.199.139.166) has joined #ceph
[22:49] * sarob (~sarob@2001:4998:effd:600:1539:ffa7:4bba:f23a) Quit (Remote host closed the connection)
[22:49] * sarob (~sarob@nat-dip27-wl-a.cfw-a-gci.corp.yahoo.com) has joined #ceph
[22:51] <s3an2> Hi, I seem to have hit an assertion failure in rbd.c that is causing a kernel panic, the logs report 'rbd_assert(img_request != NULL);' I am running a stock ubuntu 3.13 kernel - has anyone ran into this before?
[22:51] <kraken> http://i.imgur.com/H7PXV.gif
[22:53] * sherlock (sherlock@d.clients.kiwiirc.com) Quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client)
[22:55] * qhartman (~qhartman@75-151-85-53-Colorado.hfc.comcastbusiness.net) has joined #ceph
[22:56] * primechuck (~primechuc@69.170.148.179) has joined #ceph
[22:57] * vbellur (~vijay@122.172.242.189) has joined #ceph
[22:57] * sarob (~sarob@nat-dip27-wl-a.cfw-a-gci.corp.yahoo.com) Quit (Ping timeout: 480 seconds)
[22:57] * ircolle is now known as icolle-away
[23:02] * dmit2k (~Adium@balticom-131-176.balticom.lv) has joined #ceph
[23:03] <dmit2k> Hello here, can someone advise please if cache pools in current Firefly are production ready?
[23:05] * mwyatt (~mwyatt@137.152.127.16) Quit ()
[23:06] * brad_mssw (~brad@shop.monetra.com) Quit (Quit: Leaving)
[23:06] <s3an2> It is in the LTS version so it 'should' be assumed stable....
[23:08] * mashwo00 (~mashwo00@marx.geckoproject.co.uk) has joined #ceph
[23:09] * cronix1 (~cronix@5.199.139.166) Quit (Ping timeout: 480 seconds)
[23:10] * JayJ (~jayj@157.130.21.226) Quit (Quit: Computer has gone to sleep.)
[23:13] * ScOut3R (~ScOut3R@5401C4E7.dsl.pool.telekom.hu) Quit (Ping timeout: 480 seconds)
[23:18] * markbby (~Adium@168.94.245.1) Quit (Quit: Leaving.)
[23:21] * cookednoodles (~eoin@eoin.clanslots.com) Quit (Quit: Ex-Chat)
[23:22] * Knorrie (knorrie@yoshi.kantoor.mendix.nl) has left #ceph
[23:28] * sarob (~sarob@nat-dip27-wl-a.cfw-a-gci.corp.yahoo.com) has joined #ceph
[23:31] <dmit2k> s3an2: Firefly still has some new features marked as "testing"
[23:31] * kfei (~root@114-27-53-96.dynamic.hinet.net) Quit (Read error: Connection reset by peer)
[23:31] <dmit2k> just not sure of cache tier
[23:36] * sarob (~sarob@nat-dip27-wl-a.cfw-a-gci.corp.yahoo.com) Quit (Ping timeout: 480 seconds)
[23:41] * ikrstic (~ikrstic@178-222-94-242.dynamic.isp.telekom.rs) Quit (Quit: Konversation terminated!)
[23:42] <joshd> s3an2: that was mitigated in 3.14.5, and should be entirely fixed in 3.16
[23:44] * ninkotech_ (~duplo@static-84-242-87-186.net.upcbroadband.cz) has joined #ceph
[23:48] <Sysadmin88> would it be possible to have a geographically distributed cluster with a cache pool in one location to take writes and read cache? i know performance of backend read/write would be very slow... but getting sites with replicas is very important and would the cache tier compensate much for the slowness?
[23:49] * kfei (~root@114-27-89-92.dynamic.hinet.net) has joined #ceph
[23:52] * nljmo (~nljmo@64.125.103.162) Quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz???)
[23:54] * analbeard (~shw@host81-147-14-90.range81-147.btcentralplus.com) has joined #ceph
[23:56] * Nacer (~Nacer@c2s31-2-83-152-89-219.fbx.proxad.net) Quit (Remote host closed the connection)
[23:58] * sarob (~sarob@nat-dip27-wl-a.cfw-a-gci.corp.yahoo.com) has joined #ceph

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