#ceph IRC Log

Index

IRC Log for 2013-09-16

Timestamps are in GMT/BST.

[0:06] * BillK (~BillK-OFT@58-7-172-nwork.dyn.iinet.net.au) has joined #ceph
[0:07] * ScOut3R_ (~scout3r@4E5C2305.dsl.pool.telekom.hu) has joined #ceph
[0:07] * ScOut3R (~scout3r@4E5C2305.dsl.pool.telekom.hu) Quit (Read error: Connection reset by peer)
[0:21] * AfC (~andrew@2407:7800:200:1011:2ad2:44ff:fe08:a4c) has joined #ceph
[0:34] * sjustlaptop (~sam@ip174-70-103-52.no.no.cox.net) Quit (Ping timeout: 480 seconds)
[0:38] * wschulze (~wschulze@cpe-98-14-21-89.nyc.res.rr.com) has joined #ceph
[0:55] * scuttlemonkey (~scuttlemo@38.127.1.5) has joined #ceph
[0:55] * ChanServ sets mode +o scuttlemonkey
[0:58] * ScOut3R_ (~scout3r@4E5C2305.dsl.pool.telekom.hu) Quit (Remote host closed the connection)
[1:01] * malcolm_ (~malcolm@silico24.lnk.telstra.net) has joined #ceph
[1:09] * scuttlemonkey (~scuttlemo@38.127.1.5) Quit (Ping timeout: 480 seconds)
[1:12] * jjgalvez (~jjgalvez@ip72-193-217-254.lv.lv.cox.net) has joined #ceph
[1:14] * berant (~blemmenes@24-236-241-163.dhcp.trcy.mi.charter.com) Quit (Quit: berant)
[1:21] * Steki (~steki@198.199.65.141) Quit (Ping timeout: 480 seconds)
[1:25] * jjgalvez (~jjgalvez@ip72-193-217-254.lv.lv.cox.net) Quit (Read error: Connection reset by peer)
[1:26] * jjgalvez (~jjgalvez@ip72-193-217-254.lv.lv.cox.net) has joined #ceph
[1:35] * sagelap (~sage@76.89.177.113) has joined #ceph
[2:01] * malcolm_ (~malcolm@silico24.lnk.telstra.net) Quit (Ping timeout: 480 seconds)
[2:03] * houkouonchi-home (~linux@pool-108-38-63-48.lsanca.fios.verizon.net) Quit (Remote host closed the connection)
[2:06] * danieagle (~Daniel@177.133.174.145) Quit (Quit: inte+ e Obrigado Por tudo mesmo! :-D)
[2:07] * Lea (~LeaChim@host86-135-252-168.range86-135.btcentralplus.com) Quit (Read error: Operation timed out)
[2:14] * KindTwo (~KindOne@h80.208.89.75.dynamic.ip.windstream.net) has joined #ceph
[2:16] * KindOne (~KindOne@0001a7db.user.oftc.net) Quit (Ping timeout: 480 seconds)
[2:16] * KindTwo is now known as KindOne
[2:36] * malcolm_ (~malcolm@a82-3.nat.uq.edu.au) has joined #ceph
[2:51] * sglwlb (~sglwlb@221.12.27.202) Quit ()
[2:55] * yy-nm (~Thunderbi@122.233.46.14) has joined #ceph
[2:58] * wschulze (~wschulze@cpe-98-14-21-89.nyc.res.rr.com) Quit (Quit: Leaving.)
[3:06] * sleinen (~Adium@2001:620:0:25:5129:5e26:1e64:ffb) Quit (Quit: Leaving.)
[3:06] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) has joined #ceph
[3:12] * markbby (~Adium@168.94.245.3) has joined #ceph
[3:14] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) Quit (Ping timeout: 480 seconds)
[3:14] * john_barbee (~jbarbee@c-98-220-74-174.hsd1.in.comcast.net) has joined #ceph
[3:15] * nerdtron (~kenneth@202.60.8.252) has joined #ceph
[3:15] * john_barbee (~jbarbee@c-98-220-74-174.hsd1.in.comcast.net) Quit ()
[3:15] * john_barbee (~jbarbee@c-98-220-74-174.hsd1.in.comcast.net) has joined #ceph
[3:21] * glzhao (~glzhao@117.79.232.211) has joined #ceph
[3:35] * a_ (~a@pool-173-55-143-200.lsanca.fios.verizon.net) Quit (Read error: Operation timed out)
[3:36] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) has joined #ceph
[3:37] * malcolm_ (~malcolm@a82-3.nat.uq.edu.au) Quit (Ping timeout: 480 seconds)
[3:38] * Vjarjadian (~IceChat77@05453253.skybroadband.com) Quit (Ping timeout: 480 seconds)
[3:49] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) Quit (Ping timeout: 480 seconds)
[3:51] * yanzheng (~zhyan@134.134.139.74) has joined #ceph
[3:55] * markbby (~Adium@168.94.245.3) Quit (Remote host closed the connection)
[3:59] <Qu310> with the radosgw does anyone know why when using the libapache2-mod-fastcgi ubuntu package, i get in my apache error log "chunked Transfer-Encoding forbidden: /swift/v1/" but when using the module from wget http://gitbuilder.ceph.com/libapache-mod-fastcgi-deb-precise-x86_64-basic/ref/master/pool/main/liba/libapache-mod-fastcgi/libapache2-mod-fastcgi_2.4.7~0910052141-1-inktank2_amd64.deb it works
[3:59] <Qu310> fine?
[4:04] * a (~a@pool-173-55-143-200.lsanca.fios.verizon.net) has joined #ceph
[4:05] * a is now known as Guest6767
[4:07] * malcolm_ (~malcolm@silico24.lnk.telstra.net) has joined #ceph
[4:15] * eternaleye (~eternaley@2002:3284:29cb::1) Quit (Ping timeout: 480 seconds)
[4:34] * shang (~ShangWu@175.41.48.77) has joined #ceph
[4:38] * newbie (~kvirc@111.173.155.71) has joined #ceph
[4:39] * newbie is now known as huangjun
[4:41] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) has joined #ceph
[4:43] * sjustlaptop (~sam@172.56.21.10) has joined #ceph
[4:49] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) Quit (Ping timeout: 480 seconds)
[4:52] * mnash (~chatzilla@66-194-114-178.static.twtelecom.net) Quit (Ping timeout: 480 seconds)
[5:06] * fireD (~fireD@93-142-213-185.adsl.net.t-com.hr) has joined #ceph
[5:07] * fireD_ (~fireD@93-142-229-143.adsl.net.t-com.hr) Quit (Ping timeout: 480 seconds)
[5:31] * jjgalvez (~jjgalvez@ip72-193-217-254.lv.lv.cox.net) Quit (Quit: Leaving.)
[5:34] * grepory (~Adium@c-69-181-42-170.hsd1.ca.comcast.net) has joined #ceph
[5:40] * john_barbee (~jbarbee@c-98-220-74-174.hsd1.in.comcast.net) Quit (Quit: ChatZilla 0.9.90.1 [Firefox 23.0.1/20130814063812])
[5:42] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) has joined #ceph
[5:50] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) Quit (Ping timeout: 480 seconds)
[6:02] * yy-nm (~Thunderbi@122.233.46.14) Quit (Quit: yy-nm)
[6:03] * julian (~julianwa@125.70.133.27) has joined #ceph
[6:15] * a (~a@pool-173-55-143-200.lsanca.fios.verizon.net) has joined #ceph
[6:15] * Guest6767 (~a@pool-173-55-143-200.lsanca.fios.verizon.net) Quit (Read error: Connection reset by peer)
[6:16] * a is now known as Guest6776
[6:34] * eternaleye (~eternaley@2002:3284:29cb::1) has joined #ceph
[6:38] * huangjun (~kvirc@111.173.155.71) Quit (Ping timeout: 480 seconds)
[6:40] * scuttlemonkey (~scuttlemo@38.127.1.5) has joined #ceph
[6:40] * ChanServ sets mode +o scuttlemonkey
[6:42] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) has joined #ceph
[6:45] * sjustlaptop (~sam@172.56.21.10) Quit (Read error: Connection reset by peer)
[6:50] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) Quit (Ping timeout: 480 seconds)
[6:51] * grepory (~Adium@c-69-181-42-170.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[6:53] * sjustlaptop (~sam@172.56.21.10) has joined #ceph
[7:00] * huangjun (~kvirc@111.173.155.71) has joined #ceph
[7:00] * sjustlaptop (~sam@172.56.21.10) Quit (Read error: Connection reset by peer)
[7:05] * yy-nm (~Thunderbi@122.233.46.14) has joined #ceph
[7:21] * julian (~julianwa@125.70.133.27) Quit (Read error: Connection timed out)
[7:22] * julian (~julianwa@125.70.133.27) has joined #ceph
[7:29] * erice (~erice@c-98-245-48-79.hsd1.co.comcast.net) Quit (Quit: erice)
[7:36] * ScOut3R (~ScOut3R@4E5C2305.dsl.pool.telekom.hu) has joined #ceph
[7:36] * yeled (~yeled@spodder.com) Quit (Quit: brb)
[7:40] * yeled (~yeled@spodder.com) has joined #ceph
[7:43] * Kioob1 (~kioob@2a01:e35:2432:58a0:21e:8cff:fe07:45b6) Quit (Quit: Leaving.)
[7:44] * ScOut3R (~ScOut3R@4E5C2305.dsl.pool.telekom.hu) Quit (Ping timeout: 480 seconds)
[7:48] * sjm (~sjm@dhcp-108-168-18-236.cable.user.start.ca) has joined #ceph
[8:02] * foosinn (~stefan@office.unitedcolo.de) has joined #ceph
[8:03] * Guest6776 (~a@pool-173-55-143-200.lsanca.fios.verizon.net) Quit (Quit: This computer has gone to sleep)
[8:07] * mrprud (~mrprud@ANantes-554-1-246-148.w2-1.abo.wanadoo.fr) Quit (Remote host closed the connection)
[8:11] * iii8 (~Miranda@91.207.132.71) has joined #ceph
[8:15] * cofol1986 (~xwrj@110.90.119.113) has joined #ceph
[8:23] * jjgalvez (~jjgalvez@ip72-193-217-254.lv.lv.cox.net) has joined #ceph
[8:31] * scuttlemonkey (~scuttlemo@38.127.1.5) Quit (Read error: Operation timed out)
[8:35] * SpamapS (~clint@xencbyrum2.srihosting.com) Quit (Ping timeout: 480 seconds)
[8:41] * sleinen (~Adium@2001:620:0:46:e5c1:6929:543c:6f41) has joined #ceph
[8:41] * sleinen1 (~Adium@2001:620:0:2d:750b:7ea9:6e63:a6b7) has joined #ceph
[8:43] * sleinen2 (~Adium@2001:620:0:25:adf6:7263:f6b8:7f39) has joined #ceph
[8:49] * sleinen (~Adium@2001:620:0:46:e5c1:6929:543c:6f41) Quit (Ping timeout: 480 seconds)
[8:49] * gurubert (~gurubert@p4FF5BF7F.dip0.t-ipconnect.de) has joined #ceph
[8:49] * sleinen1 (~Adium@2001:620:0:2d:750b:7ea9:6e63:a6b7) Quit (Ping timeout: 480 seconds)
[8:54] * dpippenger (~riven@cpe-76-166-208-83.socal.res.rr.com) has joined #ceph
[8:54] * allsystemsarego (~allsystem@188.25.134.128) has joined #ceph
[9:00] * sleinen2 (~Adium@2001:620:0:25:adf6:7263:f6b8:7f39) Quit (Quit: Leaving.)
[9:00] * sleinen (~Adium@130.59.94.200) has joined #ceph
[9:03] * sleinen1 (~Adium@2001:620:0:25:2cc1:cb2e:d189:43d) has joined #ceph
[9:04] * AfC (~andrew@2407:7800:200:1011:2ad2:44ff:fe08:a4c) Quit (Quit: Leaving.)
[9:06] * sleinen (~Adium@130.59.94.200) Quit (Read error: Operation timed out)
[9:06] * AfC (~andrew@2407:7800:200:1011:2ad2:44ff:fe08:a4c) has joined #ceph
[9:07] * malcolm_ (~malcolm@silico24.lnk.telstra.net) Quit (Ping timeout: 480 seconds)
[9:19] * Lea (~LeaChim@host86-135-252-168.range86-135.btcentralplus.com) has joined #ceph
[9:19] * glzhao (~glzhao@117.79.232.211) Quit (Read error: Connection reset by peer)
[9:20] * BManojlovic (~steki@91.195.39.5) has joined #ceph
[9:32] * AfC1 (~andrew@101.119.14.1) has joined #ceph
[9:32] * AfC1 (~andrew@101.119.14.1) Quit ()
[9:33] * nigwil (~chatzilla@2001:44b8:5144:7b00:75cb:ed1:3b99:68b) has joined #ceph
[9:33] * AfC1 (~andrew@101.119.14.1) has joined #ceph
[9:34] * dpippenger (~riven@cpe-76-166-208-83.socal.res.rr.com) Quit (Ping timeout: 480 seconds)
[9:37] * brambles (lechuck@s0.barwen.ch) Quit (Read error: Connection reset by peer)
[9:37] * brambles (lechuck@s0.barwen.ch) has joined #ceph
[9:38] * AfC2 (~andrew@2407:7800:200:1011:2ad2:44ff:fe08:a4c) has joined #ceph
[9:38] * AfC1 (~andrew@101.119.14.1) Quit (Read error: Connection reset by peer)
[9:38] * dpippenger (~riven@cpe-76-166-208-83.socal.res.rr.com) has joined #ceph
[9:39] * AfC (~andrew@2407:7800:200:1011:2ad2:44ff:fe08:a4c) Quit (Ping timeout: 480 seconds)
[9:41] * AfC2 (~andrew@2407:7800:200:1011:2ad2:44ff:fe08:a4c) Quit ()
[9:50] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) has joined #ceph
[9:50] * ChanServ sets mode +v andreask
[9:53] * steki (~steki@198.199.65.141) has joined #ceph
[9:55] * odyssey4me (~odyssey4m@165.233.71.2) has joined #ceph
[9:56] * BManojlovic (~steki@91.195.39.5) Quit (Ping timeout: 480 seconds)
[9:58] * Cube (~Cube@cpe-76-95-217-129.socal.res.rr.com) has joined #ceph
[10:10] * erwan_ta- is now known as erwan_taf
[10:15] * dpippenger (~riven@cpe-76-166-208-83.socal.res.rr.com) Quit (Ping timeout: 480 seconds)
[10:24] * nigwil (~chatzilla@2001:44b8:5144:7b00:75cb:ed1:3b99:68b) Quit (Ping timeout: 480 seconds)
[10:29] * ScOut3R (~scout3r@4E5C2305.dsl.pool.telekom.hu) has joined #ceph
[10:30] * morse (~morse@supercomputing.univpm.it) has joined #ceph
[10:35] <joelio> Qu310: I think you answered your own question, right? The ceph version contains support for chunking I'd guess.. the ubuntu stock does not
[10:44] * hybrid5121 (~walid@LPoitiers-156-86-25-85.w193-248.abo.wanadoo.fr) has joined #ceph
[10:49] * hybrid512 (~walid@LPoitiers-156-86-25-85.w193-248.abo.wanadoo.fr) Quit (Ping timeout: 480 seconds)
[10:52] * ScOut3R_ (~scout3r@4E5C2305.dsl.pool.telekom.hu) has joined #ceph
[10:52] * ScOut3R (~scout3r@4E5C2305.dsl.pool.telekom.hu) Quit (Read error: Connection reset by peer)
[10:52] <ccourtaut> tiens petite question, j'essaye de lancer un test sur teuthology, mais ca se gauffre
[10:53] <ccourtaut> ccourtaut: sorry wrong channel :)
[10:55] <ccourtaut> thoses monday mornings...
[10:57] * glzhao (~glzhao@211.155.113.239) has joined #ceph
[11:01] * yanzheng (~zhyan@134.134.139.74) Quit (Quit: Leaving)
[11:03] * dosaboy (~dosaboy@65.93.189.91.lcy-01.canonistack.canonical.com) Quit (Quit: leaving)
[11:03] * dosaboy (~dosaboy@host109-158-232-255.range109-158.btcentralplus.com) has joined #ceph
[11:04] * yy-nm (~Thunderbi@122.233.46.14) Quit (Quit: yy-nm)
[11:05] * sleinen1 (~Adium@2001:620:0:25:2cc1:cb2e:d189:43d) Quit (Ping timeout: 480 seconds)
[11:11] <joao> lol ccourtaut
[11:11] <joao> morning all
[11:12] * sleinen (~Adium@2001:620:0:25:494:65f3:5e8d:4704) has joined #ceph
[11:12] <mech422> actually...I think I"m going to bed :-)
[11:12] <mech422> g'nite all
[11:12] * mech422 (~steve@ip68-2-159-8.ph.ph.cox.net) Quit (Quit: Leaving.)
[11:25] * jbd_ (~jbd_@2001:41d0:52:a00::77) has joined #ceph
[11:28] * lx0 (~aoliva@lxo.user.oftc.net) has joined #ceph
[11:31] * lxo (~aoliva@lxo.user.oftc.net) Quit (Ping timeout: 480 seconds)
[11:32] * Rocky (~r.nap@188.205.52.204) has left #ceph
[11:41] * sleinen1 (~Adium@2001:620:0:46:e0e3:2504:6f84:4c14) has joined #ceph
[11:43] * Cube (~Cube@cpe-76-95-217-129.socal.res.rr.com) Quit (Quit: Leaving.)
[11:44] * infinitytrapdoor (~infinityt@134.95.27.132) has joined #ceph
[11:46] * jjgalvez (~jjgalvez@ip72-193-217-254.lv.lv.cox.net) Quit (Quit: Leaving.)
[11:47] * sleinen (~Adium@2001:620:0:25:494:65f3:5e8d:4704) Quit (Ping timeout: 480 seconds)
[11:53] * claenjoy (~leggenda@37.157.33.36) has joined #ceph
[11:57] * deadsimple (~infinityt@134.95.27.132) has joined #ceph
[12:03] * infinitytrapdoor (~infinityt@134.95.27.132) Quit (Ping timeout: 480 seconds)
[12:05] * jcsp (~john@82-71-55-202.dsl.in-addr.zen.co.uk) has joined #ceph
[12:11] * infinitytrapdoor (~infinityt@134.95.27.132) has joined #ceph
[12:11] * deadsimple (~infinityt@134.95.27.132) Quit (Read error: Connection reset by peer)
[12:12] * infinitytrapdoor (~infinityt@134.95.27.132) Quit ()
[12:14] * Cube (~Cube@cpe-76-95-217-129.socal.res.rr.com) has joined #ceph
[12:15] * nerdtron (~kenneth@202.60.8.252) Quit (Remote host closed the connection)
[12:21] * huangjun (~kvirc@111.173.155.71) Quit (Read error: Connection reset by peer)
[12:21] * tserong_ (~tserong@203-57-208-89.dyn.iinet.net.au) Quit (Quit: Leaving)
[12:26] * Cube (~Cube@cpe-76-95-217-129.socal.res.rr.com) Quit (Ping timeout: 480 seconds)
[12:32] * steki (~steki@198.199.65.141) Quit (Quit: Ja odoh a vi sta 'ocete...)
[12:32] * steki (~steki@198.199.65.141) has joined #ceph
[12:35] <claenjoy> does ceph-deploy support OSD on LVM ?
[12:41] <andreask> claenjoy: not at the moment ... but someone already made feature request http://tracker.ceph.com/issues/5620
[12:48] * glzhao (~glzhao@211.155.113.239) Quit (Quit: leaving)
[12:52] * nwf (~nwf@67.62.51.95) Quit (Ping timeout: 480 seconds)
[12:56] <jtang> good morning
[13:01] * JustEra (~JustEra@89.234.148.11) has joined #ceph
[13:01] <JustEra> Hello
[13:02] <JustEra> How can I user ceph with ext4 ? the --fs-type don't work ?
[13:02] <JustEra> use*
[13:02] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) Quit (Ping timeout: 480 seconds)
[13:07] <joelio> hrm, had a mon flapping state today.. checking the logs it seems as though the faulty mon is popping up with - service_dispatch not in quorum -- drop message
[13:07] <joelio> then it's ok after a re-election, then a minute later dropas
[13:07] <joelio> repeat ad infinitum..
[13:25] <joelio> hmm, now.. health HEALTH_WARN mds cluster is degraded
[13:33] * sprachgenerator (~sprachgen@c-50-141-192-36.hsd1.il.comcast.net) has joined #ceph
[13:41] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) has joined #ceph
[13:41] * ChanServ sets mode +v andreask
[14:01] * gurubert (~gurubert@p4FF5BF7F.dip0.t-ipconnect.de) Quit (Quit: Leaving.)
[14:03] * yanzheng (~zhyan@jfdmzpr01-ext.jf.intel.com) has joined #ceph
[14:03] * gurubert (~gurubert@p4FF5BF7F.dip0.t-ipconnect.de) has joined #ceph
[14:08] * sleinen (~Adium@2001:620:0:2d:5d37:24c5:b001:ceab) has joined #ceph
[14:11] * kraken (~kraken@c-24-131-46-23.hsd1.ga.comcast.net) has joined #ceph
[14:11] * alfredodeza (~alfredode@c-24-131-46-23.hsd1.ga.comcast.net) has joined #ceph
[14:13] * sleinen1 (~Adium@2001:620:0:46:e0e3:2504:6f84:4c14) Quit (Ping timeout: 480 seconds)
[14:15] <joelio> heh, right, I think I might have got to the bottom, but really unsure why. Resetarted services, affected box to no avail.. I have another test cluster attached to the same switch (which was partially downed) but on a different subnet however. I fixed the test cluster back up and low and behold, the prod cluster is fine again
[14:15] <joelio> can't imagine why that would be the case unless there something multicast-y going on.. can't be broadcast, diffeent net.
[14:16] <joelio> unless it's a complete red-herring, but look far too related
[14:16] * sleinen (~Adium@2001:620:0:2d:5d37:24c5:b001:ceab) Quit (Ping timeout: 480 seconds)
[14:18] * Cube (~Cube@cpe-76-95-217-129.socal.res.rr.com) has joined #ceph
[14:19] * gurubert (~gurubert@p4FF5BF7F.dip0.t-ipconnect.de) Quit (Remote host closed the connection)
[14:19] * gurubert (~gurubert@p4FF5BF7F.dip0.t-ipconnect.de) has joined #ceph
[14:20] * gurubert (~gurubert@p4FF5BF7F.dip0.t-ipconnect.de) has left #ceph
[14:21] <joelio> hmm, no, spoke to soon.. flapping again
[14:21] <joelio> upgraded to ceph version 0.61.8 (a6fdcca3bddbc9f177e4e2bf0d9cdd85006b028b)
[14:22] <joelio> last week.. wasn't doing it before then
[14:30] <peetaur> I'm new to ceph, but what I'd try would be to remove the monitor, clean up its files, verify the package versions are all the same on all monitors, and create it again
[14:30] * Cube (~Cube@cpe-76-95-217-129.socal.res.rr.com) Quit (Ping timeout: 480 seconds)
[14:30] <joelio> no, that's not a great idea.. removing mons when you're not sure isn't the wisest of propositions
[14:31] <joelio> can compound the issue
[14:31] * marrusl (~mark@209-150-43-182.c3-0.wsd-ubr2.qens-wsd.ny.cable.rcn.com) has joined #ceph
[14:33] * thomnico (~thomnico@wmh38-5-78-223-116-113.fbx.proxad.net) has joined #ceph
[14:40] * thanasisk (~akostopou@p5DDB8599.dip0.t-ipconnect.de) has joined #ceph
[14:43] <thanasisk> ceph-deploy mon create mon-ceph-node <-- do I run this as ceph or as root?
[14:43] <thanasisk> hi all, btw
[14:44] <alfredodeza> thanasisk: ceph-deploy will connect to the remote host as the user that is executing ceph-deploy
[14:44] <alfredodeza> so if you are user `foo` calling ceph-deploy, then you should have a `foo` user in the remote host you can connect as
[14:44] <alfredodeza> no need for sudo locally
[14:44] <alfredodeza> as there is nothing ceph-deploy does that needs sudo permissions locally
[14:46] <thanasisk> alfredodeza, http://pastebin.com/zx3kMT33
[14:46] <thanasisk> thats my lengthy error output
[14:47] <thanasisk> all done as user ceph in directory mycluster
[14:47] <alfredodeza> thanasisk: can the user `ceph` in the remote host run commands as paswordless sudo?
[14:48] <thanasisk> yes - i just verified
[14:48] * thomnico (~thomnico@wmh38-5-78-223-116-113.fbx.proxad.net) Quit (Ping timeout: 480 seconds)
[14:49] <thanasisk> ceph@brizo:~/mycluster$ sudo /usr/sbin/service ceph start mon.brizo
[14:49] <thanasisk> === mon.brizo ===
[14:49] <thanasisk> Starting Ceph mon.brizo on brizo...
[14:49] <thanasisk> failed: 'ulimit -n 32768; /usr/bin/ceph-mon -i brizo --pid-file /var/run/ceph/mon.brizo.pid -c /etc/ceph/ceph.conf '
[14:50] <alfredodeza> yep, that is one problem
[14:50] <alfredodeza> the other problem is this one: RuntimeError: config file /etc/ceph/ceph.conf exists with different content; use --overwrite-conf to overwrite
[14:50] <alfredodeza> you probably need to overwrite the configuratino
[14:50] <alfredodeza> configuration
[14:51] <thanasisk> so how do I do that? do I pass --overwrite-conf to mon create command?
[14:51] * wschulze (~wschulze@cpe-72-229-37-201.nyc.res.rr.com) has joined #ceph
[14:52] <thanasisk> alfredodeza, some context: I am trying to add my admin node as a monitor node, is that possible?
[14:52] <alfredodeza> thanasisk: yes, just like that log line said :)
[14:52] <alfredodeza> and yes, I believe that you can make the admin node the same
[14:53] <alfredodeza> but I am not 100% sure there
[14:53] <alfredodeza> you might want to check the docs on that one
[14:53] <thanasisk> ceph-deploy mon create brizo.classmarkets.net --overwrite-conf
[14:53] <thanasisk> usage: ceph-deploy [-h] [-v | -q] [-n] [--version] [--overwrite-conf]
[14:53] <thanasisk> [--cluster NAME]
[14:53] <thanasisk> COMMAND ...
[14:53] <thanasisk> ceph-deploy: error: unrecognized arguments: --overwrite-conf
[14:53] <joelio> yea, you've got it wrong
[14:53] <joelio> not at the end
[14:54] <joelio> ceph-deploy --overwrite-conf mon create brizo.classmarkets.net
[14:54] <joelio> it does tell you, you know
[14:54] <alfredodeza> thanasisk: `ceph-deploy --overwrite-conf mon create brizo.classmarkets.net
[14:54] <joelio> reading always helps
[14:54] <alfredodeza> bah joelio wins
[14:54] <joelio> :)
[14:54] <alfredodeza> norris joelio
[14:54] <kraken> joelio was once on Wheel of Fortune and was the first to spin. The next 29 minutes of the show consisted of everyone standing around awkwardly, waiting for the wheel to stop.
[14:54] <joelio> haha
[14:55] <thanasisk> ok that takes care of the 1st error thanks a lot :)
[14:55] <thanasisk> how about the 2nd one?
[14:55] <joelio> what second one sorry?
[14:56] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) has left #ceph
[14:56] <alfredodeza> I am not sure about the ulimit one
[14:56] <joelio> that's a generic error, don't worry
[14:56] <joelio> I get that too
[14:56] <thanasisk> http://pastebin.com/hepeEjBS
[14:57] <thanasisk> thats my second error
[14:58] <thanasisk> im on debian wheezy
[14:58] <joelio> looks like a perms issue to me
[14:58] * dpippenger (~riven@cpe-76-166-208-83.socal.res.rr.com) has joined #ceph
[14:59] <thanasisk> norris thanasisk
[14:59] <kraken> There used to be a street named after thanasisk, but it was changed because nobody crosses thanasisk and lives.
[15:00] * joelio fixes flapping mon with a judicious restart of other mons
[15:00] <joelio> not convinced it won't come back mind.. will keep my eye on it
[15:03] <thanasisk> i run the command manualy
[15:05] <JustEra> how to force the use of ext4 with ceph-deploy ?
[15:05] <alfredodeza> JustEra: you cannot do that with ceph-deploy
[15:05] <alfredodeza> you just get defaults
[15:05] <JustEra> alfredodeza, the docs say to pass th --fs-type but it don't work :'(
[15:06] * dspano (~dspano@rrcs-24-103-221-202.nys.biz.rr.com) has joined #ceph
[15:06] <alfredodeza> I believe that is a bit of a confusing part of the docs
[15:06] <alfredodeza> and I think it means --fs-type for the command that ceph-deploy runs
[15:06] <alfredodeza> not for ceph-deploy
[15:07] <JustEra> hmmm
[15:07] <JustEra> I've done it manually but the osd is not up :/
[15:08] <alfredodeza> if you need to do something custom, outside of the defaults that ceph-deploy gives you then it sounds to me you are ready to move on to a more granular tool like chef/puppet/ansible/salt etc...
[15:11] <JustEra> I'm pretty news to ceph, all I want is a cluster of 3 server with ext4 :/
[15:11] * julian (~julianwa@125.70.133.27) Quit (Read error: Connection reset by peer)
[15:12] * julian (~julianwa@125.70.133.27) has joined #ceph
[15:12] <joelio> JustEra: why not XFS?
[15:12] <joelio> much better imho
[15:12] <JustEra> Joelio yeah I know but Idera don't work with that :p
[15:13] <joelio> this is for OSDs remember, nothing user facing
[15:13] <joelio> and I'm not sure I understand what you mean by Idera don't work with that
[15:14] <joelio> it's a filesystem?
[15:14] * julian (~julianwa@125.70.133.27) Quit ()
[15:14] <JustEra> Idera is a backup system
[15:14] <JustEra> it's doesnt support xfs/btrfs atm
[15:15] <joelio> sounds... erm... sucky :)
[15:17] <JustEra> If you know a better system x)
[15:17] * oliver (~oliver@p4FFFECD5.dip0.t-ipconnect.de) has joined #ceph
[15:19] * malcolm_ (~malcolm@101.165.48.42) has joined #ceph
[15:22] <joelio> JustEra: amanda, bacula.... rsync ;) quite a few tbh
[15:23] <joelio> we don't backup OSDs though, it has to be said
[15:23] <joelio> I don't bother backing up all the replication data
[15:23] <joelio> pointless
[15:23] <joelio> just backup the images and /or temporal data and leverage automation more
[15:24] <joelio> easy to rebuild a VM in puppet and just rsync back data - not that we've neede to do that yet
[15:25] <joelio> hmm, actually - may try Amanda agains radosgw
[15:25] <joelio> that'd be quite funku
[15:25] <joelio> funky
[15:27] <thanasisk> 2013-09-16 13:26:38.877290 7fa25e6f3700 -1 monclient(hunting): ERROR: missing keyring, cannot use cephx for authentication
[15:27] <thanasisk> 2013-09-16 13:26:38.877302 7fa25e6f3700 0 librados: client.admin initialization error (2) No such file or directory
[15:27] <thanasisk> Error connecting to cluster: ObjectNotFound
[15:27] <thanasisk> what does that mean? it is the output of ceph -s
[15:27] <thanasisk> as ceph user under ~/mycluster
[15:28] * gregmark (~Adium@68.87.42.115) has joined #ceph
[15:30] * sprachgenerator (~sprachgen@c-50-141-192-36.hsd1.il.comcast.net) Quit (Quit: sprachgenerator)
[15:30] * BManojlovic (~steki@91.195.39.5) has joined #ceph
[15:31] <JustEra> hmmm where I can know why osds are in but not up ?
[15:32] * sagelap (~sage@76.89.177.113) Quit (Ping timeout: 480 seconds)
[15:33] * dspano (~dspano@rrcs-24-103-221-202.nys.biz.rr.com) Quit (Ping timeout: 480 seconds)
[15:34] * steki (~steki@198.199.65.141) Quit (Ping timeout: 480 seconds)
[15:35] <alfredodeza> thanasisk: what are you running to get that?
[15:35] <alfredodeza> are you running that command with sudo (or super user privs) ?
[15:35] * sagelap (~sage@2600:1012:b00a:2d4:dc5c:a28:49c2:ca85) has joined #ceph
[15:36] <thanasisk> ive ran it as root and as ceph - identical results
[15:37] <joelio> thanasisk: did you gatherkeys
[15:37] <thanasisk> one caveat of the documentation is that it does not specify which commands needs superuser privileges
[15:37] <thanasisk> joelio, let me try it
[15:37] * dpippenger (~riven@cpe-76-166-208-83.socal.res.rr.com) Quit (Ping timeout: 480 seconds)
[15:38] * mnash (~chatzilla@66-194-114-178.static.twtelecom.net) has joined #ceph
[15:38] <joelio> thanasisk: it's part of the documentation, so you should have done that step
[15:39] <thanasisk> http://pastebin.com/vVkPS9NV
[15:40] <joelio> I've got a feeling as you've had to --overwrite-conf, you'll have cruft let about from a previous attempt at install
[15:40] <thanasisk> im working with an existing lab installation (lab = not production)
[15:40] <joelio> you need to purge all and start from scratch
[15:40] <joelio> use a new working direcory
[15:41] <joelio> you also need to leave about 30 seconds for the inital mon create to make the keyring (in my experience) - I guess it needs entorpy
[15:41] * dspano (~dspano@rrcs-24-103-221-202.nys.biz.rr.com) has joined #ceph
[15:41] <joelio> alternatively try and set that machine as an admin host (althouhg it looks like you are already on that host)
[15:42] <thanasisk> it is already an admin host
[15:43] <thanasisk> can an admin host be an OSD host at the same time? (NOT for production)
[15:44] <joelio> yea, of course
[15:44] <joelio> even in prod
[15:49] * thomnico (~thomnico@37.160.123.220) has joined #ceph
[15:50] * grepory (~Adium@50-115-70-146.static-ip.telepacific.net) has joined #ceph
[15:52] * sagelap (~sage@2600:1012:b00a:2d4:dc5c:a28:49c2:ca85) Quit (Read error: Connection reset by peer)
[15:53] * madkiss (~madkiss@089144192241.atnat0001.highway.a1.net) has joined #ceph
[15:54] <madkiss> hey there, is there any way for me to make the Ceph gateway authenticate against LDAP or Active Directory?
[15:55] <madkiss> i guess i could use the keystone backend
[15:55] <madkiss> but that would require swift mode
[15:57] * peetaur (~peter@CPEbc1401e60493-CMbc1401e60490.cpe.net.cable.rogers.com) Quit (Ping timeout: 480 seconds)
[15:58] <thanasisk> [ceph_deploy.mds][DEBUG ] Deploying mds.calypso.classmarkets.net to calypso.classmarkets.net <-- it has been like that for 5 minutes, is that normal? (atom powered servers here but the rest of the process was very fast)
[15:58] * yanzheng (~zhyan@jfdmzpr01-ext.jf.intel.com) Quit (Remote host closed the connection)
[16:01] * thomnico (~thomnico@37.160.123.220) Quit (Ping timeout: 480 seconds)
[16:01] * peetaur (~peter@CPEbc1401e60493-CMbc1401e60490.cpe.net.cable.rogers.com) has joined #ceph
[16:03] * BillK (~BillK-OFT@58-7-172-nwork.dyn.iinet.net.au) Quit (Ping timeout: 480 seconds)
[16:04] * tziOm (~bjornar@194.19.106.242) has joined #ceph
[16:04] * sagelap (~sage@2600:1012:b00a:2d4:dc5c:a28:49c2:ca85) has joined #ceph
[16:04] <thanasisk> it fails with http://pastebin.com/hhyz8tZV - pipe.fault - what gives?
[16:05] * mikedawson (~chatzilla@23-25-46-97-static.hfc.comcastbusiness.net) has joined #ceph
[16:07] <joelio> thanasisk: did you get HEALTH_OK before doing MDS?
[16:08] <thanasisk> from which command?
[16:08] <joelio> umm, ceph -s, or ceph -w
[16:08] <thanasisk> let me see
[16:08] <thanasisk> again: 2013-09-16 14:08:29.775801 7fd31da64700 -1 monclient(hunting): ERROR: missing keyring, cannot use cephx for authentication
[16:08] <thanasisk> 2013-09-16 14:08:29.775944 7fd31da64700 0 librados: client.admin initialization error (2) No such file or directory
[16:08] <thanasisk> Error connecting to cluster: ObjectNotFound
[16:09] <joelio> so, no then :)
[16:09] <thanasisk> this is under a new working directory
[16:09] <joelio> did you purge everythying, packasges, /var/lib/ceph etc.
[16:10] * jcfischer (~fischer@macjcf.switch.ch) has joined #ceph
[16:10] <thanasisk> do I really need to do that?
[16:10] <thanasisk> the joys of working with pre v1.0 software :)
[16:11] <joelio> heh, the joys of *learning how to properly use* teh software..
[16:11] <joelio> works fine for me, I can share my steps
[16:11] <joelio> you need to purge, yes, otherwise you may hit other issues.. it's not that much work, is it
[16:12] <joelio> you'll spend longer chasing your tail otherwise, truse me :)
[16:12] <thanasisk> please enlighten me :)
[16:16] <joelio> https://gist.github.com/joelio/01d017e0c8d49a6b4c20
[16:16] <joelio> n.b. I have 6 servers, vm-ds-0{1-6}
[16:16] <joelio> 3 are mons, (1,2,3)
[16:16] <joelio> all 6 have OSDs
[16:16] <joelio> I don't use ssd journals
[16:17] <joelio> each OSD host has 6 OSDs
[16:18] <joelio> the bit of bash that builds the OSDs could be more elegant, but hey ho.. it works :)
[16:19] <thanasisk> i only see a .txt file
[16:20] * thanasisk dumb
[16:24] * diegows (~diegows@190.190.11.42) has joined #ceph
[16:24] * vata (~vata@2607:fad8:4:6:156c:5b52:b19c:7128) has joined #ceph
[16:26] * tziOm (~bjornar@194.19.106.242) Quit (Remote host closed the connection)
[16:28] * akostopoulos_ (~akostopou@p5DDB97AF.dip0.t-ipconnect.de) has joined #ceph
[16:35] * KevinPerks (~Adium@cpe-066-026-252-218.triad.res.rr.com) has joined #ceph
[16:35] * PerlStalker (~PerlStalk@2620:d3:8000:192::70) has joined #ceph
[16:35] * sagelap (~sage@2600:1012:b00a:2d4:dc5c:a28:49c2:ca85) Quit (Read error: Connection reset by peer)
[16:35] * thanasisk (~akostopou@p5DDB8599.dip0.t-ipconnect.de) Quit (Ping timeout: 480 seconds)
[16:38] * nwf (~nwf@67.62.51.95) has joined #ceph
[16:40] * akostopoulos_ is now known as thanasisk
[16:42] * mozg (~oftc-webi@host217-46-236-49.in-addr.btopenworld.com) has joined #ceph
[16:42] <mozg> hello guys
[16:43] <mozg> i was wondering if anyone noticed excess number of hang tasks on the virtual machines running on qemu + rbd with 0.67.3?
[16:43] <mozg> i've recently migrated 3 Zimbra email servers and two file servers to ceph cluster
[16:44] <mozg> and noticed that the email servers have crashed after about 12 hours with a great deal of hang task messages
[16:44] <mozg> the file servers are also having hang tasks
[16:44] <mozg> is this a know issue?
[16:44] <mozg> or something I should report to the developers?
[16:45] <mikedawson> mozg: Yes
[16:46] <mikedawson> mozg: http://tracker.ceph.com/issues/6278
[16:46] <mozg> mikedawson: I see
[16:46] <mozg> is this on 0.67.3?
[16:47] <mikedawson> mozg: Try to turn off scrub and deep scrub, you may notice an improvement. Then let me know
[16:47] <mikedawson> mozg: yes, any 0.67.3 (but I've seen this issue for all Dumpling releases, and possibly Cuttlefish, too)
[16:48] <mozg> mikedawson: sorry, how do i do that? Also, would it not make my cluster less resilient?
[16:48] <mozg> mikedawson: I've not noticed these issues with cuttlefish release
[16:48] <mikedawson> mozg: 'ceph osd set noscrub' and 'ceph osd set nodeep-scrub'
[16:48] <mozg> mikedawson: have you noticed more slow requests on dumpling?
[16:48] <mozg> compared with cuttlefish?
[16:49] <mikedawson> mozg: it you want to re-enable either, s/set/unset/
[16:49] <mozg> mikedawson: is it safe to have scrubbing off?
[16:50] <mozg> i thought that scrubbing should fix issues that might cause data corruption
[16:50] <mikedawson> mozg: scrub and deep-scrub are attempt to identify issues and you won't get that functionality
[16:51] <mikedawson> mozg: I have to keep them off to keep my vms performing until the bug is fixed
[16:51] <mozg> i see
[16:51] <mozg> i will also do that
[16:52] <mozg> do you think it makes sense to reenable scrubbing during idle hours?
[16:52] <mozg> like overnight
[16:52] <mozg> and disable it during the day time?
[16:52] * scuttlemonkey (~scuttlemo@38.127.1.5) has joined #ceph
[16:52] * ChanServ sets mode +o scuttlemonkey
[16:52] <mikedawson> mozg: also check your disks for spindle contention. You can do 'iostat -x' and look for high %util.
[16:52] <joelio> mozg: you sure the extra servers you added aren't the issue, you're not maxing out IOPS?
[16:52] <joelio> ahh, snap
[16:53] <mikedawson> joelio: scrub and deep-scrub create the spindle contention
[16:53] <thanasisk> i can still mount an rbd block device but ceph does not work
[16:55] <mozg> mikedawson: i am only seeing high utilisation when there is extensive small io writes. This is typically when i am running io benchmarks. However, i've not used any benchmarks for a week, yet i had several crashed servers in the morning
[16:55] <mikedawson> mozg: when scrub or deep-scrub start, your spindles are likely thrashing. If you have rbd writeback cache on, you'll notice dips in write performance when the cache is written to the thrashing disks. Reads are slow whenever you encounter spindle contention.
[16:55] <mozg> i do have a number of backup jobs taking place overnight, which could cause higher load
[16:55] <mozg> mikedawson: I do have 512mb of cache for rbd
[16:55] <joelio> wow
[16:56] <mikedawson> mozg: whoa! 32MB is the default
[16:56] <mozg> mikedawson: you recon 512 is excessive?
[16:57] <mozg> i've got infiniband networking, which can do over 1gb/s on ip over infiniband link
[16:57] <joelio> doesn't sound too safe to me, regardless of how fast you can write
[16:57] <mozg> so I thougth that if i set caching to cover at least half of the bandwidth
[16:57] <mikedawson> mozg: so when your 512mb flushes (not sure if that is the right term), it'll take 16x longer than default
[16:58] <mikedawson> joelio: you are likely limited by your spindles, not your network
[16:58] <joelio> sure, hence why I said regardless of how fast :)
[16:58] * pyweb (~pyweb@209-197-139-35.cpe.distributel.net) has joined #ceph
[16:58] <mikedawson> sorry, that was meant for mozg
[16:58] <joelio> )
[16:59] <joelio> :)
[16:59] <mozg> mikedawson: i've got 2 osd servers with 8 spindles each
[16:59] <mozg> 16 in total
[16:59] <mozg> half of those use ssd for journaling
[16:59] <pyweb> can I ask some stupid question more or less related to ceph, I'M quite noob!
[16:59] <mozg> other half don't have the ssd journal
[16:59] <mozg> i will address that by purchasing more ssds
[17:00] * markl (~mark@tpsit.com) has joined #ceph
[17:00] <mozg> mikedawson: what do you think the rbd cache should be? do you think the default values are enough for majority of cases?
[17:00] <mozg> or should i lower it to something like 64 or 128mb?
[17:01] <mozg> pyweb: sure, go ahead,
[17:01] <mozg> i am sure for a lot of experienced ceph users a lot of questions sound stupid
[17:01] <mozg> but they help anyway
[17:01] <mikedawson> mozg: I have 72 spinners across 24 hosts on LACP bonded gigabit nics and it's all spindle contention for me (network utilization is very low)
[17:01] <mozg> as you can't learn without asking questions
[17:01] <mikedawson> pyweb: go for it
[17:01] <mozg> mikedawson: wow!
[17:02] <mozg> mikedawson: do you have high disk utilisation figures?
[17:02] * jcsp (~john@82-71-55-202.dsl.in-addr.zen.co.uk) Quit (Ping timeout: 480 seconds)
[17:02] <mikedawson> mozg: yes, sensu + graphite
[17:03] <pyweb> I was checking ceph this morning for grad school business intervention in management. I was searching for a information server,(files, DB and "stuff" to record in a business). ceph seems very nice to store information but I can't get my head around how you access the information. what kind ou user interface layer exist for ceph
[17:03] <mikedawson> mozg: ~10% to ~20% under normal conditions. Scrub, deep-scrub, recovery show complete saturation
[17:04] <joelio> pyweb: http://ceph.com/docs/next/architecture/
[17:04] <joelio> it depends on what data you want to store
[17:04] * sprachgenerator (~sprachgen@130.202.135.201) has joined #ceph
[17:04] <pyweb> the business uses windows server and directories to store data, everyone go and do their own structure and it's a big mess. the business deal with a considerable percentage of all bank and insurance in my region
[17:04] <joelio> VM images, media objects, files on a clistered file system etc.
[17:05] <pyweb> sorry poor english ;)
[17:06] <joelio> well, afaik, there are no windows based CephFS clients, which is probably more what you want, but sounds like you'd need to define your requirements more tightly
[17:06] <pyweb> they have case files with previous insurances solution to consults, project files, ideas files, rules files, law cases... all in a big horrible mess with a search engine like the one in windows wich takes forever to search around
[17:07] * malcolm_ (~malcolm@101.165.48.42) Quit (Ping timeout: 480 seconds)
[17:07] * kraken (~kraken@c-24-131-46-23.hsd1.ga.comcast.net) Quit (Remote host closed the connection)
[17:07] <pyweb> wee, just a best solution on linux idea
[17:07] <pyweb> sry
[17:07] * kraken (~kraken@c-24-131-46-23.hsd1.ga.comcast.net) has joined #ceph
[17:07] <joelio> pyweb: you don't need just cehp, you need a document management solution, like Alfresco or something
[17:07] <pyweb> ok i will search on that!
[17:08] <pyweb> right now I need words! just to know what to search on google
[17:08] * pmatulis (~peter@64.34.151.178) has joined #ceph
[17:08] <joelio> http://www.alfresco.com/ - other document management solutions are availale of course :)
[17:08] <mikedawson> mozg: I've been trying to get Inktank to prioritize http://tracker.ceph.com/issues/6278 Could you add some comments with your view of the issue?
[17:09] <pyweb> thx a lot joelio, I will find many "words" on their sites and will be able to get around from there!
[17:09] * madkiss (~madkiss@089144192241.atnat0001.highway.a1.net) Quit (Quit: Leaving.)
[17:10] <joelio> pyweb: just downloa Alfresco community edition and give it a whirl, it's quite good
[17:10] <joelio> ymmv :)
[17:10] <mozg> mikedawson: I've got two different osd servers. one server is fairly new and fast with 12 cores 24 ram - this server stays pretty idle most of the time with loads reaching to about 5 during heavy io benchmarking. The second server is an old Quad-core Xeon with 20gb of ram. This server shows loads in excess of 30 during the benchmarking and cpu utilisation reaching 900-100%
[17:10] <pyweb> I love the concept behind crush(if i remember) in ceph. I will be back if I find time to try it on small scale
[17:10] <mozg> mikedawson: will do
[17:10] * cofol19861 (~xwrj@110.90.119.113) has joined #ceph
[17:11] * claenjoy (~leggenda@37.157.33.36) Quit (Quit: Leaving.)
[17:12] * Tamil (~tamil@38.122.20.226) Quit (Read error: Connection reset by peer)
[17:12] * JustEra (~JustEra@89.234.148.11) Quit (Read error: Operation timed out)
[17:12] * Tamil (~tamil@38.122.20.226) has joined #ceph
[17:12] * clayg (~clayg@ec2-54-235-44-253.compute-1.amazonaws.com) Quit (Read error: Operation timed out)
[17:12] * phantomcircuit (~phantomci@covertinferno.org) Quit (Remote host closed the connection)
[17:13] * _Tassadar (~tassadar@tassadar.xs4all.nl) Quit (Read error: Operation timed out)
[17:13] * houkouonchi-work (~linux@12.248.40.138) Quit (Read error: Operation timed out)
[17:13] * clayg (~clayg@ec2-54-235-44-253.compute-1.amazonaws.com) has joined #ceph
[17:14] * phantomcircuit (~phantomci@covertinferno.org) has joined #ceph
[17:14] * houkouonchi-work (~linux@12.248.40.138) has joined #ceph
[17:14] * _Tassadar (~tassadar@tassadar.xs4all.nl) has joined #ceph
[17:14] * cofol1986 (~xwrj@110.90.119.113) Quit (Ping timeout: 480 seconds)
[17:15] * foosinn (~stefan@office.unitedcolo.de) Quit (Quit: Leaving)
[17:17] <mozg> mikedawson: just left the comment
[17:20] * BManojlovic (~steki@91.195.39.5) Quit (Quit: Ja odoh a vi sta 'ocete...)
[17:22] * AfC (~andrew@2001:44b8:31cb:d400:6e88:14ff:fe33:2a9c) has joined #ceph
[17:25] * frank9999 (~Frank@kantoor.transip.nl) Quit ()
[17:25] <dspano> Has anyone here tried the Samsung 840 pro SSDs on a Dell Perc h700?
[17:26] * dmsimard (~Adium@108.163.152.2) has joined #ceph
[17:33] * scuttlemonkey (~scuttlemo@38.127.1.5) Quit (Ping timeout: 480 seconds)
[17:34] * bandrus (~Adium@cpe-76-95-217-129.socal.res.rr.com) has joined #ceph
[17:34] * topro (~topro@host-62-245-142-50.customer.m-online.net) Quit (Quit: Konversation terminated!)
[17:35] * xiaoxi (~xiaoxi@192.102.204.38) has joined #ceph
[17:36] * xiaoxi (~xiaoxi@192.102.204.38) Quit ()
[17:37] * a (~a@pool-173-55-143-200.lsanca.fios.verizon.net) has joined #ceph
[17:38] * a is now known as Guest6825
[17:41] <jcfischer> we are using the Samsung 840 in Quanta servers
[17:45] * alram (~alram@38.122.20.226) has joined #ceph
[17:47] * Vjarjadian (~IceChat77@94.4.28.26) has joined #ceph
[17:48] * diegows (~diegows@190.190.11.42) Quit (Ping timeout: 480 seconds)
[17:51] <thanasisk> anyone here used glusterFS? what are the pros and cons vs ceph ?
[17:51] * dspano (~dspano@rrcs-24-103-221-202.nys.biz.rr.com) Quit (Read error: Operation timed out)
[17:52] * dspano (~dspano@rrcs-24-103-221-202.nys.biz.rr.com) has joined #ceph
[17:53] * thanasisk (~akostopou@p5DDB97AF.dip0.t-ipconnect.de) Quit (Quit: Leaving)
[17:54] * Guest6825 (~a@pool-173-55-143-200.lsanca.fios.verizon.net) Quit (Quit: This computer has gone to sleep)
[17:57] <wrencsok> mozg: those hung vm's are they doing database type operations? where potential table locking could be locking a significant number of your osd's for a single insert since some db's like SQL will lock on the entire table. or do they have lots of outstanding io's against the cluster perchance? I can DoS my ceph clusters with outstanding io's. I have very poor sql type db performance. hopefully i can tune and configure that problem away. i s
[17:58] <mozg> wrencsok: not really sure as this has started happening recently
[17:58] <mozg> i've noticed this after upgrading to 0.67.3
[17:58] * oliver (~oliver@p4FFFECD5.dip0.t-ipconnect.de) has left #ceph
[17:58] <mozg> and not done much debugging
[17:58] <mozg> not sure what is the default configuration behaviour of zimbra email servers
[17:59] <mozg> the file servers use the standar sql server config with ownCloud service running on top
[17:59] <mozg> wrencsok: do you know what could be done to address the DoS issue you've refered to?
[18:00] <wrencsok> ah I don't know either. ok. I do know for small io's and db related workflows performance against my clusters is not well, in other areas its impressivley fast. i recently updated to that build too. but i have those two issues i described. hopefully i can tune those out. not sure yet. its on my plate internally to pull out the debuggers and profilers do deep dive into that and figure out why.
[18:00] * jjgalvez (~jjgalvez@ip72-193-217-254.lv.lv.cox.net) has joined #ceph
[18:01] * AfC (~andrew@2001:44b8:31cb:d400:6e88:14ff:fe33:2a9c) Quit (Quit: Leaving.)
[18:02] * madkiss (~madkiss@089144192241.atnat0001.highway.a1.net) has joined #ceph
[18:03] <wrencsok> i have theory's on the outstanding io's and aborting them before they are flushed from the journals. and with the db problem and potential locking issues. but i'll know later in the week. had similar problems to at a past startup.
[18:03] * sagelap (~sage@2600:1010:b012:ccdd:dc5c:a28:49c2:ca85) has joined #ceph
[18:08] * dpippenger (~riven@cpe-76-166-208-83.socal.res.rr.com) has joined #ceph
[18:09] * jcfischer (~fischer@macjcf.switch.ch) Quit (Quit: jcfischer)
[18:10] * Cube (~Cube@cpe-76-95-217-129.socal.res.rr.com) has joined #ceph
[18:11] * adam2 (~adam@46-65-111-12.zone16.bethere.co.uk) has joined #ceph
[18:11] * angdraug (~angdraug@204.11.231.50.static.etheric.net) has joined #ceph
[18:16] * sjustlaptop (~sam@204.57.119.28) has joined #ceph
[18:16] * sagelap (~sage@2600:1010:b012:ccdd:dc5c:a28:49c2:ca85) Quit (Read error: Connection reset by peer)
[18:17] * adam1 (~adam@46-65-111-12.zone16.bethere.co.uk) Quit (Ping timeout: 480 seconds)
[18:18] * sjustlaptop1 (~sam@172.56.21.10) has joined #ceph
[18:21] * xmltok_ (~xmltok@pool101.bizrate.com) has joined #ceph
[18:24] * sjustlaptop (~sam@204.57.119.28) Quit (Ping timeout: 480 seconds)
[18:25] * a_ (~a@209.12.169.218) has joined #ceph
[18:25] <dspano> jcfischer: I'm getting extremely crappy performance with these in a Dell R515 with a Perc H700. I.E - 30-70MB/s and 1200 iops. I've got a feeling the Dell controller doesn't like them because they're not Dell drives.
[18:26] <dspano> jcfischer: The 15k 600GB Dell SAS drives, are much faster on the same controller.
[18:26] * xarses (~andreww@c-71-202-167-197.hsd1.ca.comcast.net) Quit (Ping timeout: 480 seconds)
[18:27] * skm (~smiley@205.153.36.170) has joined #ceph
[18:27] * BManojlovic (~steki@fo-d-130.180.254.37.targo.rs) has joined #ceph
[18:28] * dpippenger (~riven@cpe-76-166-208-83.socal.res.rr.com) Quit (Quit: Leaving.)
[18:29] <dmsimard> dspano: Have you tried various I/O scenarios ? I've had issues in the past with raid cards that came with Dell C6220's - really bad random write performance (MySQL database, for instance)
[18:30] <dmsimard> dspano: Ended up having the cards replaced, they were forced write through and nothing I could do about it
[18:34] <dmsimard> replaced by another model, that is
[18:37] * xarses (~andreww@204.11.231.50.static.etheric.net) has joined #ceph
[18:39] * xdeller (~xdeller@91.218.144.129) Quit (Ping timeout: 480 seconds)
[18:41] * tsnider (~tsnider@nat-216-240-30-23.netapp.com) has joined #ceph
[18:41] * ircolle (~Adium@c-67-165-237-235.hsd1.co.comcast.net) has joined #ceph
[18:42] * gregaf1 (~Adium@38.122.20.226) Quit (Quit: Leaving.)
[18:43] * gregaf (~Adium@2607:f298:a:607:5eb:7fe6:865e:ac5) has joined #ceph
[18:48] <dspano> dmsimard: I tried write through, writeback, all the different read policies, etc.. My next step is to try them in a different machine. I've been thinking about a different card in the future. What one did you replace it with?
[18:48] <dmsimard> I believe we're using LSI 9266-8i's
[18:49] <dspano> dmsimard: I'll take a look at them. Thanks.
[18:49] * xdeller (~xdeller@91.218.144.129) has joined #ceph
[18:50] * wrencsok (~wrencsok@wsip-174-79-34-244.ph.ph.cox.net) has left #ceph
[18:50] * wrencsok (~wrencsok@wsip-174-79-34-244.ph.ph.cox.net) has joined #ceph
[18:50] * wrencsok (~wrencsok@wsip-174-79-34-244.ph.ph.cox.net) has left #ceph
[18:51] <r0r_taga> i'd use the rebadged SLI 3ware cards or the MegaRAIDs, imo
[18:51] <r0r_taga> SLI = LSI
[18:51] * wrencsok (~wrencsok@wsip-174-79-34-244.ph.ph.cox.net) has joined #ceph
[18:51] * joelio curses the pg_num default of 8
[18:51] <joelio> why?
[18:52] <joelio> even the docs say this is not recommended.. so why set it? Sane defaults guys.. :)
[18:52] * mjblw (~mbaysek@wsip-174-79-34-244.ph.ph.cox.net) Quit (Remote host closed the connection)
[18:54] * odyssey4me (~odyssey4m@165.233.71.2) Quit (Ping timeout: 480 seconds)
[18:54] * yehuda_hm (~yehuda@2602:306:330b:1410:9de:8b13:f9a3:9b8) Quit (Ping timeout: 480 seconds)
[18:57] * sagelap (~sage@2600:1010:b022:4ad2:2cc7:832:e20a:387e) has joined #ceph
[18:58] * mozg (~oftc-webi@host217-46-236-49.in-addr.btopenworld.com) Quit (Remote host closed the connection)
[19:00] * sagelap (~sage@2600:1010:b022:4ad2:2cc7:832:e20a:387e) Quit (Read error: No route to host)
[19:01] * sagelap (~sage@2600:1010:b022:4ad2:11f6:aa83:5d3b:eb99) has joined #ceph
[19:02] * Vjarjadian (~IceChat77@94.4.28.26) Quit (Quit: If you think nobody cares, try missing a few payments)
[19:03] * nhm (~nhm@184-97-187-196.mpls.qwest.net) has joined #ceph
[19:03] * ChanServ sets mode +o nhm
[19:06] * yehuda_hm (~yehuda@99-48-177-65.lightspeed.irvnca.sbcglobal.net) has joined #ceph
[19:07] * [1]pyweb (~pyweb@209-197-139-35.cpe.distributel.net) has joined #ceph
[19:10] * madkiss1 (~madkiss@srvnet-01-055.ikbnet.co.at) has joined #ceph
[19:14] * pyweb (~pyweb@209-197-139-35.cpe.distributel.net) Quit (Ping timeout: 480 seconds)
[19:14] * [1]pyweb is now known as pyweb
[19:17] * madkiss (~madkiss@089144192241.atnat0001.highway.a1.net) Quit (Ping timeout: 480 seconds)
[19:18] * ismell (~ismell@host-64-17-89-79.beyondbb.com) has joined #ceph
[19:20] <peetaur> so, should we feel safe enough with btfs, an experimental file system, because if it blows up, the cluster can recover just like if a disk died some other way? or is a single btrfs device OSD safe enough?
[19:21] <decede> give it a go and let us know ;)
[19:21] <peetaur> I've had btfs on my root system at home, and had a few meltdowns in kernel 3.4, and using an old fs updated to newest kernel still had quirks (and total hang after updating to kernel 3.10, but no hangs with a new mkfs.btrfs).
[19:21] <peetaur> well, why is it suggested all over the place?
[19:22] <joelio> faster mechanisms than using older fs's with XATTRS
[19:22] <joelio> but not (imho) battle hardened
[19:22] <sjustlaptop1> peetaur: only catch is that the btrfs fses aren't really independent, if a particular bug kills one disk you might see the same bug on other disks
[19:22] <joelio> although I guess that's open to interprestation
[19:23] <peetaur> if I decided to use btrfs, I'd make it one btrfs fs per OSD, so then they are independent ... right? (other than kernel bugs)
[19:23] <sjustlaptop1> other than kernel bugs, sure
[19:23] <sjustlaptop1> I am merely pointing out that if a kernel bug hits one osd, it may hit others as well
[19:23] * jbd_ (~jbd_@2001:41d0:52:a00::77) has left #ceph
[19:24] <dmsimard> There's this old saying somewhere, saying btrfs is always 2 years from being production ready :)
[19:25] * sagelap (~sage@2600:1010:b022:4ad2:11f6:aa83:5d3b:eb99) Quit (Quit: Leaving.)
[19:25] <peetaur> okay then side question, does anyone have any benchmarks on comparing ext4 with separated journal to btrfs?
[19:25] <sjustlaptop1> you do need a journal for both
[19:25] <sjustlaptop1> also, we usually suggest xfs over ext4
[19:25] <dmsimard> peetaur: You might find these interesting: http://ceph.com/community/ceph-performance-part-1-disk-controller-write-throughput/ /// http://ceph.com/community/ceph-performance-part-2-write-throughput-without-ssd-journals/
[19:26] <dmsimard> The only sad thing is that these are on older builds of ceph
[19:26] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) has joined #ceph
[19:26] * ChanServ sets mode +v andreask
[19:26] <dmsimard> I'd be interested to see the same tests but done on dumpling
[19:28] * thomnico (~thomnico@mtl93-h02-31-39-189-146.dsl.sta.abo.bbox.fr) has joined #ceph
[19:28] * PerlStalker (~PerlStalk@2620:d3:8000:192::70) Quit (Quit: ...)
[19:29] <peetaur> ah I see, so the test (part 1) I read before was with separate SSD journals, and part 2 is without
[19:30] <nhm> peetaur: btw, those are pretty old
[19:33] * jcsp (~john@82-71-55-202.dsl.in-addr.zen.co.uk) has joined #ceph
[19:33] <peetaur> yes but without new info, ... what should I conclude?
[19:34] <peetaur> how many journals can I put on one SSD?
[19:36] <peetaur> is it a good idea to build something like: os software raid1 on 2 ssds, the rest of the space for journals, 6 OSD disks, with 3 partitions on each ssd for their journals, no raid for journals?
[19:37] <nhm> peetaur: As many as you want. There are two things to keep in mind: 1) Ceph writes all data to the journal, so it can become a bottleneck quickly with multiple OSDs using the same journal device. 2) An OSD will stop working when a journal device fails. The more OSDs with journals on 1 SSD, the greater the effect when the SSD fails.
[19:38] <peetaur> if it writes all data to the SSD, that sounds like the SSD will run out of writes too fast
[19:40] <nhm> peetaur: Maybe, it depnds on the number of disks and what the bottleneck is. IE if you only have 1 10GbE link and you have 3 400MB/s SSDs in the node for journals, adding more SSDs won't likely improve large write performance much because the SSDs are already faster in aggregate than the network.
[19:43] <peetaur> I mean the wear leveling won't be able to keep the SSD alive, and it'll die an old age death in a short time
[19:45] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) Quit (Ping timeout: 480 seconds)
[19:46] <peetaur> here's a line from an SSD I used for 2 years as a ZFS ZIL 202 Perc_Rated_Life_Used 0x0018 044 044 001 Old_age Offline - 56
[19:46] * ScOut3R_ (~scout3r@4E5C2305.dsl.pool.telekom.hu) Quit (Remote host closed the connection)
[19:48] * dspano (~dspano@rrcs-24-103-221-202.nys.biz.rr.com) Quit (Read error: Operation timed out)
[19:49] * dspano (~dspano@rrcs-24-103-221-202.nys.biz.rr.com) has joined #ceph
[19:50] * buck (~buck@c-24-6-91-4.hsd1.ca.comcast.net) has joined #ceph
[19:50] * jcfischer (~fischer@macjcf.switch.ch) has joined #ceph
[19:50] * rturk-away is now known as rturk
[19:52] * KevinPerks (~Adium@cpe-066-026-252-218.triad.res.rr.com) Quit (Quit: Leaving.)
[19:52] * rturk is now known as rturk-away
[19:52] * rturk-away is now known as rturk
[19:58] * sjustlaptop1 (~sam@172.56.21.10) Quit (Read error: Connection reset by peer)
[20:06] * dpippenger (~riven@tenant.pas.idealab.com) has joined #ceph
[20:13] <jlogan> I'm looking at a ceph -s report where I have: 14 pgs stuck unclean. It has been this way since Dumpling. http://pastebin.com/YX2Dk7sR
[20:13] <jlogan> What can I do to try to get the health monitor to return clena
[20:18] * pyweb (~pyweb@209-197-139-35.cpe.distributel.net) Quit (Quit: HydraIRC -> http://www.hydrairc.com <- *I* use it, so it must be good!)
[20:18] * nwf (~nwf@67.62.51.95) Quit (Ping timeout: 480 seconds)
[20:24] * roald (~oftc-webi@87.209.150.214) has joined #ceph
[20:28] * thomnico (~thomnico@mtl93-h02-31-39-189-146.dsl.sta.abo.bbox.fr) Quit (Quit: Ex-Chat)
[20:51] * davidzlap (~Adium@ip68-5-239-214.oc.oc.cox.net) has joined #ceph
[20:56] * sjm (~sjm@dhcp-108-168-18-236.cable.user.start.ca) has left #ceph
[21:02] * sarob (~sarob@nat-dip28-wl-b.cfw-a-gci.corp.yahoo.com) has joined #ceph
[21:03] * sarob (~sarob@nat-dip28-wl-b.cfw-a-gci.corp.yahoo.com) Quit (Remote host closed the connection)
[21:04] * sarob (~sarob@nat-dip28-wl-b.cfw-a-gci.corp.yahoo.com) has joined #ceph
[21:05] * sarob_ (~sarob@nat-dip28-wl-b.cfw-a-gci.corp.yahoo.com) has joined #ceph
[21:05] * sarob (~sarob@nat-dip28-wl-b.cfw-a-gci.corp.yahoo.com) Quit (Read error: Connection reset by peer)
[21:05] * sarob_ (~sarob@nat-dip28-wl-b.cfw-a-gci.corp.yahoo.com) Quit (Remote host closed the connection)
[21:05] * sarob (~sarob@nat-dip4.cfw-a-gci.corp.yahoo.com) has joined #ceph
[21:06] * sagelap (~sage@63-237-196-66.dia.static.qwest.net) has joined #ceph
[21:12] * mikedawson (~chatzilla@23-25-46-97-static.hfc.comcastbusiness.net) Quit (Ping timeout: 480 seconds)
[21:15] * Tamil1 (~Adium@cpe-108-184-67-162.socal.res.rr.com) has joined #ceph
[21:20] * diegows (~diegows@200.68.116.185) has joined #ceph
[21:23] * sjustlaptop (~sam@172.56.21.10) has joined #ceph
[21:30] * sagelap (~sage@63-237-196-66.dia.static.qwest.net) Quit (Ping timeout: 480 seconds)
[21:31] * rturk is now known as rturk-away
[21:31] * jcsp (~john@82-71-55-202.dsl.in-addr.zen.co.uk) Quit (Ping timeout: 480 seconds)
[21:34] * jcsp (~john@82-71-55-202.dsl.in-addr.zen.co.uk) has joined #ceph
[21:35] * \ask (~ask@oz.develooper.com) Quit (Ping timeout: 480 seconds)
[21:36] * \ask (~ask@oz.develooper.com) has joined #ceph
[21:37] * JustEra (~JustEra@ALille-555-1-93-207.w90-7.abo.wanadoo.fr) has joined #ceph
[21:43] * houkouonchi-home (~linux@pool-71-165-8-99.lsanca.fios.verizon.net) has joined #ceph
[21:43] * KevinPerks (~Adium@cpe-066-026-252-218.triad.res.rr.com) has joined #ceph
[21:44] * scuttlemonkey (~scuttlemo@204.57.119.28) has joined #ceph
[21:44] * ChanServ sets mode +o scuttlemonkey
[21:48] * madkiss1 (~madkiss@srvnet-01-055.ikbnet.co.at) Quit (Quit: Leaving.)
[21:48] * rturk-away is now known as rturk
[21:50] * JustEra (~JustEra@ALille-555-1-93-207.w90-7.abo.wanadoo.fr) Quit (Quit: This computer has gone to sleep)
[21:55] * sjustlaptop (~sam@172.56.21.10) Quit (Ping timeout: 480 seconds)
[22:02] * `Kevin (~kevin@sys0.chi0.birchbox.com) has joined #ceph
[22:04] <loicd> glowell: ping ?
[22:04] <glowell> Hello
[22:04] * JustEra (~JustEra@ALille-555-1-93-207.w90-7.abo.wanadoo.fr) has joined #ceph
[22:08] * sagelap (~sage@63-237-196-66.dia.static.qwest.net) has joined #ceph
[22:11] * mtl1 (~Adium@c-67-176-54-246.hsd1.co.comcast.net) has joined #ceph
[22:18] * madkiss (~madkiss@srvnet-01-055.ikbnet.co.at) has joined #ceph
[22:25] * scuttlemonkey (~scuttlemo@204.57.119.28) Quit (Ping timeout: 480 seconds)
[22:28] * allsystemsarego (~allsystem@188.25.134.128) Quit (Quit: Leaving)
[22:31] * madkiss (~madkiss@srvnet-01-055.ikbnet.co.at) Quit (Ping timeout: 480 seconds)
[22:31] * sagelap1 (~sage@63-237-196-66.dia.static.qwest.net) has joined #ceph
[22:32] <mtl1> Hi, quick question. Is there any reason I can't map a cloned boot image with rbd? When I try, I get "rbd: add failed: (6) No such device or address". I'm able to map block device images just fine though.
[22:36] * sagelap (~sage@63-237-196-66.dia.static.qwest.net) Quit (Ping timeout: 480 seconds)
[22:38] <joshd> mtl1: are you using linux 3.10 ? format 2 support is pretty new in the kernel rbd driver
[22:39] <mtl1> I just installed 3.9.
[22:45] * sjustlaptop (~sam@172.56.21.10) has joined #ceph
[22:47] <loicd> glowell: hi :-) Sorry for the lag : I'm used to only react when my nick is included in the sentence.
[22:47] <glowell> Sorry, what's up
[22:48] <glowell> loicd: ^^
[22:48] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) has joined #ceph
[22:48] * ChanServ sets mode +v andreask
[22:48] <loicd> glowell: I would like to have your opinion on "fabfile: add python-nose build dependency " https://github.com/ceph/autobuild-ceph/pull/2 . I'm not sure this is the right solution...
[22:49] <glowell> loicd: I'll have a look
[22:49] <loicd> cool :-)
[22:50] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) has left #ceph
[22:52] <athrift> Is there an easy way to display the journal path and status, kind of like "ceph osd tree" but that shows the journal path ?
[22:53] * sjustlaptop (~sam@172.56.21.10) Quit (Ping timeout: 480 seconds)
[22:57] * scuttlemonkey (~scuttlemo@204.57.119.28) has joined #ceph
[22:57] * ChanServ sets mode +o scuttlemonkey
[22:59] * sjm (~sjm@dhcp-108-168-18-236.cable.user.start.ca) has joined #ceph
[22:59] <nhm> yehuda_hm: assuming my results are correct, wip-6286 is about 8x faster for 4k PUTS, at least initially.
[22:59] <joshd> athrift: not for all osds, no. best way would probably involve doing a 'config get' on all their admin sockets
[23:00] <yehuda_hm> nhm: cool!
[23:00] <yehuda_hm> how's the cpu doing?
[23:00] <athrift> joshd: thanks Josh. We will work something out around that
[23:00] <nhm> yehuda_hm: radosgw is using about 400% CPU
[23:01] <nhm> yehuda_hm: spread out between malloc, memcpy, free, some various other bits.
[23:01] * tziOm (~bjornar@ti0099a340-dhcp0395.bb.online.no) has joined #ceph
[23:01] <gregaf> geeze, how many calls to what size lists were we doing?
[23:02] <yehuda_hm> nhm: 400%? how many cpus these machine have?
[23:02] <yehuda_hm> also, how many ops per second are we talking?
[23:06] <yehuda_hm> gregaf: we were doing 1 call for every operation to 10k sized list
[23:06] <yehuda_hm> well, at least 1 call, might be more
[23:06] * DarkAceZ (~BillyMays@50.107.55.36) Quit (Ping timeout: 480 seconds)
[23:07] <gregaf> and that was able to take up 87.5% of CPU time spent on the request? That seems odd
[23:08] * mtl1 (~Adium@c-67-176-54-246.hsd1.co.comcast.net) has left #ceph
[23:09] * sarob_ (~sarob@nat-dip28-wl-b.cfw-a-gci.corp.yahoo.com) has joined #ceph
[23:10] <nhm> yehuda_hm: we have 6 cores on the burnupi nodes
[23:11] <nhm> yehuda_hm: right now we are doing about 680ops/s
[23:11] <nhm> yehuda_hm: up from 80 ops/s previously.
[23:11] <nhm> actually, it's slowing down, about 650ops/s now.
[23:11] * scuttlemonkey (~scuttlemo@204.57.119.28) Quit (Ping timeout: 480 seconds)
[23:12] <yehuda_hm> nhm: you should really turn off or quiet down logging
[23:13] <yehuda_hm> also, did you increase the thread pool size?
[23:13] <nhm> yehuda_hm: eventually, but I wanted to run this test the same as the previous one.
[23:13] <yehuda_hm> yeah
[23:13] <yehuda_hm> did the cpu load change with the ops slowing down?
[23:14] <nhm> yehuda_hm: what's the default RGW logging level?
[23:14] <yehuda_hm> 20
[23:14] <nhm> yehuda_hm: Is that more or less kitchen sink?
[23:14] * dmick (~androirc@63-237-196-66.dia.static.qwest.net) has joined #ceph
[23:14] * ChanServ sets mode +o dmick
[23:15] <yehuda_hm> yeah, that's the highest effective value
[23:15] * dmick (~androirc@63-237-196-66.dia.static.qwest.net) Quit ()
[23:15] <nhm> yehuda_hm: tough to say regarding CPU load. It's fluxating a bit. So far it's stll about the same.
[23:15] <nhm> maybe a bit less.
[23:16] * sagelap (~sage@63-237-196-66.dia.static.qwest.net) has joined #ceph
[23:16] * sagelap1 (~sage@63-237-196-66.dia.static.qwest.net) Quit (Quit: Leaving.)
[23:16] * sagelap (~sage@63-237-196-66.dia.static.qwest.net) Quit ()
[23:16] <nhm> yehuda_hm: might be worth changing it to be lower if we expect it has a big impact on performance.
[23:16] <nhm> yehuda_hm: IE if we think most users aren't going to use that kind of debugging.
[23:16] * sarob (~sarob@nat-dip4.cfw-a-gci.corp.yahoo.com) Quit (Ping timeout: 480 seconds)
[23:17] <yehuda_hm> nhm: well, let's see what your investigation comes up with
[23:17] <yehuda_hm> nhm: that's exactly why the default is that high
[23:17] <yehuda_hm> we were targeting at stability over performance, it helped with trimming the bugs a lot
[23:17] * AndroUser (~androirc@63-237-196-66.dia.static.qwest.net) has joined #ceph
[23:17] * sarob_ (~sarob@nat-dip28-wl-b.cfw-a-gci.corp.yahoo.com) Quit (Ping timeout: 480 seconds)
[23:18] * AndroUser is now known as dmick
[23:19] <loicd> glowell: It's late here in France but I'll read you in the morning :-) Night !
[23:24] * madkiss (~madkiss@srvnet-01-055.ikbnet.co.at) has joined #ceph
[23:28] * dspano (~dspano@rrcs-24-103-221-202.nys.biz.rr.com) Quit (Quit: Leaving)
[23:30] * ScOut3R (~ScOut3R@4E5C2305.dsl.pool.telekom.hu) has joined #ceph
[23:31] * ScOut3R_ (~scout3r@4E5C2305.dsl.pool.telekom.hu) has joined #ceph
[23:31] * Cube (~Cube@cpe-76-95-217-129.socal.res.rr.com) Quit (Quit: Leaving.)
[23:32] * madkiss (~madkiss@srvnet-01-055.ikbnet.co.at) Quit (Ping timeout: 480 seconds)
[23:35] * mnash (~chatzilla@66-194-114-178.static.twtelecom.net) Quit (Ping timeout: 480 seconds)
[23:38] * gregmark (~Adium@68.87.42.115) Quit (Quit: Leaving.)
[23:42] * sprachgenerator (~sprachgen@130.202.135.201) Quit (Quit: sprachgenerator)
[23:42] * sjm (~sjm@dhcp-108-168-18-236.cable.user.start.ca) has left #ceph
[23:46] * roald (~oftc-webi@87.209.150.214) Quit (Quit: Page closed)
[23:47] * AndroUser (~androirc@63-237-196-66.dia.static.qwest.net) has joined #ceph
[23:48] * AndroUser (~androirc@63-237-196-66.dia.static.qwest.net) Quit ()
[23:50] * ScOut3R (~ScOut3R@4E5C2305.dsl.pool.telekom.hu) Quit (Remote host closed the connection)
[23:51] * Cube (~Cube@12.248.40.138) has joined #ceph
[23:52] * dmick (~androirc@63-237-196-66.dia.static.qwest.net) Quit (Ping timeout: 480 seconds)
[23:53] * tziOm (~bjornar@ti0099a340-dhcp0395.bb.online.no) Quit (Remote host closed the connection)
[23:54] * carif (~mcarifio@ip-207-145-81-212.nyc.megapath.net) has joined #ceph
[23:57] * ScOut3R (~ScOut3R@4E5C2305.dsl.pool.telekom.hu) has joined #ceph
[23:58] * bandrus (~Adium@cpe-76-95-217-129.socal.res.rr.com) Quit (Quit: Leaving.)
[23:59] * ScOut3R (~ScOut3R@4E5C2305.dsl.pool.telekom.hu) Quit (Remote host closed the connection)
[23:59] * bandrus (~Adium@cpe-76-95-217-129.socal.res.rr.com) has joined #ceph

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