#ceph IRC Log

Index

IRC Log for 2015-10-28

Timestamps are in GMT/BST.

[0:00] * TMM (~hp@178-84-46-106.dynamic.upc.nl) Quit (Ping timeout: 480 seconds)
[0:00] <moloney> Looks like I have joined the group of people suffering from inexplicable loss of all write speed :(
[0:02] * TMM (~hp@178-84-46-106.dynamic.upc.nl) has joined #ceph
[0:02] * abhi (~yaaic@zz2012427835719DFD82.userreverse.dion.ne.jp) Quit (Ping timeout: 480 seconds)
[0:03] * yguang11 (~yguang11@66.228.162.44) Quit (Remote host closed the connection)
[0:03] * yguang11 (~yguang11@66.228.162.44) has joined #ceph
[0:04] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[0:04] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[0:07] * lcurtis (~lcurtis@47.19.105.250) Quit (Remote host closed the connection)
[0:07] * abhi (~yaaic@zz2012427835719DFD82.userreverse.dion.ne.jp) has joined #ceph
[0:08] * nickpappin (~nickpappi@wl-nat109.it.wsu.edu) has joined #ceph
[0:10] * TMM (~hp@178-84-46-106.dynamic.upc.nl) Quit (Ping timeout: 480 seconds)
[0:16] * sudocat (~dibarra@192.185.1.20) Quit (Ping timeout: 480 seconds)
[0:28] * Gorazd (Gorazd@89-212-99-37.dynamic.t-2.net) Quit (Ping timeout: 480 seconds)
[0:28] * daviddcc (~dcasier@84.197.151.77.rev.sfr.net) Quit (Ping timeout: 480 seconds)
[0:29] * daviddcc (~dcasier@84.197.151.77.rev.sfr.net) has joined #ceph
[0:30] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[0:30] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[0:38] * daviddcc (~dcasier@84.197.151.77.rev.sfr.net) Quit (Ping timeout: 480 seconds)
[0:39] * moore (~moore@64.202.160.88) Quit (Remote host closed the connection)
[0:44] * xarses (~xarses@zz2012427835719DFD82.userreverse.dion.ne.jp) has joined #ceph
[0:44] * xarses (~xarses@zz2012427835719DFD82.userreverse.dion.ne.jp) Quit (Read error: Connection reset by peer)
[0:45] * xarses (~xarses@zz2012427835719DFD82.userreverse.dion.ne.jp) has joined #ceph
[0:45] * xarses (~xarses@zz2012427835719DFD82.userreverse.dion.ne.jp) Quit (Read error: Connection reset by peer)
[0:46] * abhi (~yaaic@zz2012427835719DFD82.userreverse.dion.ne.jp) Quit (Ping timeout: 480 seconds)
[0:47] * xarses (~xarses@zz2012427835719DFD82.userreverse.dion.ne.jp) has joined #ceph
[0:47] * xarses (~xarses@zz2012427835719DFD82.userreverse.dion.ne.jp) Quit (Remote host closed the connection)
[0:48] * xarses (~xarses@zz2012427835719DFD82.userreverse.dion.ne.jp) has joined #ceph
[0:48] * xarses (~xarses@zz2012427835719DFD82.userreverse.dion.ne.jp) Quit (Read error: Connection reset by peer)
[0:49] <cholcombe> is there some way to get the cache mode a pool is using other than ceph osd dump and parsing it?
[0:49] * xarses (~xarses@zz2012427835719DFD82.userreverse.dion.ne.jp) has joined #ceph
[0:49] * erhudy (uid89730@id-89730.ealing.irccloud.com) Quit (Quit: Connection closed for inactivity)
[0:49] * xarses (~xarses@zz2012427835719DFD82.userreverse.dion.ne.jp) Quit (Remote host closed the connection)
[0:50] * xarses (~xarses@zz2012427835719DFD82.userreverse.dion.ne.jp) has joined #ceph
[0:57] * xarses_ (~xarses@2402:c800:ff64:300:6e88:14ff:fe71:3fc4) has joined #ceph
[0:59] * andreww (~xarses@163.138.224.216) has joined #ceph
[1:00] * andreww (~xarses@163.138.224.216) Quit (Remote host closed the connection)
[1:04] * andreww (~xarses@2402:c800:ff64:300:4661:62d9:c96f:eb1c) has joined #ceph
[1:04] * xarses (~xarses@zz2012427835719DFD82.userreverse.dion.ne.jp) Quit (Ping timeout: 480 seconds)
[1:04] * abhi (~yaaic@zz2012427835719DFD82.userreverse.dion.ne.jp) has joined #ceph
[1:06] * xarses (~xarses@2402:c800:ff64:300:4661:62d9:c96f:eb1c) has joined #ceph
[1:07] * xarses_ (~xarses@2402:c800:ff64:300:6e88:14ff:fe71:3fc4) Quit (Ping timeout: 480 seconds)
[1:09] * rendar (~I@host146-182-dynamic.252-95-r.retail.telecomitalia.it) Quit (Quit: std::lower_bound + std::less_equal *works* with a vector without duplicates!)
[1:12] * xarses_ (~xarses@163.138.224.216) has joined #ceph
[1:12] * xarses_ (~xarses@163.138.224.216) Quit (Remote host closed the connection)
[1:13] * xarses_ (~xarses@2402:c800:ff64:300:4661:62d9:c96f:eb1c) has joined #ceph
[1:14] * andreww (~xarses@2402:c800:ff64:300:4661:62d9:c96f:eb1c) Quit (Ping timeout: 480 seconds)
[1:14] * hgichon (~hgichon@220.90.135.162) has joined #ceph
[1:16] * nickpappin (~nickpappi@wl-nat109.it.wsu.edu) Quit (Quit: nickpappin)
[1:19] * terje (~root@135.109.216.239) has joined #ceph
[1:20] * xarses (~xarses@2402:c800:ff64:300:4661:62d9:c96f:eb1c) Quit (Ping timeout: 480 seconds)
[1:22] * abhi (~yaaic@zz2012427835719DFD82.userreverse.dion.ne.jp) Quit (Ping timeout: 480 seconds)
[1:23] * brutuscat (~brutuscat@105.34.133.37.dynamic.jazztel.es) Quit (Remote host closed the connection)
[1:24] * brutuscat (~brutuscat@105.34.133.37.dynamic.jazztel.es) has joined #ceph
[1:28] * xarses_ (~xarses@2402:c800:ff64:300:4661:62d9:c96f:eb1c) Quit (Ping timeout: 480 seconds)
[1:36] * oms101 (~oms101@p20030057EA019B00C6D987FFFE4339A1.dip0.t-ipconnect.de) Quit (Ping timeout: 480 seconds)
[1:39] * kafkane (~kafkane@00020e74.user.oftc.net) has joined #ceph
[1:39] * kafkane (~kafkane@00020e74.user.oftc.net) has left #ceph
[1:39] * LeaChim (~LeaChim@host86-143-17-156.range86-143.btcentralplus.com) Quit (Ping timeout: 480 seconds)
[1:42] * LeonaS (~quaLity@130.193.234.134) has joined #ceph
[1:43] * LeonaS (~quaLity@130.193.234.134) Quit ()
[1:45] * oms101 (~oms101@p20030057EA015F00C6D987FFFE4339A1.dip0.t-ipconnect.de) has joined #ceph
[1:50] * fdmanana (~fdmanana@2001:8a0:6dfd:6d01:d980:d368:6369:fcaa) Quit (Ping timeout: 480 seconds)
[1:50] * sudocat (~dibarra@2602:306:8bc7:4c50::1f) has joined #ceph
[1:52] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[1:52] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[1:52] * abhi (~yaaic@zz2012427835719DFD82.userreverse.dion.ne.jp) has joined #ceph
[1:55] * loicd reading http://community.redhat.com/blog/2015/10/ceph-launches-new-community-advisory-board/
[1:58] * yguang11 (~yguang11@66.228.162.44) Quit ()
[2:00] * abhi (~yaaic@zz2012427835719DFD82.userreverse.dion.ne.jp) Quit (Ping timeout: 480 seconds)
[2:02] * segutier (~segutier@sfo-vpn1.shawnlower.net) Quit (Ping timeout: 480 seconds)
[2:09] * jcsp (~jspray@2402:c800:ff64:300:df3d:9e5b:2e44:13e7) has joined #ceph
[2:10] * abhi (~yaaic@zz2012427835719DFD82.userreverse.dion.ne.jp) has joined #ceph
[2:13] * sileht (~sileht@sileht.net) Quit (Ping timeout: 480 seconds)
[2:17] * swami1 (~swami@163.138.224.180) has joined #ceph
[2:18] * georgem (~Adium@69-196-163-180.dsl.teksavvy.com) has joined #ceph
[2:21] * ibravo (~ibravo@72.83.69.64) has joined #ceph
[2:22] * abhi2 (~yaaic@zz2012427835719DFD82.userreverse.dion.ne.jp) has joined #ceph
[2:22] * abhi (~yaaic@zz2012427835719DFD82.userreverse.dion.ne.jp) Quit (Ping timeout: 480 seconds)
[2:25] * fred`` (fred@earthli.ng) Quit (Quit: +++ATH0)
[2:28] * brutuscat (~brutuscat@105.34.133.37.dynamic.jazztel.es) Quit (Remote host closed the connection)
[2:29] * swami1 (~swami@163.138.224.180) Quit (Quit: Leaving.)
[2:33] * kefu (~kefu@114.86.210.253) has joined #ceph
[2:34] * abhi2 (~yaaic@zz2012427835719DFD82.userreverse.dion.ne.jp) Quit (Ping timeout: 480 seconds)
[2:38] * ibravo (~ibravo@72.83.69.64) Quit (Quit: Leaving)
[2:44] * abhi (~yaaic@2402:c800:ff64:300:ab:d665:5099:397c) has joined #ceph
[2:49] * jcsp (~jspray@2402:c800:ff64:300:df3d:9e5b:2e44:13e7) Quit (Ping timeout: 480 seconds)
[2:52] * kefu (~kefu@114.86.210.253) Quit (Max SendQ exceeded)
[2:53] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[2:53] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[2:53] * kefu (~kefu@114.86.210.253) has joined #ceph
[2:55] * fred`` (fred@earthli.ng) has joined #ceph
[2:56] * abhi (~yaaic@2402:c800:ff64:300:ab:d665:5099:397c) Quit (Ping timeout: 480 seconds)
[2:57] * jcsp (~jspray@2402:c800:ff64:300:df3d:9e5b:2e44:13e7) has joined #ceph
[3:01] * georgem (~Adium@69-196-163-180.dsl.teksavvy.com) Quit (Quit: Leaving.)
[3:01] * abhi (~yaaic@2402:c800:ff64:300:ab:d665:5099:397c) has joined #ceph
[3:02] * georgem (~Adium@206.108.127.16) has joined #ceph
[3:03] * abhi (~yaaic@2402:c800:ff64:300:ab:d665:5099:397c) Quit (Read error: Connection reset by peer)
[3:08] * abhi (~yaaic@2402:c800:ff64:300:ab:d665:5099:397c) has joined #ceph
[3:10] * kefu (~kefu@114.86.210.253) Quit (Max SendQ exceeded)
[3:10] * kefu (~kefu@114.86.210.253) has joined #ceph
[3:11] * moloney (~oftc-webi@137.53.91.160) Quit (Quit: Page closed)
[3:16] * zhaochao (~zhaochao@125.39.8.235) has joined #ceph
[3:21] * kefu (~kefu@114.86.210.253) Quit (Max SendQ exceeded)
[3:22] * kefu (~kefu@114.86.210.253) has joined #ceph
[3:23] * georgem (~Adium@206.108.127.16) Quit (Quit: Leaving.)
[3:24] * georgem (~Adium@206.108.127.16) has joined #ceph
[3:30] * brutuscat (~brutuscat@105.34.133.37.dynamic.jazztel.es) has joined #ceph
[3:30] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[3:30] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[3:30] * kefu (~kefu@114.86.210.253) Quit (Max SendQ exceeded)
[3:31] * kefu (~kefu@114.86.210.253) has joined #ceph
[3:32] * marrusl (~mark@209-150-46-243.c3-0.wsd-ubr2.qens-wsd.ny.cable.rcn.com) Quit (Remote host closed the connection)
[3:33] * abhi (~yaaic@2402:c800:ff64:300:ab:d665:5099:397c) Quit (Ping timeout: 480 seconds)
[3:35] * xarses (~xarses@2402:c800:ff64:300:4661:62d9:c96f:eb1c) has joined #ceph
[3:38] * abhi (~yaaic@2402:c800:ff64:300:ab:d665:5099:397c) has joined #ceph
[3:40] * kefu is now known as kefu|afk
[3:40] * kefu|afk is now known as kefu
[3:44] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[3:44] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[3:44] * Kupo1 (~tyler.wil@23.111.254.159) Quit (Read error: Connection reset by peer)
[3:44] * yanzheng (~zhyan@125.71.108.204) has joined #ceph
[3:50] * badone_ is now known as badone
[3:58] * abhi (~yaaic@2402:c800:ff64:300:ab:d665:5099:397c) Quit (Ping timeout: 480 seconds)
[4:03] * brutuscat (~brutuscat@105.34.133.37.dynamic.jazztel.es) Quit (Ping timeout: 480 seconds)
[4:06] * elder (~elder@60-250-199-132.HINET-IP.hinet.net) has joined #ceph
[4:07] * abhi (~yaaic@2402:c800:ff64:300:ab:d665:5099:397c) has joined #ceph
[4:08] * qman (~rohroh@2600:3c00::f03c:91ff:fe69:92af) Quit (Remote host closed the connection)
[4:11] * yguang11 (~yguang11@nat-dip15.fw.corp.yahoo.com) has joined #ceph
[4:11] * yguang11 (~yguang11@nat-dip15.fw.corp.yahoo.com) Quit ()
[4:12] * nhm (~nhm@c-50-171-139-246.hsd1.mn.comcast.net) Quit (Ping timeout: 480 seconds)
[4:14] * elder (~elder@60-250-199-132.HINET-IP.hinet.net) Quit (Ping timeout: 480 seconds)
[4:16] * elder (~elder@60-250-199-132.HINET-IP.hinet.net) has joined #ceph
[4:27] * babilen (~babilen@babilen.user.oftc.net) Quit (Ping timeout: 480 seconds)
[4:28] * ilken (~ilknest@2602:63:c2a2:af00:c5e6:5f94:e665:ad4) Quit (Ping timeout: 480 seconds)
[4:29] * qman (~rohroh@2600:3c00::f03c:91ff:fe69:92af) has joined #ceph
[4:29] * overclk (~overclk@121.244.87.117) has joined #ceph
[4:36] * ilken (~ilknest@2602:63:c2a2:af00:5cbd:7490:9b41:4b8e) has joined #ceph
[4:40] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[4:40] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[4:45] * georgem (~Adium@206.108.127.16) Quit (Quit: Leaving.)
[4:45] <mech422> anyone around that might be able to help with a wierd, 'fails if I try to create volumes too fast' problem with openstack/cinder on ceph ?
[4:47] * abhi (~yaaic@2402:c800:ff64:300:ab:d665:5099:397c) Quit (Ping timeout: 480 seconds)
[4:51] * elder (~elder@60-250-199-132.HINET-IP.hinet.net) Quit (Ping timeout: 480 seconds)
[4:51] * elder (~elder@60-250-199-132.HINET-IP.hinet.net) has joined #ceph
[4:51] * Psi-Jack (~psi-jack@lhmon.linux-help.org) Quit (Quit: ZNC shutting down. I'll be back!)
[4:55] * brutuscat (~brutuscat@105.34.133.37.dynamic.jazztel.es) has joined #ceph
[4:56] * jcsp (~jspray@2402:c800:ff64:300:df3d:9e5b:2e44:13e7) Quit (Ping timeout: 480 seconds)
[4:57] * bene2 (~bene@2601:18c:8300:f3ae:ea2a:eaff:fe08:3c7a) Quit (Quit: Konversation terminated!)
[5:00] * xarses (~xarses@2402:c800:ff64:300:4661:62d9:c96f:eb1c) Quit (Ping timeout: 480 seconds)
[5:03] * Psi-Jack (~psi-jack@lhmon.linux-help.org) has joined #ceph
[5:03] * Psi-Jack (~psi-jack@lhmon.linux-help.org) Quit ()
[5:07] * babilen (~babilen@babilen.user.oftc.net) has joined #ceph
[5:10] * kefu is now known as kefu|afk
[5:11] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[5:11] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[5:12] * Vacuum__ (~Vacuum@i59F79B0F.versanet.de) has joined #ceph
[5:14] * kefu|afk is now known as kefu
[5:14] * rdas (~rdas@122.168.210.191) has joined #ceph
[5:19] * Vacuum_ (~Vacuum@i59F79526.versanet.de) Quit (Ping timeout: 480 seconds)
[5:24] * abhi (~yaaic@2402:c800:ff64:300:ab:d665:5099:397c) has joined #ceph
[5:25] * kanagaraj (~kanagaraj@121.244.87.117) has joined #ceph
[5:30] * abhi (~yaaic@2402:c800:ff64:300:ab:d665:5099:397c) Quit (Read error: Connection reset by peer)
[5:39] * abhi (~yaaic@zz2012427835719DFD82.userreverse.dion.ne.jp) has joined #ceph
[5:43] * abhi (~yaaic@zz2012427835719DFD82.userreverse.dion.ne.jp) Quit (Read error: Connection reset by peer)
[5:46] * elder (~elder@60-250-199-132.HINET-IP.hinet.net) Quit (Ping timeout: 480 seconds)
[5:47] * kefu is now known as kefu|afk
[5:48] * hoonetorg (~hoonetorg@77.119.226.254.static.drei.at) Quit (Ping timeout: 480 seconds)
[5:49] * elder (~elder@60-250-199-132.HINET-IP.hinet.net) has joined #ceph
[5:50] * LPG (~LPG@c-50-181-212-148.hsd1.wa.comcast.net) Quit (Remote host closed the connection)
[5:54] * xarses (~xarses@2402:c800:ff64:300:4661:62d9:c96f:eb1c) has joined #ceph
[5:55] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[5:55] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[5:56] * hoonetorg (~hoonetorg@77.119.226.254.static.drei.at) has joined #ceph
[5:57] * rakeshgm (~rakesh@121.244.87.117) has joined #ceph
[5:58] * xarses (~xarses@2402:c800:ff64:300:4661:62d9:c96f:eb1c) Quit (Remote host closed the connection)
[5:58] * xarses (~xarses@2402:c800:ff64:300:4661:62d9:c96f:eb1c) has joined #ceph
[5:59] * brutuscat (~brutuscat@105.34.133.37.dynamic.jazztel.es) Quit (Ping timeout: 480 seconds)
[6:04] * elder (~elder@60-250-199-132.HINET-IP.hinet.net) Quit (Ping timeout: 480 seconds)
[6:08] * lxo (~aoliva@lxo.user.oftc.net) has joined #ceph
[6:09] * jcsp (~jspray@2402:c800:ff64:300:df3d:9e5b:2e44:13e7) has joined #ceph
[6:19] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[6:19] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[6:26] * remy1991 (~ravi@122.179.187.127) has joined #ceph
[6:38] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[6:38] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[6:38] * nihilifer (nihilifer@s6.mydevil.net) Quit (Quit: WeeChat 1.3)
[6:38] * nihilifer (nihilifer@s6.mydevil.net) has joined #ceph
[6:39] * squ (~Thunderbi@00020d26.user.oftc.net) has joined #ceph
[6:47] * remy1991 (~ravi@122.179.187.127) Quit (Ping timeout: 480 seconds)
[6:47] * jcsp (~jspray@2402:c800:ff64:300:df3d:9e5b:2e44:13e7) Quit (Ping timeout: 480 seconds)
[6:51] * sileht (~sileht@sileht.net) has joined #ceph
[6:58] * kefu|afk is now known as kefu
[7:00] * lincolnb (~lincoln@c-71-57-68-189.hsd1.il.comcast.net) Quit (Remote host closed the connection)
[7:00] * lincolnb (~lincoln@c-71-57-68-189.hsd1.il.comcast.net) has joined #ceph
[7:15] * rdas (~rdas@122.168.210.191) Quit (Ping timeout: 480 seconds)
[7:27] * rdas (~rdas@182.70.145.80) has joined #ceph
[7:28] * nardial (~ls@dslb-088-072-094-085.088.072.pools.vodafone-ip.de) has joined #ceph
[7:35] * i_m (~ivan.miro@nat-5-carp.hcn-strela.ru) has joined #ceph
[7:38] * xarses (~xarses@2402:c800:ff64:300:4661:62d9:c96f:eb1c) Quit (Remote host closed the connection)
[7:41] * fridim_ (~fridim@56-198-190-109.dsl.ovh.fr) has joined #ceph
[7:41] * DV (~veillard@2001:41d0:1:d478::1) has joined #ceph
[7:46] * xarses (~xarses@2402:c800:ff64:300:4661:62d9:c96f:eb1c) has joined #ceph
[7:46] * xarses (~xarses@2402:c800:ff64:300:4661:62d9:c96f:eb1c) Quit (Remote host closed the connection)
[7:46] * xarses (~xarses@2402:c800:ff64:300:4661:62d9:c96f:eb1c) has joined #ceph
[7:47] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[7:47] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[7:47] * DV__ (~veillard@2001:41d0:a:f29f::1) Quit (Ping timeout: 480 seconds)
[7:47] * kefu is now known as kefu|afk
[7:48] * kefu|afk is now known as kefu
[7:52] * brutuscat (~brutuscat@105.34.133.37.dynamic.jazztel.es) has joined #ceph
[7:56] * Icey (~chris@0001bbad.user.oftc.net) Quit (Ping timeout: 480 seconds)
[7:59] * shylesh__ (~shylesh@121.244.87.124) has joined #ceph
[8:00] * brutuscat (~brutuscat@105.34.133.37.dynamic.jazztel.es) Quit (Ping timeout: 480 seconds)
[8:00] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[8:00] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[8:01] * Icey (~chris@0001bbad.user.oftc.net) has joined #ceph
[8:03] * karnan (~karnan@121.244.87.117) has joined #ceph
[8:08] * bitserker (~toni@81.184.9.72.dyn.user.ono.com) has joined #ceph
[8:12] * remy1991 (~ravi@115.114.59.182) has joined #ceph
[8:19] * Be-El (~quassel@fb08-bcf-pc01.computational.bio.uni-giessen.de) has joined #ceph
[8:20] * kawa2014 (~kawa@151.33.10.211) has joined #ceph
[8:21] * dan_ (~dan@2001:1458:202:225::101:124a) has joined #ceph
[8:21] * dan (~dan@dvanders-pro.cern.ch) Quit (Read error: Connection reset by peer)
[8:21] * remy1991 (~ravi@115.114.59.182) Quit (Ping timeout: 480 seconds)
[8:22] * Kurt (~Adium@2001:628:1:5:c8af:1a35:2958:d490) Quit (Quit: Leaving.)
[8:25] <Be-El> hi
[8:26] <boolman> ellu
[8:30] * remy1991 (~ravi@115.114.59.182) has joined #ceph
[8:31] * enax (~enax@hq.ezit.hu) has joined #ceph
[8:36] * xarses (~xarses@2402:c800:ff64:300:4661:62d9:c96f:eb1c) Quit (Ping timeout: 480 seconds)
[8:37] * T1w (~jens@node3.survey-it.dk) has joined #ceph
[8:37] * xarses (~xarses@zz2012427835719DFD82.userreverse.dion.ne.jp) has joined #ceph
[8:39] * Kurt (~Adium@2001:628:1:5:10de:b96b:8621:1be9) has joined #ceph
[8:42] * pam (~pam@193.106.183.1) has joined #ceph
[8:42] * xarses (~xarses@zz2012427835719DFD82.userreverse.dion.ne.jp) Quit (Remote host closed the connection)
[8:42] * xarses (~xarses@zz2012427835719DFD82.userreverse.dion.ne.jp) has joined #ceph
[8:43] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[8:43] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[8:45] * linjan (~linjan@86.62.112.22) has joined #ceph
[8:50] * _NiC (~kristian@aeryn.ronningen.no) Quit (Remote host closed the connection)
[8:51] * _NiC (~kristian@aeryn.ronningen.no) has joined #ceph
[8:51] * LeaChim (~LeaChim@host86-143-17-156.range86-143.btcentralplus.com) has joined #ceph
[8:52] * b0e (~aledermue@213.95.25.82) has joined #ceph
[8:55] * fdmanana (~fdmanana@2001:8a0:6dfd:6d01:dc49:daec:a0a3:2f0f) has joined #ceph
[8:56] * nils_ (~nils_@doomstreet.collins.kg) Quit (Quit: This computer has gone to sleep)
[9:00] * ngoswami (~ngoswami@121.244.87.116) has joined #ceph
[9:03] * lightspeed (~lightspee@fw-carp-wan.ext.lspeed.org) Quit (Ping timeout: 480 seconds)
[9:10] * analbeard (~shw@support.memset.com) has joined #ceph
[9:11] * Wielebny (~Icedove@cl-927.waw-01.pl.sixxs.net) has joined #ceph
[9:12] * lightspeed (~lightspee@fw-carp-wan.ext.lspeed.org) has joined #ceph
[9:14] * stxShadow (~jens@92.39.19.242.fixip.bitel.net) has joined #ceph
[9:15] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[9:15] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[9:16] * rdas_ (~rdas@122.168.168.184) has joined #ceph
[9:19] * ade (~abradshaw@tmo-100-229.customers.d1-online.com) has joined #ceph
[9:19] * rdas (~rdas@182.70.145.80) Quit (Ping timeout: 480 seconds)
[9:21] * vbellur (~vijay@122.172.57.91) Quit (Ping timeout: 480 seconds)
[9:23] * lightspeed (~lightspee@fw-carp-wan.ext.lspeed.org) Quit (Ping timeout: 480 seconds)
[9:25] * fsimonce (~simon@host30-173-dynamic.23-79-r.retail.telecomitalia.it) has joined #ceph
[9:26] * pabluk_ is now known as pabluk
[9:26] * rotbeard (~redbeard@185.32.80.238) has joined #ceph
[9:28] * legion (~Nijikokun@216.17.99.183) has joined #ceph
[9:32] * lightspeed (~lightspee@fw-carp-wan.ext.lspeed.org) has joined #ceph
[9:32] * garphy`aw is now known as garphy
[9:34] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[9:34] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[9:36] * daviddcc (~dcasier@84.197.151.77.rev.sfr.net) has joined #ceph
[9:37] * xarses_ (~xarses@2402:c800:ff64:300:4661:62d9:c96f:eb1c) has joined #ceph
[9:40] * xarses (~xarses@zz2012427835719DFD82.userreverse.dion.ne.jp) Quit (Ping timeout: 480 seconds)
[9:44] * jasuarez (~jasuarez@243.Red-81-39-64.dynamicIP.rima-tde.net) has joined #ceph
[9:45] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[9:45] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Remote host closed the connection)
[9:45] * kefu is now known as kefu|afk
[9:46] * kefu|afk is now known as kefu
[9:49] * Ganebal (~Ganebal@bzq-82-81-163-130.red.bezeqint.net) has joined #ceph
[9:50] * dgurtner (~dgurtner@178.197.231.141) has joined #ceph
[9:51] <Ganebal> Hi, I'm the new one here :). We have a strange problem with osd balancing. We have 24 osds with the same size. Number of PGs is 4096 per pool and total number is 55984. After using of ceph 8 months we see that some of osds isn't balanced. For example osd.14 is using more disk space than other osds (~10%). We ran ceph osd reweight-by-utilization in order to fix the problem, but is there another way to let the ceph to fix it by itself ? M
[9:51] <Ganebal> aybe to increase number of PGs per pool ? Thanks in advance.
[9:51] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[9:51] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[9:52] <Be-El> Ganebal: ceph uses hashing to distribute objects over pgs. and hashing introduces a certain amount of imbalance
[9:52] <Be-El> Ganebal: more pgs might help, but will increase the amount of resources consumed by an osd process (memory, cpu)
[9:53] <kiranos> for 24 osd's 55984 pg's seems alot already
[9:53] * dan (~dan@2001:1458:202:225::102:124a) has joined #ceph
[9:53] <Be-El> and 10% is rather low, we have between 20% and 30% on our cluster
[9:55] * rendar (~I@host235-46-dynamic.31-79-r.retail.telecomitalia.it) has joined #ceph
[9:57] <Ganebal> thank you for your quick answers :) @Be-El but we had other clusters and diff between the osd was ~1-2%.
[9:57] * dan_ (~dan@2001:1458:202:225::101:124a) Quit (Ping timeout: 480 seconds)
[9:58] <Be-El> Ganebal: maybe the object names had a different structure on that clusters
[9:58] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[9:58] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[9:58] * legion (~Nijikokun@4Z9AAAHG5.tor-irc.dnsbl.oftc.net) Quit ()
[9:59] * Nacer (~Nacer@LCaen-656-1-72-185.w80-13.abo.wanadoo.fr) has joined #ceph
[10:02] * Amto_res (~amto_res@ks312256.kimsufi.com) Quit (Quit: leaving)
[10:03] <Ganebal> :Be-El what do you mean " the object names had a different structure" ? :)
[10:04] <T1w> how big should a MONs data directory be?
[10:04] <Be-El> Ganebal: hashing is based on the object names. if the name distribution leads to an imbalanced hashing, the data will be imbalanced across the osds
[10:04] <T1w> 5G? 10G?
[10:05] * Amto_res (~amto_res@ks312256.kimsufi.com) has joined #ceph
[10:05] <T1w> I'm creating a new LV for it, and I'd like to size it resonable
[10:05] <Be-El> T1w: there were some recent discussion (here, mailing list?) with some people having problems to store the mon databases due to limited space
[10:05] <Be-El> T1w: and they reported sizes > 20 or 30 GB
[10:06] * rakeshgm (~rakesh@121.244.87.117) Quit (Quit: Leaving)
[10:06] <Be-El> especially during backfilling/recovery the database may be significantly larger than 5 GB
[10:06] * rakeshgm (~rakesh@121.244.87.117) has joined #ceph
[10:06] <T1w> Be-El: mkay - but I assume it's also tied up to how many OSDs there are in play - and probably a lot of other stuff
[10:06] <T1w> mkay..
[10:06] * alrick (~alrick@91.218.144.129) has joined #ceph
[10:07] <Be-El> T1w: number of OSDs, number of PGs etc....there are a lot of factors involved
[10:07] <T1w> I've got ~97GB free that should be distributed between 2x OSD journal, MON data dir and then a bit of reserve
[10:07] <T1w> yeah..
[10:07] <Be-El> T1w: tl;dr: make an educated guess and use 2/3 times the size ;-)
[10:07] <T1w> haha
[10:07] <T1w> it would be a _very_ uneducated guess.. :0
[10:07] <T1w> :) even
[10:08] <mfa298_> T1w: normally it should be small, but if the cluster is degraded for a while it can grow quickly (we're at 130GB currently - buts thats 2200 OSDs and it's been degraded for a fortnight)
[10:08] * andreww (~xarses@zz2012427835719DFD82.userreverse.dion.ne.jp) has joined #ceph
[10:08] * xarses_ (~xarses@2402:c800:ff64:300:4661:62d9:c96f:eb1c) Quit (Ping timeout: 480 seconds)
[10:08] <T1w> eeeeeks
[10:08] <T1w> ok
[10:09] <Be-El> mfa298_: whoops...that sounds like a lot of fun
[10:09] <T1w> well.. 3 physical hosts - each with 2 OSDs
[10:09] <T1w> and growing to perhaps 10s oh hosts and perhaps 100s of OSDs over the comming years
[10:09] <T1w> oh = of
[10:10] <T1w> 20GB "should be enough" for now
[10:10] <T1w> I'd conservatively recon
[10:11] <T1w> and OSD journal LV size - 10G?
[10:11] <T1w> for a 4TB disk
[10:11] <Ganebal> Be-El: as you said before the increasing number of PGs can help ?
[10:12] <Be-El> T1w: "640kb should be enough for everyone"
[10:12] <T1w> :D
[10:12] <Be-El> Ganebal: you already have a large number of pgs
[10:12] <Be-El> Ganebal: and you cannot reduce the number of pgps afterwards
[10:16] * andreww (~xarses@zz2012427835719DFD82.userreverse.dion.ne.jp) Quit (Ping timeout: 480 seconds)
[10:17] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[10:17] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[10:18] * shawniverson (~shawniver@192.69.183.61) has joined #ceph
[10:24] <T1w> is it reccommended to make the OSD journal a block device that the OSD manages itself or should I make it a filesystem?
[10:25] * infernix (nix@spirit.infernix.net) Quit (Ping timeout: 480 seconds)
[10:27] <mfa298_> T1w: it would normally be a block device.
[10:27] <kiranos> T1w: it depends on your use case, the best way is to have journal on the local device as its own partition, but if you want speed its better to have it on an external ssd device
[10:28] <T1w> kiranos: I think you misunderstood..
[10:28] * hgichon (~hgichon@220.90.135.162) Quit (Ping timeout: 480 seconds)
[10:29] <mfa298_> T1w: kiranos' point is valid though, if the journal is on the same SSD as the mon store and OS you might hit contention issues that slow the whole thing down.
[10:29] <T1w> I'm placing the journals on ssd's (shared with the os), but I'd like to know if I should make /var/lib/ceph/osd/cluster-0/journal (etc etc..) a file on an LV mounted to /var/lib/ceph/osd/cluster-0 or if I should make the LV mounted directly to /var/lib/ceph/osd/cluster-0/journal
[10:29] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[10:29] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[10:30] <T1w> .. and I'm using Intel S3710 ssd's, so I should have enough IOPS for 2 OSDs
[10:31] <T1w> right now I'm prepping the LVs on the individual nodes before I'm letting ceph-deploy have a go at them
[10:35] * infernix (nix@spirit.infernix.net) has joined #ceph
[10:40] * fen (~fen@HSI-KBW-217-008-056-240.hsi.kabelbw.de) has joined #ceph
[10:40] <fen> hey :-)
[10:40] * dan_ (~dan@dvanders-pro.cern.ch) has joined #ceph
[10:41] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[10:41] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[10:41] * haodong (~haodong@103.10.86.234) has joined #ceph
[10:46] * kawa2014 (~kawa@151.33.10.211) Quit (Ping timeout: 480 seconds)
[10:47] * dan (~dan@2001:1458:202:225::102:124a) Quit (Ping timeout: 480 seconds)
[10:47] * brutuscat (~brutuscat@41.Red-83-47-113.dynamicIP.rima-tde.net) has joined #ceph
[10:47] * brutuscat (~brutuscat@41.Red-83-47-113.dynamicIP.rima-tde.net) Quit (Remote host closed the connection)
[10:47] * brutuscat (~brutuscat@41.Red-83-47-113.dynamicIP.rima-tde.net) has joined #ceph
[10:50] <T1w> hm..
[10:51] <T1w> I'm having a brainfart right now and I can't see a way around it..
[10:51] <T1w> I've got a LV /dev/VGsys0/mon
[10:51] <T1w> argh..
[10:52] <T1w> it's /dev/VGsys0/jour1
[10:52] <T1w> I need it to be mapped as a block device to /var/lib/ceph/osd/cluster-0/journal
[10:52] <T1w> there is something I'm missing the be able to do that
[10:53] * daviddcc (~dcasier@84.197.151.77.rev.sfr.net) Quit (Ping timeout: 480 seconds)
[10:55] <mfa298_> T1w: if you're creating a new OSD you should be able to feed the OSD drive and journal device into ceph-deploy or ceph-disk
[10:55] <T1w> ahhhhhh
[10:55] <T1w> thanks
[10:56] <mfa298_> If it's an existing OSD (that will already have a journal) the process is a bit more involved (you need to flush the original journal first)
[10:56] * rdas_ (~rdas@122.168.168.184) Quit (Quit: Leaving)
[10:56] <T1w> mmm - it's not - starting from an all clean slate
[10:57] * TMM (~hp@sams-office-nat.tomtomgroup.com) has joined #ceph
[11:06] <Ganebal> Ok. if I already have a enough pg in the cluster, is there any way to prevent disbalance in the further ? Thx
[11:06] * rdas (~rdas@122.168.168.184) has joined #ceph
[11:10] * todin (tuxadero@kudu.in-berlin.de) Quit (Ping timeout: 480 seconds)
[11:13] * kefu is now known as kefu|afk
[11:24] * thomnico (~thomnico@2a01:e35:8b41:120:4884:9cb6:f7cc:487b) has joined #ceph
[11:26] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[11:26] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[11:27] * rdas (~rdas@122.168.168.184) Quit (Quit: Leaving)
[11:35] * i_m (~ivan.miro@nat-5-carp.hcn-strela.ru) Quit (Ping timeout: 480 seconds)
[11:44] * todin (tuxadero@kudu.in-berlin.de) has joined #ceph
[11:44] * brutuscat (~brutuscat@41.Red-83-47-113.dynamicIP.rima-tde.net) Quit (Read error: No route to host)
[11:45] * brutuscat (~brutuscat@41.Red-83-47-113.dynamicIP.rima-tde.net) has joined #ceph
[11:51] * overclk (~overclk@121.244.87.117) Quit (Ping timeout: 480 seconds)
[11:54] <fen> can i run the mon and the osd daemons on the same physical node if i take care that there's enough memory and osd and mons use different disks?
[11:55] * alrick (~alrick@91.218.144.129) Quit (Remote host closed the connection)
[11:55] <stxShadow> sure ... why not ?
[11:55] <kiranos> fen: "Ceph provides a default path where Ceph Monitors store data. For optimal performance in a production Ceph Storage Cluster, we recommend running Ceph Monitors on separate hosts and drives from Ceph OSD Daemons. Ceph Monitors do lots of fsync(), which can interfere with Ceph OSD Daemon workloads."
[11:55] <fen> i'd like to start with three nodes total (mon + osds)
[11:55] <fen> yeah but the cluster is not that i/o intense
[11:55] <kiranos> then its ok
[11:55] <fen> and if i use different disks the fsync should not matter or should it
[11:56] <stxShadow> you can move the services later if the cluster grows
[11:56] <fen> yes - that too
[11:56] <fen> ok - then i can start with three nodes which is perfect due to budget and space constraints
[12:01] * georgem (~Adium@69-196-163-180.dsl.teksavvy.com) has joined #ceph
[12:02] * karnan (~karnan@121.244.87.117) Quit (Remote host closed the connection)
[12:04] * Nacer (~Nacer@LCaen-656-1-72-185.w80-13.abo.wanadoo.fr) Quit (Remote host closed the connection)
[12:05] * Nacer (~Nacer@176.31.89.99) has joined #ceph
[12:07] * alrick (~alrick@91.218.144.129) has joined #ceph
[12:07] * Nacer (~Nacer@176.31.89.99) Quit (Read error: Connection reset by peer)
[12:07] * Nacer (~Nacer@176.31.89.99) has joined #ceph
[12:10] * georgem (~Adium@69-196-163-180.dsl.teksavvy.com) Quit (Quit: Leaving.)
[12:14] * vbellur (~vijay@122.172.57.91) has joined #ceph
[12:15] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[12:15] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[12:16] * Nacer (~Nacer@176.31.89.99) Quit (Ping timeout: 480 seconds)
[12:23] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[12:23] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[12:27] * zhaochao (~zhaochao@125.39.8.235) Quit (Quit: ChatZilla 0.9.92 [Iceweasel 38.3.0/20150922225347])
[12:29] * i_m (~ivan.miro@deibp9eh1--blueice4n2.emea.ibm.com) has joined #ceph
[12:32] * danieagle (~Daniel@177.9.73.197) has joined #ceph
[12:37] * brutuscat (~brutuscat@41.Red-83-47-113.dynamicIP.rima-tde.net) Quit (Remote host closed the connection)
[12:38] * brutuscat (~brutuscat@41.Red-83-47-113.dynamicIP.rima-tde.net) has joined #ceph
[12:39] * shinobu (~oftc-webi@pdf874b16.tokynt01.ap.so-net.ne.jp) has joined #ceph
[12:44] * kawa2014 (~kawa@89.184.114.246) has joined #ceph
[12:44] * ira (~ira@nat-pool-bos-t.redhat.com) has joined #ceph
[12:50] * georgem (~Adium@207.164.79.45) has joined #ceph
[12:50] * georgem (~Adium@207.164.79.45) Quit ()
[12:50] * georgem (~Adium@206.108.127.16) has joined #ceph
[12:52] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[12:52] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[12:59] * daviddcc (~dcasier@LAubervilliers-656-1-16-160.w217-128.abo.wanadoo.fr) has joined #ceph
[12:59] * overclk (~overclk@121.244.87.117) has joined #ceph
[13:00] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[13:00] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[13:00] * shawniverson (~shawniver@192.69.183.61) Quit (Remote host closed the connection)
[13:03] * karnan (~karnan@106.51.240.17) has joined #ceph
[13:08] * brutuscat (~brutuscat@41.Red-83-47-113.dynamicIP.rima-tde.net) Quit (Remote host closed the connection)
[13:10] * brutuscat (~brutuscat@41.Red-83-47-113.dynamicIP.rima-tde.net) has joined #ceph
[13:18] * brutuscat (~brutuscat@41.Red-83-47-113.dynamicIP.rima-tde.net) Quit (Ping timeout: 480 seconds)
[13:30] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[13:30] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[13:34] <T1w> hm, do I need to add repos to the individual nodes in order for ceph-deploy to work?
[13:34] <T1w> I'm getting
[13:34] <T1w> [ceph1][INFO ] Running command: sudo yum -y install ceph-osd ceph-mds ceph-mon ceph-radosgw
[13:34] <T1w> [ceph1][DEBUG ] No package ceph-osd available.
[13:34] <T1w> [ceph1][DEBUG ] No package ceph-mds available.
[13:34] <T1w> [ceph1][DEBUG ] No package ceph-mon available.
[13:34] <T1w> [ceph1][DEBUG ] Package 1:ceph-radosgw-0.80.7-0.6.el7.x86_64 already installed and latest version
[13:35] <T1w> on the admin-node where I'm running ceph-deploy from I've only got the ceph-noarc repo defined (from where I installed ceph-deploy)
[13:35] <T1w> I was under the impression, that ceph-deploy knew about the repos
[13:37] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[13:37] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[13:38] <fen> are hot swappable drives important for osds or is it ok to shutdown a node for drive repair
[13:38] * shaunm (~shaunm@208.102.161.229) has joined #ceph
[13:40] * brutuscat (~brutuscat@41.Red-83-47-113.dynamicIP.rima-tde.net) has joined #ceph
[13:42] * kanagaraj (~kanagaraj@121.244.87.117) Quit (Quit: Leaving)
[13:43] * georgem (~Adium@206.108.127.16) Quit (Quit: Leaving.)
[13:45] * brutuscat (~brutuscat@41.Red-83-47-113.dynamicIP.rima-tde.net) Quit (Remote host closed the connection)
[13:45] * brutuscat (~brutuscat@41.Red-83-47-113.dynamicIP.rima-tde.net) has joined #ceph
[13:47] <fen> ... and what ratio is the ssd journal to spinning disk space for an osd
[13:49] <via> fen: if you make your failure domain a host, you can take a host down for maintainance easily
[13:49] <via> that said, hotswap is always gonna be easier
[13:49] <fen> ok..
[13:49] <via> i use 10G journal or 3T disks
[13:49] <T1w> wtf..
[13:49] <T1w> there are no ceph-osd packages available under the hammer repo
[13:50] <via> T1w: the package is just 'ceph'
[13:50] <via> it has all the daemons
[13:50] <fen> via: so the journal is usually not very big
[13:50] <via> no. there's a doc page on journal sizing, trying to find it
[13:50] * haodong (~haodong@103.10.86.234) Quit (Quit: Leaving)
[13:51] <T1w> via: then I don't understand why ceph-deploy issues this command
[13:51] <T1w> [ceph1][INFO ] Running command: sudo yum -y install ceph-osd ceph-mds ceph-mon ceph-radosgw
[13:52] <via> T1w: i don't know, i only used ceph-deploy once a while ago and it instlaled the right thing
[13:52] <T1w> I'm expecting ceph-deploy to install hammer releases, but apparently I'm getting ceph-common 0.80.7
[13:52] <via> you can specify the release on the commandline iirc
[13:53] <via> fen: "osd journal size = {2 * (expected throughput * filestore max sync interval)}"
[13:53] <via> where the latter is 5s default
[13:53] <fen> via: thanks
[13:54] <T1w> via: I'd assume a simple "ceph-deploy install <node>" would install the latest stable version
[13:54] <via> yeah, i would too
[13:54] <via> maybe you have an ancient ceph-deploy?
[13:54] <T1w> and the quick start guide (http://docs.ceph.com/docs/master/start/quick-ceph-deploy/) never mentions anything about installing a specific release
[13:55] <T1w> running 1.5.28
[13:55] <T1w> from http://download.ceph.com/rpm-hammer/rhel7/noarch/
[13:56] <via> i don't know then, sorry, i'm not very familiar with ceph-deploy. i created my cluster before it existed
[13:56] <T1w> quite allright - you're just the only one answering.. :)
[13:57] <T1w> the preflight page (http://docs.ceph.com/docs/master/start/quick-start-preflight/) mentions that the repository should just be added to the admin-node
[13:58] <via> i know it used to work, i feel like with redhat now running the place people would have noticed if that broke
[13:58] <via> did ceph-deploy install the repos on the target machines?
[13:58] <T1w> indeed
[13:59] <T1w> no, the individual nodes does not have the repo
[13:59] <T1w> only a few RHEL ones and EPEL
[13:59] <T1w> (as instructed in the setup notes)
[13:59] <via> well, ceph-deploy installs the repos iirc
[13:59] <via> so i dunno, thats weird
[14:02] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[14:02] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[14:05] * vegas3 (~FNugget@93.115.84.202) has joined #ceph
[14:07] * shinobu (~oftc-webi@pdf874b16.tokynt01.ap.so-net.ne.jp) Quit (Quit: Page closed)
[14:09] * bene2 (~bene@nat-pool-bos-t.redhat.com) has joined #ceph
[14:10] * georgem (~Adium@206.108.127.16) has joined #ceph
[14:12] <T1w> hmpf.. adding the repo to the node helped - I got 0.94.5 installed now
[14:12] <T1w> but I'm still getting
[14:12] <T1w> [ceph1][DEBUG ] No package ceph-osd available.
[14:12] <T1w> [ceph1][DEBUG ] No package ceph-mds available.
[14:12] <T1w> [ceph1][DEBUG ] No package ceph-mon available.
[14:14] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[14:14] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[14:14] * rakeshgm (~rakesh@121.244.87.117) Quit (Ping timeout: 480 seconds)
[14:16] <kiranos> https://git.ceph.com/?p=ceph.git;a=blob_plain;f=keys/release.asc
[14:16] <kiranos> is down?=
[14:16] <kiranos> extremely slow for me again
[14:16] <kiranos> is it under attack?
[14:16] <T1w> not down, just sooooo slow
[14:16] <T1w> was also slow yesterday
[14:17] <kiranos> must be something with the webserver
[14:17] <kiranos> can we ping somone?
[14:17] <kiranos> T1w:
[14:17] <kiranos> btw
[14:18] <kiranos> ceph-deploy install ceph01-osd02 --release hammer
[14:18] <kiranos> worked great for me, just had to wait for the release.asc
[14:18] <T1w> kiranos: hm, ok - I'll try that
[14:23] * erhudy (uid89730@id-89730.ealing.irccloud.com) has joined #ceph
[14:24] <kiranos> T1w: did it work?
[14:24] <T1w> waiting for key import.. :)
[14:24] <kiranos> lol :)
[14:24] <T1w> ealier today I had to retry once due to timeout
[14:25] * rakeshgm (~rakesh@121.244.87.124) has joined #ceph
[14:26] <T1w> dang.. it always imports the key
[14:28] <alfredodeza> T1w: kiranos ceph-deploy can use any key you might want to point it at
[14:28] <alfredodeza> that key is also here: http://download.ceph.com/keys/release.asc
[14:29] <kiranos> yes but the default should be hosted on a stable server
[14:29] <alfredodeza> ceph-deploy install --help, should point to the right flag
[14:29] <T1w> alfredodeza: .. and not an overloaded gitweb server
[14:29] <alfredodeza> kiranos: stable servers can become unstable :)
[14:30] <alfredodeza> download.ceph.com is (currently) a stable server
[14:30] <kiranos> :) yes but it was the same yesterday, a simple icinga/nagios check would alert
[14:30] <T1w> curl: (22) The requested URL returned error: 504 Gateway Time-out
[14:30] <T1w> error: https://git.ceph.com/?p=ceph.git: import read failed(2).
[14:30] <alfredodeza> and who would receive that alart kiranos :D
[14:30] <kiranos> ok great
[14:30] <alfredodeza> like I mentioned, you can point directly to another location for the key
[14:30] <alfredodeza> http://download.ceph.com/keys/release.asc works for me fine and it is not overloaded
[14:31] <T1w> hm.. I can't the the option for that
[14:31] <alfredodeza> you can also default to that one using cephdeploy.conf
[14:31] <T1w> there is --repo-url and --gpg-url
[14:31] <kiranos> alfredodeza: yes I understand, I've used it before, but its often the servers are overloaded
[14:31] <kiranos> perhaps offer a mirror list
[14:31] <T1w> doh.. gpg-url it is
[14:31] <alfredodeza> *and that is why this is an option*
[14:32] <kiranos> we could host an official mirror
[14:32] <kiranos> I bet alot of ceph users could
[14:32] <kiranos> and perhaps do a git pull to change the default release.asc from git.ceph
[14:32] * nhm (~nhm@c-50-171-139-246.hsd1.mn.comcast.net) has joined #ceph
[14:32] * ChanServ sets mode +o nhm
[14:32] <T1w> aha
[14:32] <T1w> does not work..
[14:33] <T1w> ceph-deploy install ceph1 --release hammer --gpg-url http://download.ceph.com/keys/release.asc
[14:33] <T1w> ...
[14:33] <T1w> [ceph1][INFO ] Running command: sudo rpm --import https://git.ceph.com/?p=ceph.git;a=blob_plain;f=keys/release.asc
[14:34] <alfredodeza> T1w: would you mind pasting the full output of ceph-deploy somewhere so I can take a look?
[14:34] * thomnico (~thomnico@2a01:e35:8b41:120:4884:9cb6:f7cc:487b) Quit (Read error: Connection reset by peer)
[14:34] <T1w> alfredodeza: sure - hang on
[14:35] <kiranos> alfredodeza: http://mirror-status.centos.org/
[14:35] <kiranos> is a nice example
[14:35] * vegas3 (~FNugget@93.115.84.202) Quit ()
[14:35] <T1w> alfredodeza: http://pastebin.com/qny1wQtY
[14:36] * ade (~abradshaw@tmo-100-229.customers.d1-online.com) Quit (Ping timeout: 480 seconds)
[14:36] * thehoffau (~thehoffau@ns3000816.ip-5-196-74.eu) has joined #ceph
[14:37] * ade (~abradshaw@tmo-100-152.customers.d1-online.com) has joined #ceph
[14:38] <alfredodeza> ah you are using redhat, there is a caveat if you are using redhat and want to install community bits
[14:38] <alfredodeza> one sec
[14:38] <T1w> huh ok..
[14:39] <T1w> at a later time we'll probably add centos to the mix, but for now we're running everything off rhel - I hope that will work
[14:39] * marrusl (~mark@209-150-46-243.c3-0.wsd-ubr2.qens-wsd.ny.cable.rcn.com) has joined #ceph
[14:39] <alfredodeza> it works, but I think there is something you are required to do
[14:39] <alfredodeza> one sec
[14:39] <T1w> mkay
[14:40] <alfredodeza> aha, you are supposed to specify --release when using RHEL (which you are)
[14:40] <alfredodeza> trying to dig further why you are getting the other url
[14:40] <magicrobotmonkey> pro tip: If you have a bunch of pgs stuck in active+remapped or just active, check that you haven't run reweight by utilization too many times and all your osd weights are too low to place pgs
[14:40] <T1w> ok
[14:40] <T1w> magicrobotmonkey: >.<
[14:41] <magicrobotmonkey> I ran into that and reweighted everything back to 1 and now I'm at 12736 active+clean
[14:41] * TheSov (~TheSov@cip-248.trustwave.com) has joined #ceph
[14:45] <alfredodeza> T1w: so this is a bug. Kind of. Let me explain
[14:45] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[14:45] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[14:45] <T1w> alfredodeza: heh, okay
[14:45] <alfredodeza> the gpg-url and repo-url where meant to be used together, in the sense that a user would want to use a mirror
[14:45] <alfredodeza> or a proxy
[14:45] <alfredodeza> so those options where kind of "use these both options if --repo-url is passed in"
[14:46] <alfredodeza> in your case this shouldn't be, it is a corner case though, but a bug nonetheless
[14:46] <alfredodeza> T1w: to go around it you can pass --repo-url
[14:46] <alfredodeza> and point to the hammer release
[14:46] <T1w> ah, so if I had also specified repo-url it should have worked
[14:46] <alfredodeza> yes I think so
[14:46] <alfredodeza> can you try?
[14:46] <T1w> okay, nice to have found a bug.. :)
[14:46] <kiranos> magicrobotmonkey: yes always try to use ceph crush reweight, not just ceph reweight
[14:47] <kiranos> I've ran into the same issue
[14:47] <T1w> while you were looking into things the normal key from git was actually fetched
[14:47] <T1w> so I seem to have gotten my ceph1 node installed
[14:48] <TheSov> we need like a freenas distro for ceph osds
[14:48] * thomnico (~thomnico@2a01:e35:8b41:120:4884:9cb6:f7cc:487b) has joined #ceph
[14:48] <TheSov> something similar to that
[14:48] <kiranos> TheSov: supermicro works great :)
[14:48] <TheSov> yes supermicro's are awesome!
[14:48] <TheSov> but i mean we need a distribution of linux that essentially runs ceph osd out of the box
[14:49] <T1w> alfredodeza: I'll retry with --repo-url and --gpg-key then I install the next node - hopefully I'll be able to get to that today (a few mails are pressing them into my workload atm..)
[14:49] <T1w> themselves even
[14:49] <TheSov> like you stick in the cd, install it to a bootable flash, set the IP, point it at a ceph monitor, supply username and password, pick the hard drives and bam, its done.
[14:49] <TheSov> is that too much to ask
[14:50] * jrankin (~jrankin@d53-64-170-236.nap.wideopenwest.com) has joined #ceph
[14:50] <Aeso_> well no matter what you'd have to copy over the keyrings first
[14:50] * vbellur (~vijay@122.172.57.91) Quit (Ping timeout: 480 seconds)
[14:50] <TheSov> thats the point of pointing it to a monitor
[14:50] <TheSov> it will ssh and copy the keyring :)
[14:51] <TheSov> Aeso_, eh?
[14:51] <kiranos> TheSov: look at ansible
[14:51] <alfredodeza> SmartOS does something like that TheSov, you just slap a usb key with the minimal os in it
[14:51] <TheSov> im not versed at all in ansible :(
[14:51] <alfredodeza> you do need to run a couple of things the first time around but it is very very simple
[14:51] <TheSov> or puppet
[14:51] <TheSov> or chef
[14:52] <alfredodeza> issue 13629
[14:52] <kraken> alfredodeza might be talking about http://tracker.ceph.com/issues/13629 [using --gpg-url alone doesn't do anything]
[14:52] * swami1 (~swami@zz2012427835719DFD82.userreverse.dion.ne.jp) has joined #ceph
[14:52] <alfredodeza> T1w: ^ ^
[14:52] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[14:52] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[14:52] <T1w> alfredodeza: :)
[14:52] <alfredodeza> thanks for helping get some output of your issue
[14:53] <T1w> glad to have helped.. :)
[14:53] <kiranos> TheSov: if you dont know any of these tool, ansible is by far the easiest to learn
[14:54] <m0zes> ansible is pure ssh. whee :D
[14:54] <TheSov> do you have a resource for getting started?
[14:54] * Ganebal (~Ganebal@bzq-82-81-163-130.red.bezeqint.net) Quit (Quit: Leaving)
[14:54] <TheSov> m0zes, ooooh i like that
[14:54] <TheSov> I'm not too bad at scripting
[14:54] <m0zes> http://docs.ansible.com/ansible/intro_getting_started.html
[14:55] <TheSov> literally getting started LOL
[14:55] <TheSov> thanks m0z
[14:55] * lcurtis (~lcurtis@47.19.105.250) has joined #ceph
[14:55] <m0zes> their docs are pretty good. I'm still moving all my stuff from cfengine2 to ansible, but it is going very well.
[14:56] * squ (~Thunderbi@00020d26.user.oftc.net) Quit (Quit: squ)
[14:56] <Tetard> ansible ftw
[14:57] <kiranos> yes ansible is very nice, not like puppet, its hell to get into a large play
[14:57] <m0zes> eventually I'll have a playbook that can build an hpc cluster, configure dns, icinga, ganglia, dhcp. probably need to throw ceph in there. I don't know. I might leave ceph out of the "hpc" playbook.
[14:57] <kiranos> how all dependencies etc fit togehter
[14:57] <thehoffau> there are already good ansible playbooks out there, we just tweaked them a bit and have been perfect. https://github.com/ceph/ceph-ansible
[14:57] <kiranos> m0zes: I would split them up to multiple playbooks
[14:57] <kiranos> import roles which fit on all instead
[14:58] * linjan (~linjan@86.62.112.22) Quit (Ping timeout: 480 seconds)
[14:58] <m0zes> they are, more or less. define the role you want things to have, and you can pick and choose roles to deploy at that point in time.
[14:58] <kiranos> ok nice
[15:01] * danieagle (~Daniel@177.9.73.197) Quit (Quit: Obrigado por Tudo! :-) inte+ :-))
[15:01] * karnan (~karnan@106.51.240.17) Quit (Quit: Leaving)
[15:09] <magicrobotmonkey> I have a pretty nice set of ansible roles for dealing with dmcrypt and getting ceph all installed I'm working on releasing
[15:09] <magicrobotmonkey> 800 osd cluster with 1 command!
[15:10] * linjan (~linjan@86.62.112.22) has joined #ceph
[15:10] <rkeene> My system is a little simpler, as it requires no commands at all
[15:11] <rkeene> (Just PXE boot a box and it adds all its disks as OSDs)
[15:16] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[15:16] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[15:19] <rotbeard> magicrobotmonkey, sounds pretty good ;)
[15:19] * bpkroth (~bpkroth@rolo.cae.wisc.edu) has joined #ceph
[15:19] * rotbeard should switch from saltstack to ansible then
[15:21] <magicrobotmonkey> yea rkeene I'm working towards that
[15:22] <magicrobotmonkey> plug and play ceph racks ftw
[15:22] <magicrobotmonkey> im thinking of adding an ansible-pull to my preseed
[15:23] <boolman> I didnt get the playboot to work on vagrant with v9.1 rc
[15:25] * mattbenjamin (~mbenjamin@aa2.linuxbox.com) has joined #ceph
[15:26] * overclk (~overclk@121.244.87.117) Quit (Quit: Leaving)
[15:27] <rkeene> magicrobotmonkey, I have that right now :-)
[15:29] <rkeene> magicrobotmonkey, It's real fancy too, comes with an installer and everything
[15:29] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[15:29] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[15:30] * LobsterRoll (~LobsterRo@140.247.242.44) has joined #ceph
[15:30] <LobsterRoll> ceph
[15:31] <LobsterRoll> what happened to http://download.ceph.com/calamari ?
[15:35] * Wielebny (~Icedove@cl-927.waw-01.pl.sixxs.net) Quit (Quit: Wielebny)
[15:36] <TheSov> redhat happened
[15:37] <TheSov> half the ceph docs dont work and now some of the downloads
[15:38] <TheSov> im getting a sudden urge to mirror the debian repo's
[15:39] <LobsterRoll> Gotcha, they are having some downtime i guess
[15:40] * kawa2014 (~kawa@89.184.114.246) Quit (Ping timeout: 480 seconds)
[15:40] * linjan (~linjan@86.62.112.22) Quit (Ping timeout: 480 seconds)
[15:40] * kawa2014 (~kawa@89.184.114.246) has joined #ceph
[15:40] * rakeshgm (~rakesh@121.244.87.124) Quit (Ping timeout: 480 seconds)
[15:40] <T1w> things got a lot worse since the security incident
[15:42] <Heebie> What would be considered "good" latency levels and "bad" latency levels in rados bench?
[15:43] <Heebie> "the security incident" ?
[15:44] <bpkroth> probably not related, but i recently noticed that a number of the publications links are dead/unresponsive as well
[15:46] <rkeene> Security incident ?
[15:48] * dyasny (~dyasny@198.251.59.55) Quit (Ping timeout: 480 seconds)
[15:48] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[15:48] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[15:48] <T1w> alfredodeza: I'll be back with the test of both parameters tomorrow..
[15:48] <T1w> I'm a going home!
[15:49] * linjan (~linjan@86.62.112.22) has joined #ceph
[15:50] * rakeshgm (~rakesh@121.244.87.117) has joined #ceph
[15:53] * sudocat (~dibarra@2602:306:8bc7:4c50::1f) Quit (Ping timeout: 480 seconds)
[15:54] <TheSov> T1w, what security incident?
[15:57] * debian112 (~bcolbert@24.126.201.64) has joined #ceph
[15:57] <LobsterRoll> https://ceph.com/releases/important-security-notice-regarding-signing-key-and-binary-downloads-of-ceph/
[15:58] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[15:58] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[15:58] * georgem1 (~Adium@206.108.127.16) has joined #ceph
[15:59] * shylesh__ (~shylesh@121.244.87.124) Quit (Remote host closed the connection)
[15:59] * TheSov2 (~TheSov@204.13.200.248) has joined #ceph
[15:59] * T1w (~jens@node3.survey-it.dk) Quit (Ping timeout: 480 seconds)
[15:59] * georgem (~Adium@206.108.127.16) Quit (Read error: Connection reset by peer)
[16:04] * TheSov (~TheSov@cip-248.trustwave.com) Quit (Ping timeout: 480 seconds)
[16:08] * vata (~vata@207.96.182.162) has joined #ceph
[16:11] <Heebie> sounds like those of us who only installed ceph the week before last shouldn't have a security issue to worry about in the downloads?
[16:12] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[16:12] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[16:12] * hydro-b (~stefan@basseroet.kooman.org) Quit (Remote host closed the connection)
[16:14] * davidj (sid20957@id-20957.ealing.irccloud.com) has joined #ceph
[16:14] <rkeene> I don't understand, how can they not trust the signing key any more ? Is it not on an HSM like all other signing keys ?
[16:15] <TheSov2> i got dc'd i missed it, what happened with security?
[16:18] <Heebie> https://ceph.com/releases/important-security-notice-regarding-signing-key-and-binary-downloads-of-ceph/ << LobsterRoll provided this link
[16:19] * enax (~enax@hq.ezit.hu) Quit (Ping timeout: 480 seconds)
[16:19] * jasuarez (~jasuarez@243.Red-81-39-64.dynamicIP.rima-tde.net) Quit (Quit: WeeChat 1.2)
[16:19] * yanzheng (~zhyan@125.71.108.204) Quit (Quit: This computer has gone to sleep)
[16:20] <TheSov2> fak seriously?
[16:20] * neurodrone (~neurodron@108.60.145.130) Quit (Quit: neurodrone)
[16:21] <TheSov2> gonna shutdown my test cluster
[16:21] * neurodrone (~neurodron@162.243.191.67) has joined #ceph
[16:23] * sudocat (~dibarra@192.185.1.20) has joined #ceph
[16:23] * Quackie (~theghost9@kiljoy00.arvixevps.com) has joined #ceph
[16:25] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[16:25] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[16:25] <rkeene> I don't know how what they say is possible unless they were doing something *REAL* dumb like having private keys outside of an HSM
[16:25] <rkeene> (But does anyone really do that ? It's just so easy to use an HSM)
[16:25] * vbellur (~vijay@122.172.57.91) has joined #ceph
[16:25] <TheSov2> this is not instilling much confidence
[16:26] * brutuscat (~brutuscat@41.Red-83-47-113.dynamicIP.rima-tde.net) Quit (Remote host closed the connection)
[16:27] <qman> I'm guessing that's why ipv6 wasn't working when I was trying to install last week
[16:28] <qman> On download.ceph.com
[16:28] <davidj> Ouch.
[16:29] * swami1 (~swami@zz2012427835719DFD82.userreverse.dion.ne.jp) Quit (Quit: Leaving.)
[16:31] <TheSov2> so are current downloads safe?
[16:31] <kiranos> it has been fixed
[16:32] <TheSov2> but are the binaries in the repos verified?
[16:32] <kiranos> yes
[16:32] <TheSov2> ok sweet
[16:32] <TheSov2> but i still shut down my existing cluster just in case
[16:32] <TheSov2> ill rebuild it from scratch
[16:32] <zenpac1> Anyone have access to Calarmari Archives? I'm unable to access.
[16:33] <kiranos> I've not looked at calamari, seems pretty unstable :)
[16:33] <fen> can ceph compress the object store data?
[16:33] * alfredodeza (~alfredode@198.206.133.89) has left #ceph
[16:34] <TheSov2> fen, do compression on the FS side
[16:34] <fen> using?
[16:34] <TheSov2> btrfs
[16:34] <TheSov2> or zfsa
[16:35] <kiranos> you can look at http://docs.ceph.com/docs/v0.86/dev/erasure-coded-pool/
[16:38] <fen> is btrfs production stable?
[16:38] <pam> zenpac1: me too. getting a forbidden
[16:39] <fen> We recommend btrfs for testing, development, and any non-critical deployments.
[16:39] <fen> seems not likely
[16:39] * nickpappin (~nickpappi@wl-nat109.it.wsu.edu) has joined #ceph
[16:43] <bpkroth> i've been using btrfs for a while - my only recommendation so far is to add the autodefrag mount option
[16:43] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[16:43] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[16:44] <magicrobotmonkey> is there a way to list all currently configured cache tiers?
[16:47] <magicrobotmonkey> zenpac1: https://github.com/ceph/calamari ?
[16:47] <magicrobotmonkey> also https://github.com/ceph/calamari-clients
[16:47] <magicrobotmonkey> I've found it to be a bit of a pita to get up and running and quite unstable
[16:48] <magicrobotmonkey> but i think the unstability may be because i didn't give its vm enough memory
[16:48] * fen (~fen@HSI-KBW-217-008-056-240.hsi.kabelbw.de) Quit (Quit: fen)
[16:48] * kanagaraj (~kanagaraj@27.7.8.173) has joined #ceph
[16:48] <magicrobotmonkey> http://calamari.readthedocs.org/en/latest/operations/index.html
[16:48] <magicrobotmonkey> this is what i used to install it
[16:51] * moore (~moore@64.202.160.88) has joined #ceph
[16:53] * Quackie (~theghost9@4Z9AAAH1E.tor-irc.dnsbl.oftc.net) Quit ()
[16:54] * Aeso (~Genji221@2601:404:c200:420:88f2:9738:5b45:ebd8) Quit (Ping timeout: 480 seconds)
[16:54] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[16:54] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Remote host closed the connection)
[16:56] * brutuscat (~brutuscat@41.Red-83-47-113.dynamicIP.rima-tde.net) has joined #ceph
[16:57] * fen (~fen@dslb-094-216-233-194.094.216.pools.vodafone-ip.de) has joined #ceph
[16:57] * b0e (~aledermue@213.95.25.82) Quit (Quit: Leaving.)
[16:58] * kefu|afk (~kefu@114.86.210.253) Quit (Quit: My Mac has gone to sleep. ZZZzzz???)
[16:58] <kiranos> magicrobotmonkey: packages and repo or no go
[16:59] <kiranos> :)
[16:59] <kiranos> I'm using centos7
[16:59] * erhudy (uid89730@id-89730.ealing.irccloud.com) Quit (Quit: Connection closed for inactivity)
[17:01] * Destreyf (~quassel@email.newagecomputers.info) has joined #ceph
[17:04] * kefu (~kefu@114.86.210.253) has joined #ceph
[17:06] * linjan (~linjan@86.62.112.22) Quit (Ping timeout: 480 seconds)
[17:12] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[17:12] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[17:13] * yguang11 (~yguang11@66.228.162.44) has joined #ceph
[17:13] * analbeard (~shw@support.memset.com) Quit (Quit: Leaving.)
[17:18] * kefu (~kefu@114.86.210.253) Quit (Max SendQ exceeded)
[17:18] * pam (~pam@193.106.183.1) Quit (Quit: pam)
[17:19] * kefu (~kefu@114.86.210.253) has joined #ceph
[17:23] * UnknightMalta (~Blackjack@40.127.130.88) Quit (Read error: Connection reset by peer)
[17:23] * Aeso_ (~AesoSpade@c-68-37-97-11.hsd1.mi.comcast.net) Quit (Quit: Leaving)
[17:27] <fen> bpkroth: in production?
[17:28] * Aeso (~AesoSpade@c-68-37-97-11.hsd1.mi.comcast.net) has joined #ceph
[17:28] <debian112> Can someone help on ceph monitors? I had a ceph cluster working fine with 1 monitor, but when I try to install the second monitor, it completely froze the systems from running any ceph commands. How do I troubleshoot this?
[17:29] <fen> anybody knows a not so expensive 48 port switch which does distributed lacp (lacp over multiple switches)
[17:31] * pam (~pam@5.170.7.172) has joined #ceph
[17:32] * remy1991 (~ravi@115.114.59.182) Quit (Ping timeout: 480 seconds)
[17:33] <fen> for hp the lowest series supporting that seems to be the 2920 which are not so cheap
[17:34] * bitserker (~toni@81.184.9.72.dyn.user.ono.com) Quit (Quit: Leaving.)
[17:34] * bitserker (~toni@81.184.9.72.dyn.user.ono.com) has joined #ceph
[17:35] * mattbenjamin (~mbenjamin@aa2.linuxbox.com) Quit (Remote host closed the connection)
[17:35] <m0zes> gigabit or 10-gigabit?
[17:35] <m0zes> I'd go with a switch that supports cumulus linux with clagd. penguin computing has some nice ones.
[17:36] <fen> gigabit
[17:36] <fen> 10gb is too expensive
[17:36] * bitserker (~toni@81.184.9.72.dyn.user.ono.com) Quit ()
[17:36] * bitserker (~toni@81.184.9.72.dyn.user.ono.com) has joined #ceph
[17:37] * kefu (~kefu@114.86.210.253) Quit (Quit: My Mac has gone to sleep. ZZZzzz???)
[17:37] * bitserker (~toni@81.184.9.72.dyn.user.ono.com) Quit ()
[17:37] * bitserker (~toni@81.184.9.72.dyn.user.ono.com) has joined #ceph
[17:38] <Be-El> fen: we use three 2920. it is not that expensive, and the stacking features are really nice. for ceph the throughput might not be sufficient (depending on workload, number of osd and number of clients)
[17:39] <lookcrabs> I have a weird one. I am checking the quotas in ceph S3 via radosgw-admin user stats --uid; and it is around half the usage displayed in swift stat for the same user. Both are in bytes so I don't know how this could be or which one is right.
[17:39] * bitserker (~toni@81.184.9.72.dyn.user.ono.com) Quit ()
[17:39] * ade (~abradshaw@tmo-100-152.customers.d1-online.com) Quit (Quit: Too sexy for his shirt)
[17:39] * bitserker (~toni@81.184.9.72.dyn.user.ono.com) has joined #ceph
[17:40] <fen> be-el: i hope gb is enough - we do not expect a high io load on ceph. we only use it as object store with a mild workload
[17:41] <fen> it's more or a less a big data dump
[17:41] * brutuscat (~brutuscat@41.Red-83-47-113.dynamicIP.rima-tde.net) Quit (Remote host closed the connection)
[17:42] <mfa298_> debian112: at the point you tell ceph about the new monitor it probably will freeze as you don't have quorum, it would probably recover once the new monitor is started - however I've never tried a cluster like that, I've only ever deployed with a minimum of 3 mons
[17:43] <lookcrabs> http://pastebin.com/8BT2tKPj
[17:43] <debian112> it's been froze for over 24 hours
[17:43] <magicrobotmonkey> debian112: check out quanta, then can do cumulus linux and also have their own os
[17:44] * TMM (~hp@sams-office-nat.tomtomgroup.com) Quit (Quit: Ex-Chat)
[17:47] * rakeshgm (~rakesh@121.244.87.117) Quit (Quit: Leaving)
[17:47] <mfa298_> debian112: is the ceph-mon process running on both machines ?
[17:49] * mattbenjamin (~mbenjamin@aa2.linuxbox.com) has joined #ceph
[17:49] * pam (~pam@5.170.7.172) Quit (Read error: Connection reset by peer)
[17:49] <debian112> mfa298_: yes
[17:50] <mfa298_> you might want to see if there's anything in the logs about whether they're having issues. It sounds like they've not joined properly so you don't have quorum
[17:50] <debian112> but any ceph commands are not working
[17:51] <lookcrabs> debian112: the ceph commands are going to try to talk to the primary monitor and probably time out. I would check the logs on your monitor and possibly query the admin daemon
[17:51] <mfa298_> ceph commands won't work when you don't have quorum,
[17:51] <debian112> the first one shows this: http://paste.debian.net/318582/
[17:52] <lookcrabs> "quorum": [],
[17:52] <lookcrabs> "outside_quorum": [
[17:52] <lookcrabs> "sac-mon1"
[17:52] <debian112> the second shows this: http://paste.debian.net/318583/
[17:53] <lookcrabs> debian112: You're looking for something like this::
[17:53] <lookcrabs> http://paste.debian.net/318584/
[17:54] <lookcrabs> your monitors are not talking/sync'd. What does the log say in /var/log/ceph/?
[17:54] <debian112> mon1 log: http://paste.debian.net/318586/
[17:55] <debian112> mon2 log: http://paste.debian.net/318587/
[17:57] <lookcrabs> what is in /var/lib/ceph/mon/*mon*
[17:58] <lookcrabs> sounds like your monitors have two different keys or not talking. Check the keyring and check your monitor keys
[17:58] <fen> is it a problem if the files stored in ceph are really small?
[17:58] <TheSov2> damn the security breach! now i have to rebuild my test cluster!
[17:59] <lookcrabs> TheSov2: what?
[17:59] <fen> and does ceph deduplication
[17:59] <TheSov2> i build my ceph test cluster on hammer.2
[17:59] <TheSov2> fen, no, but you can do that from the FS side
[17:59] <debian112> /var/lib/ceph/mon/ceph-sac-mon1/keyring needs to be the same on all monitors?
[18:01] <lookcrabs> http://docs.ceph.com/docs/master/install/manual-deployment/#monitor-bootstrapping
[18:01] <lookcrabs> debian112:
[18:01] <TheSov2> i often wonder why people would build out a ceph cluster and get all that commodity pricing on disk and still want things like dedupe
[18:02] <fen> thesov2: i just wanted to know :)
[18:02] * ChrisNBlum (~ChrisNBlu@dhcp-ip-217.dorf.rwth-aachen.de) Quit (Ping timeout: 480 seconds)
[18:02] <TheSov2> oh well then , NO!
[18:02] <TheSov2> lol
[18:02] <davidj> @TheSov2 to save even more funds. :p
[18:02] <cephalobot> davidj: Error: "TheSov2" is not a valid command.
[18:02] <lookcrabs> TheSov2: is there a link to the breach?
[18:02] <davidj> @cephalobot: Thanks.
[18:02] <cephalobot> davidj: Error: "cephalobot:" is not a valid command.
[18:03] <TheSov2> davidj, dont use @ unless u want the bot to look something up :)
[18:03] <davidj> I realize that now ;)
[18:03] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[18:03] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[18:05] * ChrisNBlum (~ChrisNBlu@storchris.dorf.rwth-aachen.de) has joined #ceph
[18:06] <lookcrabs> TheSov2: sorry to bug you. Can you elaborate on the security issue with hammer? Is it the key repo issue?
[18:07] <kiranos> yes the https://ceph.com/releases/important-security-notice-regarding-signing-key-and-binary-downloads-of-ceph/ issue
[18:07] <lookcrabs> oh thanks kiranos .. I was freaking out "oh no not another one"
[18:07] <lookcrabs> thanks kiranos ^_^
[18:07] <kiranos> np :) it was the same one
[18:07] <mfa298_> you should be able to just update the keys on the boxes without needing to rebuild everything
[18:08] * Kioob`Taff (~plug-oliv@2a01:e35:2e8a:1e0::42:10) Quit (Quit: Leaving.)
[18:10] <rkeene> Still no closer to knowing how the key could even be compromised
[18:12] <debian112> hey all the: mon keys were different, once I sync them up things worked fine
[18:12] <debian112> thanks everyone!
[18:13] * garphy is now known as garphy`aw
[18:16] * kanagaraj (~kanagaraj@27.7.8.173) Quit (Quit: Leaving)
[18:17] <LobsterRoll> do you guys use rados bench or cbt for benchmarkign your clusters?
[18:21] * nickpappin_ (~nickpappi@wl-nat107.it.wsu.edu) has joined #ceph
[18:24] * nickpappin (~nickpappi@wl-nat109.it.wsu.edu) Quit (Read error: Connection reset by peer)
[18:27] * rendar (~I@host235-46-dynamic.31-79-r.retail.telecomitalia.it) Quit (Ping timeout: 480 seconds)
[18:28] * stxShadow (~jens@92.39.19.242.fixip.bitel.net) Quit (Remote host closed the connection)
[18:28] * i_m (~ivan.miro@deibp9eh1--blueice4n2.emea.ibm.com) Quit (Ping timeout: 480 seconds)
[18:30] * brutuscat (~brutuscat@105.34.133.37.dynamic.jazztel.es) has joined #ceph
[18:30] * brutuscat (~brutuscat@105.34.133.37.dynamic.jazztel.es) Quit (Remote host closed the connection)
[18:30] * rendar (~I@host235-46-dynamic.31-79-r.retail.telecomitalia.it) has joined #ceph
[18:31] * brutuscat (~brutuscat@105.34.133.37.dynamic.jazztel.es) has joined #ceph
[18:32] * nickpappin_ (~nickpappi@wl-nat107.it.wsu.edu) Quit (Ping timeout: 480 seconds)
[18:33] * nils_ (~nils_@doomstreet.collins.kg) has joined #ceph
[18:33] * remy1991 (~ravi@122.179.187.127) has joined #ceph
[18:34] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[18:34] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[18:35] * mykola (~Mikolaj@91.225.202.254) has joined #ceph
[18:36] * diq (~diq@2620:11c:f:2:c23f:d5ff:fe62:112c) has joined #ceph
[18:37] <diq> sorry, topic doesn't say anything but it appears that lists.ceph.com isn't working
[18:37] <diq> 404's
[18:42] * bitserker (~toni@81.184.9.72.dyn.user.ono.com) Quit (Quit: Leaving.)
[18:42] * bitserker (~toni@81.184.9.72.dyn.user.ono.com) has joined #ceph
[18:43] * adam1 (~adam@2601:1c2:300:beff:be5f:f4ff:fe55:a58f) has joined #ceph
[18:43] <via> i created a cache pool in front of a pool (test2 in front of test1), adn ran rados bench to create a bunch of objects in test2
[18:44] <via> but even with dirty_ratio set to 0.0 they don't seem to be going to test1 (backing store)
[18:44] * adam1 is now known as voxadam
[18:44] <via> flushing the cache does work, but shouldn't 0.0 suggest they go sooner rather than later?
[18:44] <via> also, is there a way to set a max dirty age?
[18:47] * ngoswami (~ngoswami@121.244.87.116) Quit (Quit: Leaving)
[18:48] * Kupo1 (~tyler.wil@23.111.254.159) has joined #ceph
[18:52] * Be-El (~quassel@fb08-bcf-pc01.computational.bio.uni-giessen.de) Quit (Remote host closed the connection)
[18:53] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[18:53] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[18:55] * kawa2014 (~kawa@89.184.114.246) Quit (Quit: Leaving)
[19:00] * pabluk is now known as pabluk_
[19:03] * dmick1 (~dmick@206.169.83.146) has joined #ceph
[19:03] * TMM (~hp@178-84-46-106.dynamic.upc.nl) has joined #ceph
[19:06] * alrick (~alrick@91.218.144.129) Quit (Remote host closed the connection)
[19:13] * dmick1 is now known as dmick
[19:14] * derjohn_mob (~aj@nl8x.mullvad.net) has joined #ceph
[19:14] * derjohn_mob (~aj@nl8x.mullvad.net) Quit (autokilled: This host may be infected. Mail support@oftc.net with questions. BOPM (2015-10-28 14:14:01))
[19:14] * TMM (~hp@178-84-46-106.dynamic.upc.nl) Quit (Ping timeout: 480 seconds)
[19:19] * TMM (~hp@178-84-46-106.dynamic.upc.nl) has joined #ceph
[19:23] * i_m (~ivan.miro@nat-5-carp.hcn-strela.ru) has joined #ceph
[19:24] * rakeshgm (~rakesh@106.51.29.129) has joined #ceph
[19:29] * thomnico (~thomnico@2a01:e35:8b41:120:4884:9cb6:f7cc:487b) Quit (Quit: Ex-Chat)
[19:30] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[19:30] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[19:31] * remy1991 (~ravi@122.179.187.127) Quit (Ping timeout: 480 seconds)
[19:33] * DV (~veillard@2001:41d0:1:d478::1) Quit (Remote host closed the connection)
[19:34] * DV (~veillard@2001:41d0:a:f29f::1) has joined #ceph
[19:35] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[19:35] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[19:39] * linjan (~linjan@176.195.227.255) has joined #ceph
[19:41] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[19:41] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[19:42] * Craig1 (~Adium@199-119-120-158.contegix.com) has joined #ceph
[19:43] <TheSov2> i dont understand why all the osd's on a host dont just use a pgmap application instead of making individual pgmap caches for each osd
[19:44] * dgurtner (~dgurtner@178.197.231.141) Quit (Ping timeout: 480 seconds)
[19:45] * clusterfudge (~Roy@pei69-1-78-193-103-77.fbxo.proxad.net) has joined #ceph
[19:47] <Craig1> Has anyone used ceph-disk (directly or via ceph-deploy) to setup an OSD with a separate journal disk in hammer?
[19:49] <Craig1> What we have used in firefly (and verified to work in giant as well) appears to be broken in hammer. There were some changes in ceph-disk which appear to assume that you are passing in a /dev/sdb3 type of device rather than a /dev/disk/by-id/???. device.
[19:50] * nardial (~ls@dslb-088-072-094-085.088.072.pools.vodafone-ip.de) Quit (Ping timeout: 480 seconds)
[19:50] <Craig1> ceph-disk -v prepare --cluster ceph --fs-type xfs -- /dev/sdg3 /dev/disk/by-id/scsi-3690b11c021dce1001db2a79c08366fbb-part4
[19:51] <Craig1> ^^^ that is the command which worked in firefly/giant and fails in hammer.
[19:52] <TheSov2> Craig1, yes i have done taht
[19:53] <Craig1> with success in hammer? if so, would I love to know how.
[19:54] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[19:54] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[19:54] <TheSov2> use by uuid
[20:02] * pabluk_ is now known as pabluk
[20:06] <nils_> I thought you only get an UUID on a formatted filesystem? Or do you mean partuuid?
[20:06] <Craig1> thought so as well, but also tried partuuid just now. I am seeing the same failed results in that case.
[20:07] <nils_> I've only used it without partitions so far
[20:07] <nils_> which works with disk by id
[20:07] <Craig1> for the record, this is the type of error I am seeing. the same parsing issue at the first numeric character of the device path.
[20:07] <Craig1> # ceph-disk -v prepare --cluster ceph --fs-type xfs -- /dev/sdg3 /dev/disk/by-partuuid/d1aaa603-6e74-481d-8a47-9ffd8a7ad119
[20:07] <Craig1> INFO:ceph-disk:Running command: /usr/bin/ceph-osd --cluster=ceph --show-config-value=fsid
[20:07] <Craig1> INFO:ceph-disk:Running command: /usr/bin/ceph-conf --cluster=ceph --name=osd. --lookup osd_mkfs_options_xfs
[20:07] <Craig1> INFO:ceph-disk:Running command: /usr/bin/ceph-conf --cluster=ceph --name=osd. --lookup osd_fs_mkfs_options_xfs
[20:07] <Craig1> INFO:ceph-disk:Running command: /usr/bin/ceph-conf --cluster=ceph --name=osd. --lookup osd_mount_options_xfs
[20:08] <Craig1> INFO:ceph-disk:Running command: /usr/bin/ceph-conf --cluster=ceph --name=osd. --lookup osd_fs_mount_options_xfs
[20:08] <Craig1> INFO:ceph-disk:Running command: /usr/bin/ceph-osd --cluster=ceph --show-config-value=osd_journal_size
[20:08] <Craig1> INFO:ceph-disk:Running command: /usr/bin/ceph-conf --cluster=ceph --name=osd. --lookup osd_cryptsetup_parameters
[20:08] <Craig1> INFO:ceph-disk:Running command: /usr/bin/ceph-conf --cluster=ceph --name=osd. --lookup osd_dmcrypt_key_size
[20:08] <Craig1> INFO:ceph-disk:Running command: /usr/bin/ceph-conf --cluster=ceph --name=osd. --lookup osd_dmcrypt_type
[20:08] <Craig1> DEBUG:ceph-disk:Journal /dev/disk/by-partuuid/d1aaa603-6e74-481d-8a47-9ffd8a7ad119 is a partition
[20:08] <Craig1> WARNING:ceph-disk:OSD will not be hot-swappable if journal is not the same device as the osd data
[20:08] <Craig1> INFO:ceph-disk:Running command: /sbin/blkid -p -o udev /dev/disk/by-partuuid/d1aaa603-6e74-481d-8a47-9ffd8a7ad119
[20:08] <Craig1> DEBUG:ceph-disk:Journal /dev/disk/by-partuuid/d1aaa603-6e74-481d-8a47-9ffd8a7ad119 was previously prepared with ceph-disk. Reusing it.
[20:08] <Craig1> INFO:ceph-disk:Running command: /sbin/sgdisk -i 1 /dev/disk/by-partuuid/d
[20:08] <Craig1> Problem opening /dev/disk/by-partuuid/d for reading! Error is 2.
[20:08] <Craig1> The specified file does not exist!
[20:08] <Craig1> DEBUG:ceph-disk:Reusing journal with uuid None
[20:08] <Craig1> Traceback (most recent call last):
[20:08] <Craig1> File "/usr/sbin/ceph-disk", line 2994, in <module>
[20:08] <Craig1> main()
[20:08] * enax (~enax@94-21-125-43.pool.digikabel.hu) has joined #ceph
[20:08] <Craig1> File "/usr/sbin/ceph-disk", line 2972, in main
[20:08] <Craig1> args.func(args)
[20:08] <Craig1> File "/usr/sbin/ceph-disk", line 1644, in main_prepare
[20:08] <Craig1> luks=luks
[20:08] <Craig1> File "/usr/sbin/ceph-disk", line 1284, in prepare_journal
[20:08] <Craig1> return prepare_journal_dev(data, journal, journal_size, journal_uuid, journal_dm_keypath, cryptsetup_parameters, luks)
[20:08] <Craig1> File "/usr/sbin/ceph-disk", line 1123, in prepare_journal_dev
[20:08] <Craig1> assert os.path.exists(journal_symlink)
[20:08] <Craig1> AssertionError
[20:08] <Craig1> nils_, I have multiple osds, so will need multiple journals per SSD.
[20:08] <voxadam> Pastebin?
[20:08] <nils_> Craig1, it'll create a new partition automagically
[20:09] * rotbeard (~redbeard@185.32.80.238) Quit (Quit: Leaving)
[20:09] <nils_> in accordance with your journal size
[20:10] <voxadam> When using a 'replicated pool' are objects stored in their entirety across the specified number of placement groups?
[20:10] <LobsterRoll> so I just started testing with rados bench, and am testing on one pool with a replica level of 2 and another wiht a replica level of 10 (I expected the replica level of 10 to have 1/5 the measure bandwidth of the relica level 2 pool) However they had almost the identical bandwidth. How is that possible?
[20:10] <Craig1> Sorry voxadam. Accidentally pasted here.
[20:11] <voxadam> :)
[20:11] <nils_> LobsterRoll, just a hunch but given enough resources one could write all replica in parallel.
[20:11] * fen (~fen@dslb-094-216-233-194.094.216.pools.vodafone-ip.de) Quit (Quit: fen)
[20:12] <LobsterRoll> I only have 2 osd hosts (with 12 drives in them each)
[20:12] <LobsterRoll> nils_: is it possible its not able to obtain the replica ten for the size=10 pool
[20:12] <LobsterRoll> because i dont have 10 osd
[20:13] <LobsterRoll> but is actually replicating twice as well?
[20:13] <LobsterRoll> Craig1: I experienced the same issue but it was with the journal on the same disk
[20:13] * ira (~ira@nat-pool-bos-t.redhat.com) Quit (Ping timeout: 480 seconds)
[20:13] <nils_> you can't get 10 replica with 2 osd...
[20:14] <LobsterRoll> it appears to use the wrong parition uuid for the symlink in /var/lib/ceph/osd/journal
[20:14] <LobsterRoll> nils_: 2 hosts, 24 disks so 24 osds
[20:14] <LobsterRoll> to clarifuy
[20:14] * clusterfudge (~Roy@4K6AACA5G.tor-irc.dnsbl.oftc.net) Quit ()
[20:15] <LobsterRoll> I dont have any object errors in my ceph status (though I do have degraded and undersized pgs)
[20:15] <nils_> LobsterRoll, yeah, because you can't satisfy the crush map with only 2 hosts, by default it'll try to place replica on different hosts/failure domains
[20:16] <nils_> so you're undersized and degraded (also the title of my autobiography)
[20:16] <LobsterRoll> you forgot misplaced as well
[20:17] <lookcrabs> Can someone point me in the right direction. I had a weird issue where a single hosts OSDs filled up in ceph. I am using 3x replication and have distribution by host. The odd thing is these osds filled up to 91% while the rest of the osds were at 5-11% full. Is there a way to see what objects were mapped to those osds or what may be the cause?
[20:18] <LobsterRoll> are you custom writing your crush map lookcrabs ?
[20:19] * x303 (~Kaervan@4Z9AAAID8.tor-irc.dnsbl.oftc.net) has joined #ceph
[20:21] * dmick (~dmick@206.169.83.146) has left #ceph
[20:23] <Craig1> Thanks all. I see the issue now. Our wrappers calling ceph-deploy (and ultimately ceph-disk) were looking up the device by id when passing it in to ensure that the symlink wasn't pointing to the numbered device. The good news is that hammer now does that lookup for us, but we now just need to adjust our wrappers since hammer will choke on someone passing in the by-id reference for the device.
[20:29] * rakeshgm (~rakesh@106.51.29.129) Quit (Quit: Leaving)
[20:30] * brutuscat (~brutuscat@105.34.133.37.dynamic.jazztel.es) Quit (Remote host closed the connection)
[20:32] * daviddcc (~dcasier@LAubervilliers-656-1-16-160.w217-128.abo.wanadoo.fr) Quit (Ping timeout: 480 seconds)
[20:49] * x303 (~Kaervan@4Z9AAAID8.tor-irc.dnsbl.oftc.net) Quit ()
[20:51] * janos_ (~messy@static-71-176-211-4.rcmdva.fios.verizon.net) Quit (Read error: Connection reset by peer)
[20:52] * janos_ (~messy@static-71-176-211-4.rcmdva.fios.verizon.net) has joined #ceph
[20:53] * enax (~enax@94-21-125-43.pool.digikabel.hu) has left #ceph
[20:54] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[20:54] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[20:55] * TheSov3 (~TheSov@204.13.200.248) has joined #ceph
[21:01] * mykola (~Mikolaj@91.225.202.254) Quit (Quit: away)
[21:03] * TheSov2 (~TheSov@204.13.200.248) Quit (Ping timeout: 480 seconds)
[21:08] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[21:08] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[21:08] * garphy`aw is now known as garphy
[21:13] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[21:13] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[21:27] * sudocat (~dibarra@192.185.1.20) Quit (Ping timeout: 480 seconds)
[21:29] * Icey (~chris@0001bbad.user.oftc.net) Quit (Quit: Leaving)
[21:32] * linjan (~linjan@176.195.227.255) Quit (Ping timeout: 480 seconds)
[21:32] * sudocat (~dibarra@66.196.218.45) has joined #ceph
[21:36] * daviddcc (~dcasier@84.197.151.77.rev.sfr.net) has joined #ceph
[21:42] * linjan (~linjan@176.195.227.255) has joined #ceph
[21:43] * georgem1 (~Adium@206.108.127.16) Quit (Quit: Leaving.)
[21:47] * Craig1 (~Adium@199-119-120-158.contegix.com) Quit (Quit: Leaving.)
[21:53] * Nats__ (~natscogs@114.31.195.238) has joined #ceph
[21:59] * nardial (~ls@dslb-088-072-094-085.088.072.pools.vodafone-ip.de) has joined #ceph
[21:59] * LobsterRoll (~LobsterRo@140.247.242.44) Quit (Ping timeout: 480 seconds)
[22:00] * Nats_ (~natscogs@114.31.195.238) Quit (Ping timeout: 480 seconds)
[22:01] * olid11 (~olid1982@185.17.206.92) has joined #ceph
[22:01] <olid11> hi, where to report a bug ? :)
[22:01] * rendar (~I@host235-46-dynamic.31-79-r.retail.telecomitalia.it) Quit (Ping timeout: 480 seconds)
[22:01] * lathiat (~lathiat@2403:dc00::117) Quit (Ping timeout: 480 seconds)
[22:02] <m0zes> tracker.ceph.com
[22:03] <lurbs> Drunk dial Sage.
[22:04] * rendar (~I@host235-46-dynamic.31-79-r.retail.telecomitalia.it) has joined #ceph
[22:07] <Aeso> So Accelio supports the ability to reconnect if an interface goes down, but it has to be supported in code. Is there any point (from a high availability perspective) in bonding interfaces and running Ceph's internal network across those bonds?
[22:08] <Aeso> Obviously you won't get the RDMA throughput from both links, but in the event of a switch failure it'd be nice to have some backend redundancy.
[22:10] * DV (~veillard@2001:41d0:a:f29f::1) Quit (Remote host closed the connection)
[22:10] <cetex> i'd say yes
[22:11] <cetex> bond 2 interfaces, preferrably to 2 different switches
[22:11] <cetex> mlag on for example arista comes to mind
[22:11] <cetex> or stacking if you rather want that, but i don't think that's real redundancy.. :>
[22:11] * DV (~veillard@2001:41d0:1:d478::1) has joined #ceph
[22:12] <cetex> we're kinda doing that with other services that need availability, but we only have active+standby (those services can't handle half the bandwidth, very important that we don't oversubscribe and then end up with everything broken when one redundancy-path goes down)
[22:13] * bitserker (~toni@81.184.9.72.dyn.user.ono.com) Quit (Quit: Leaving.)
[22:13] <cetex> so i'd recommend you to actually kill one path once in a while to see that everything is working as it should, and to know the consequences of it..
[22:15] <kiranos> I using bonding
[22:16] <kiranos> Bonding Mode: fault-tolerance
[22:16] <kiranos> for now but will probably go to LACP
[22:16] * nardial (~ls@dslb-088-072-094-085.088.072.pools.vodafone-ip.de) Quit (Quit: Leaving)
[22:16] <kiranos> 802.3ad
[22:16] <Aeso> Sure. But my concern is with RDMA. If I'm running transmitting via RDMA over a bonded netowrk and the link carrying that RDMA traffic goes down, what state does that leave Ceph in? Obviously Ceph will time-out and reinitialize eventually, but what happens to any writes in flight?
[22:19] <kiranos> I'm not sure but there should be some ack if the writes has been written ok, if not there will be an error
[22:19] <kiranos> as expected
[22:19] <Aeso> Hm, true.
[22:23] * jrankin (~jrankin@d53-64-170-236.nap.wideopenwest.com) Quit (Quit: Leaving)
[22:25] * garphy is now known as garphy`aw
[22:25] * yguang11 (~yguang11@66.228.162.44) Quit (Remote host closed the connection)
[22:27] * yguang11 (~yguang11@66.228.162.44) has joined #ceph
[22:30] * dgurtner (~dgurtner@217-162-119-191.dynamic.hispeed.ch) has joined #ceph
[22:32] * lathiat (~lathiat@2403:dc00::117) has joined #ceph
[22:32] * TheSov3 (~TheSov@204.13.200.248) Quit (Read error: Connection reset by peer)
[22:33] <cetex> hm..
[22:33] * voxadam (~adam@2601:1c2:300:beff:be5f:f4ff:fe55:a58f) Quit (Ping timeout: 480 seconds)
[22:33] <cetex> question about ssd's then
[22:33] <cetex> anyone know if those ram on pci-card things are available still?
[22:33] <cetex> i just need for the journal
[22:33] <cetex> ~30GB of ram would be nice
[22:34] <cetex> since according to my calculation i'll write ~1TB through those 30GB used very day, and we're going to scale up soonish.
[22:35] <cetex> i've been looking for pci-e based ssd's as well, intel 3700 series would work, but the smallest variety they come in is 400GB which is roughly $900 over here.
[22:35] <cetex> (sweden)
[22:35] <cetex> i'd like one of those in 100gb or 200gb.
[22:35] <cetex> to keep costs down
[22:36] <cetex> would mean one ssd per 3 hdd's.
[22:36] <cetex> otherwise i need to sacrifice one of the three disks-slots on the blade for the ssd..
[22:38] * Icey (~chris@0001bbad.user.oftc.net) has joined #ceph
[22:39] <qman> I don't know of any RAM disk devices that big that you can actually buy
[22:39] <qman> you can get the gigabyte i-RAM, but that's limtied to 4GB of DDR400
[22:39] <qman> there was a bigger ADATA one for DDR2 but I've never found it actually for sale anywhere, and even that's still pretty limited in capacity
[22:40] * fred`` (fred@earthli.ng) Quit (Quit: +++ATH0)
[22:40] <qman> if disk slots are the concern, I was looking into SATA DOMs, disk on module
[22:40] <qman> I've seen 64GB ones
[22:41] <cetex> yeah..
[22:41] <cetex> problem is TBW limit of ssd's
[22:41] <qman> yeah
[22:41] <cetex> intel's top of the line product is gonna last 3.5 years max in our setup, if we continue scaling as we've done earlier we're looking at 1.5-2years..
[22:43] <cetex> we're basically gonna write 8TB of data to each 8TB hdd weekly for now.
[22:43] * dneary (~dneary@nat-pool-bos-u.redhat.com) Quit (Ping timeout: 480 seconds)
[22:44] <qman> I'd love to be able to buy a DDR3 or DDR4 RAM disk device, but as far as I know none exist, mostly due to the market failure of previous devices and the ubiquity of SSDs
[22:44] <cetex> as in, ceph-cluster is going to be treated as a huge ring-buffer, so continuously writing and deleting data, cluster will be at ~70% disk-usage constantly, and we'll have deleted / rewritten all data each week.
[22:44] * Destreyf_ (~quassel@host-208-117-122-114.beyondbb.com) has joined #ceph
[22:45] <lathiat> TBW of SSDs is quite low when you look at it :~
[22:45] <lathiat> my desktop one here has a limit of like 150
[22:45] * Destreyf__ (~quassel@email.newagecomputers.info) has joined #ceph
[22:45] * Destreyf__ (~quassel@email.newagecomputers.info) Quit ()
[22:45] <lathiat> i know people doing like video editing/recording/etc that have to replace them pretty rapidly, inside 1 year, because theyre writing a TB every time they use it, daily
[22:45] * Destreyf__ (~quassel@email.newagecomputers.info) has joined #ceph
[22:45] * linjan (~linjan@176.195.227.255) Quit (Ping timeout: 480 seconds)
[22:45] * Destreyf (~quassel@email.newagecomputers.info) Quit (Read error: Connection reset by peer)
[22:46] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[22:46] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[22:46] <cetex> yeah.
[22:48] <cetex> the 3700 has 1825TBW/100GB
[22:48] <cetex> and we'll kill them in short time still..
[22:48] * fred`` (fred@earthli.ng) has joined #ceph
[22:48] * nhm (~nhm@c-50-171-139-246.hsd1.mn.comcast.net) Quit (Quit: Lost terminal)
[22:49] <cetex> 1825TBW = 1TB/day over 5 years.
[22:49] * mtanski (~mtanski@65.244.82.98) Quit (Quit: mtanski)
[22:49] <cetex> if we scale up a bit and write 2TB/day on them (or add more than 3 8TB hdd's to one SSD) it's gonna drop rapidly
[22:49] <cetex> other ssd manufacturers worth looking into than intel?
[22:49] <cetex> *are there other ...
[22:50] <qman> I use samsung for the most part but I don't think they have anything with that high of a write life
[22:50] <cetex> I like the cost of the 3700 series 100GB ssd, it's like $200 or something, but it requires a disk-slot. so something pci-based of same size, and up to $300 would be extremely nice.
[22:50] <qman> I don't have a lot of info on the enterprise-marketed stuff
[22:52] * Destreyf_ (~quassel@host-208-117-122-114.beyondbb.com) Quit (Ping timeout: 480 seconds)
[22:53] * mattbenjamin (~mbenjamin@aa2.linuxbox.com) Quit (Quit: Leaving.)
[22:53] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[22:53] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[23:00] * georgem (~Adium@184.151.179.22) has joined #ceph
[23:01] * georgem (~Adium@184.151.179.22) Quit ()
[23:05] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[23:05] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[23:05] * Icey (~chris@0001bbad.user.oftc.net) Quit (Ping timeout: 480 seconds)
[23:12] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[23:12] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[23:14] * Icey (~chris@0001bbad.user.oftc.net) has joined #ceph
[23:20] * pabluk is now known as pabluk_
[23:22] * fridim_ (~fridim@56-198-190-109.dsl.ovh.fr) Quit (Ping timeout: 480 seconds)
[23:23] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[23:23] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[23:27] <JoeJulian> I know you're supposed to use raw images with ceph/openstack. Does anybody know if you can use an iso? Aren't they essentially treated as raw images anyway?
[23:27] * mattbenjamin (~mbenjamin@76-206-42-105.lightspeed.livnmi.sbcglobal.net) has joined #ceph
[23:31] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[23:31] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[23:31] * sileht (~sileht@sileht.net) Quit (Read error: No route to host)
[23:32] * lcurtis (~lcurtis@47.19.105.250) Quit (Remote host closed the connection)
[23:36] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[23:36] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[23:42] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[23:42] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[23:47] * alfredodeza (~alfredode@198.206.133.89) has joined #ceph
[23:48] * Ceph-Log-Bot (~logstash@185.66.248.215) has joined #ceph
[23:48] * Ceph-Log-Bot (~logstash@185.66.248.215) Quit (Read error: Connection reset by peer)
[23:54] * daviddcc (~dcasier@84.197.151.77.rev.sfr.net) Quit (Remote host closed the connection)
[23:59] * LobsterRoll (~LobsterRo@209-6-180-200.c3-0.hdp-ubr1.sbo-hdp.ma.cable.rcn.com) has joined #ceph

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