#ceph IRC Log

Index

IRC Log for 2014-08-20

Timestamps are in GMT/BST.

[0:02] * xarses (~andreww@12.164.168.117) has joined #ceph
[0:07] * markbby (~Adium@168.94.245.4) Quit (Quit: Leaving.)
[0:07] * Eco (~Eco@adsl-99-105-53-247.dsl.pltn13.sbcglobal.net) has joined #ceph
[0:12] * Midnightmyth_ (~quassel@93-167-84-102-static.dk.customer.tdc.net) has joined #ceph
[0:17] * garphy is now known as garphy`aw
[0:18] * dneary (~dneary@nat-pool-bos-u.redhat.com) Quit (Ping timeout: 480 seconds)
[0:18] * Midnightmyth (~quassel@93-167-84-102-static.dk.customer.tdc.net) Quit (Ping timeout: 480 seconds)
[0:20] * qhartman (~qhartman@den.direwolfdigital.com) has joined #ceph
[0:22] * gregmark (~Adium@cet-nat-254.ndceast.pa.bo.comcast.net) Quit (Quit: Leaving.)
[0:22] * Nacer (~Nacer@203-206-190-109.dsl.ovh.fr) Quit (Remote host closed the connection)
[0:25] <qhartman> I'm looking at enabling data striping for my Openstack cluster that is storing images in ceph, as in: http://ceph.com/docs/master/architecture/#data-striping
[0:25] * JC (~JC@AMontpellier-651-1-27-181.w90-57.abo.wanadoo.fr) Quit (Quit: Leaving.)
[0:26] <qhartman> However, all the information I've found relates to using cephfs access, rather than librbd that I believe qemu is using. Does anyone have any experience with this and/or poitners to docs?
[0:28] <cookednoodles> qhartman, its automatic
[0:29] <gleam> rbd already breaks the devices into 4mb(?) chunks, so that's basically your stripe size
[0:30] <gleam> i guess 4mb is the default, and it's presumably configurable
[0:30] <cookednoodles> it is
[0:32] <qhartman> hm, it sounds like my use case (large monolithic VM images) would benefit from a non-default config though according to what I read at the link above.
[0:33] <qhartman> If it automatic, I'd be interested in discovering what the settings are and how to manipulate them
[0:35] <qhartman> aha: http://ceph.com/docs/next/rbd/rbd-openstack/
[0:35] <qhartman> This gives and example of a striping config, though it does not discuss it much
[0:38] <qhartman> hm, maybe not
[0:39] <qhartman> According to this paper from cern, tweaking it is not possible within the context of Openstack: http://zenodo.org/record/7570/files/CERN_openlab_report_Mcgilvary.pdf
[0:39] <qhartman> (at least as of last year)
[0:39] * sironside (~sironside@2a00:a600:603:0:8e70:5aff:fec9:ee14) has joined #ceph
[0:44] * sputnik13 (~sputnik13@207.8.121.241) Quit (Quit: My MacBook has gone to sleep. ZZZzzz???)
[0:44] * rendar (~I@host67-179-dynamic.7-87-r.retail.telecomitalia.it) Quit ()
[0:44] * hasues (~hasues@kwfw01.scrippsnetworksinteractive.com) Quit (Quit: Leaving.)
[0:44] * PerlStalker (~PerlStalk@2620:d3:8000:192::70) Quit (Quit: ...)
[0:47] * gero_santos (~gero_sant@111.68.38.95) has joined #ceph
[0:47] * gero_santos (~gero_sant@111.68.38.95) Quit ()
[0:56] * Concubidated (~Adium@66-87-66-136.pools.spcsdns.net) Quit (Ping timeout: 480 seconds)
[0:59] * zack_dolby (~textual@p8505b4.tokynt01.ap.so-net.ne.jp) Quit (Quit: My MacBook has gone to sleep. ZZZzzz???)
[1:04] * lofejndif (~lsqavnbok@99WAAAWG6.tor-irc.dnsbl.oftc.net) Quit (Quit: gone)
[1:04] * aknapp_ (~aknapp@fw125-01-outside-active.ent.mgmt.glbt1.secureserver.net) has joined #ceph
[1:05] * aknapp (~aknapp@fw125-01-outside-active.ent.mgmt.glbt1.secureserver.net) Quit (Read error: Operation timed out)
[1:05] * carmstrong (sid22558@id-22558.charlton.irccloud.com) has joined #ceph
[1:06] * t4nk193 (~oftc-webi@66.109.98.194) Quit (Quit: Page closed)
[1:06] <carmstrong> hi Cephers, I'm working through a lab/workshop setup of Ceph to mess around with it, but am stuck with `HEALTH_WARN 192 pgs degraded; 192 pgs stuck unclean`
[1:06] <carmstrong> it's been over an hour. I've read the docs and have investigated, and it looks like things are in a recovering state
[1:07] <carmstrong> but it doesn't seem like anything's happening
[1:09] * pentabular (~Adium@2601:9:4980:a01:78c5:c2e0:afdb:3610) has joined #ceph
[1:09] * aknapp (~aknapp@fw125-01-outside-active.ent.mgmt.glbt1.secureserver.net) has joined #ceph
[1:09] * aknapp_ (~aknapp@fw125-01-outside-active.ent.mgmt.glbt1.secureserver.net) Quit (Read error: Connection reset by peer)
[1:11] <lurbs> carmstrong: What do 'ceph osd tree' and 'ceph osd dump | grep size' say?
[1:12] <carmstrong> https://www.irccloud.com/pastebin/WeZQm0Do
[1:12] <lurbs> I'm guessing that you have fewer hosts containing OSDs than your replication level requires.
[1:12] * andreask (~andreask@h081217017238.dyn.cm.kabsi.at) has joined #ceph
[1:12] * ChanServ sets mode +v andreask
[1:12] <lurbs> Yep.
[1:12] <carmstrong> that would make sense, although I followed the installation guide where it says to change the OSD requirement to 2
[1:13] <lurbs> Depending on when you did that, it would have only affected newly created pools - not the default pre-created ones.
[1:13] <lurbs> Easy enough to resize them, though.
[1:14] * andreask (~andreask@h081217017238.dyn.cm.kabsi.at) Quit ()
[1:14] <lurbs> 'ceph osd pool set $pool size 2'
[1:14] <lurbs> Or add another OSD host.
[1:14] * andreask (~andreask@h081217017238.dyn.cm.kabsi.at) has joined #ceph
[1:14] * ChanServ sets mode +v andreask
[1:14] <carmstrong> makes sense. I configured it before I did anything else, though
[1:14] <lurbs> Where $pool is data, metadata and rbd.
[1:14] <carmstrong> here is what I"m writing: https://github.com/carmstrong/multinode-ceph-vagrant
[1:14] * lofejndif (~lsqavnbok@tor-exit.server6.tvdw.eu) has joined #ceph
[1:15] <carmstrong> oh wow
[1:15] <lurbs> You have 'osd pool default size = 2' in your config? If so, you might need to set it earlier in the build process, or make sure it's runtime set with something like:
[1:15] <carmstrong> sd pool default size = 2
[1:15] <carmstrong> no 'o'
[1:15] <carmstrong> :/
[1:15] <lurbs> Heh. vim-fail? ;)
[1:16] <carmstrong> looks like it
[1:16] <carmstrong> okay, this makes sense
[1:16] <carmstrong> can you explain how you saw that from my output?
[1:16] <carmstrong> trying to learn
[1:16] <lurbs> Only two OSD hosts (ceph-server-{2,3}) and each pool in the 'ceph osd dump' has 'size 3'.
[1:17] <lurbs> You may also want to change min_size to 1, depending on your requirements.
[1:17] <carmstrong> ah okay, and you got size from the second command
[1:17] <carmstrong> lemme give this a go. thanks!
[1:17] <lurbs> http://ceph.com/docs/master/rados/operations/pools/#set-pool-values
[1:17] <lurbs> min_size is described there.
[1:18] <lurbs> Seems to be a very common problem since the Firefly release, when a size of 3 became the default.
[1:18] <carmstrong> thanks!
[1:18] <carmstrong> I understood this as I read the documentation, but just fatfingered the config
[1:18] <carmstrong> unfortunately couldn't debug the issue myself
[1:19] * andreask (~andreask@h081217017238.dyn.cm.kabsi.at) Quit ()
[1:19] * linjan (~linjan@176.195.29.108) Quit (Ping timeout: 480 seconds)
[1:19] * andreask (~andreask@h081217017238.dyn.cm.kabsi.at) has joined #ceph
[1:19] * ChanServ sets mode +v andreask
[1:24] <bens> where are sercret_ids generated fro cephx
[1:24] * chuffpdx (~chuffpdx@208.186.186.51) has joined #ceph
[1:27] <bens> chasing an error here
[1:27] <bens> which I will continue to chase tomorrow
[1:27] * andreask (~andreask@h081217017238.dyn.cm.kabsi.at) Quit (Ping timeout: 480 seconds)
[1:31] * Midnightmyth_ (~quassel@93-167-84-102-static.dk.customer.tdc.net) Quit (Ping timeout: 480 seconds)
[1:31] <dmick> context for 'secret_ids', bens?
[1:32] * Nacer (~Nacer@203-206-190-109.dsl.ovh.fr) has joined #ceph
[1:32] * Nacer (~Nacer@203-206-190-109.dsl.ovh.fr) Quit (Remote host closed the connection)
[1:34] * LeaChim (~LeaChim@host86-159-115-162.range86-159.btcentralplus.com) Quit (Read error: Operation timed out)
[1:35] * lofejndif (~lsqavnbok@37PAAA20C.tor-irc.dnsbl.oftc.net) Quit (Quit: gone)
[1:37] * cookednoodles (~eoin@eoin.clanslots.com) Quit (Quit: Ex-Chat)
[1:37] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) has joined #ceph
[1:51] * KevinPerks (~Adium@2606:a000:80a1:1b00:c081:8fe8:8a8b:d4f4) Quit (Quit: Leaving.)
[1:51] * Lorne (Citadel@modemcable158.99-20-96.mc.videotron.ca) Quit (Read error: Connection reset by peer)
[1:51] * Lorne (Citadel@modemcable158.99-20-96.mc.videotron.ca) has joined #ceph
[1:55] * oms101 (~oms101@p20030057EA0A3700EEF4BBFFFE0F7062.dip0.t-ipconnect.de) Quit (Ping timeout: 480 seconds)
[1:56] * zack_dolby (~textual@e0109-114-22-95-89.uqwimax.jp) has joined #ceph
[1:56] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) Quit (Read error: Connection reset by peer)
[1:57] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) has joined #ceph
[1:57] * codice (~toodles@97-94-175-73.static.mtpk.ca.charter.com) Quit (Read error: Operation timed out)
[1:58] * codice (~toodles@97-94-175-73.static.mtpk.ca.charter.com) has joined #ceph
[2:01] * aknapp (~aknapp@fw125-01-outside-active.ent.mgmt.glbt1.secureserver.net) Quit (Remote host closed the connection)
[2:03] * oms101 (~oms101@p20030057EA09DD00EEF4BBFFFE0F7062.dip0.t-ipconnect.de) has joined #ceph
[2:04] * wschulze (~wschulze@cpe-69-206-251-158.nyc.res.rr.com) Quit (Quit: Leaving.)
[2:05] * aknapp_ (~aknapp@fw125-01-outside-active.ent.mgmt.glbt1.secureserver.net) has joined #ceph
[2:08] * aknapp_ (~aknapp@fw125-01-outside-active.ent.mgmt.glbt1.secureserver.net) Quit (Remote host closed the connection)
[2:08] * aknapp (~aknapp@fw125-01-outside-active.ent.mgmt.glbt1.secureserver.net) has joined #ceph
[2:09] * thb (~me@0001bd58.user.oftc.net) Quit (Ping timeout: 480 seconds)
[2:10] * aknapp (~aknapp@fw125-01-outside-active.ent.mgmt.glbt1.secureserver.net) Quit (Remote host closed the connection)
[2:10] * aknapp (~aknapp@fw125-01-outside-active.ent.mgmt.glbt1.secureserver.net) has joined #ceph
[2:16] * sputnik13 (~sputnik13@wsip-68-105-248-60.sd.sd.cox.net) has joined #ceph
[2:18] * aknapp (~aknapp@fw125-01-outside-active.ent.mgmt.glbt1.secureserver.net) Quit (Ping timeout: 480 seconds)
[2:18] * rmoe (~quassel@12.164.168.117) Quit (Ping timeout: 480 seconds)
[2:18] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) Quit (Quit: Leaving.)
[2:19] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) has joined #ceph
[2:23] * rmoe (~quassel@173-228-89-134.dsl.static.sonic.net) has joined #ceph
[2:34] * Eco (~Eco@adsl-99-105-53-247.dsl.pltn13.sbcglobal.net) Quit (Ping timeout: 480 seconds)
[2:34] * yguang11 (~yguang11@vpn-nat.peking.corp.yahoo.com) has joined #ceph
[2:35] * xarses (~andreww@12.164.168.117) Quit (Read error: Operation timed out)
[2:35] * yguang11 (~yguang11@vpn-nat.peking.corp.yahoo.com) Quit ()
[2:40] * gary7ny (~jelabarre@c-67-189-183-231.hsd1.ny.comcast.net) Quit (Quit: Leaving)
[2:41] * pentabular (~Adium@2601:9:4980:a01:78c5:c2e0:afdb:3610) Quit (Quit: Leaving.)
[2:42] * Qu310 (~Qu310@ip-121-0-1-110.static.dsl.onqcomms.net) Quit (Remote host closed the connection)
[2:44] * lucas1 (~Thunderbi@222.240.148.130) has joined #ceph
[2:45] * Eco (~Eco@adsl-99-105-53-247.dsl.pltn13.sbcglobal.net) has joined #ceph
[2:49] * Qu310 (~Qu310@ip-121-0-1-110.static.dsl.onqcomms.net) has joined #ceph
[2:51] * yanzheng (~zhyan@134.134.139.72) has joined #ceph
[2:51] * sputnik13 (~sputnik13@wsip-68-105-248-60.sd.sd.cox.net) Quit (Quit: My MacBook has gone to sleep. ZZZzzz???)
[2:52] * yanzheng1 (~zhyan@180.166.196.218) has joined #ceph
[2:56] * angdraug (~angdraug@12.164.168.117) Quit (Quit: Leaving)
[3:00] * lucas1 (~Thunderbi@222.240.148.130) Quit (Read error: Connection reset by peer)
[3:03] * yanzheng1 (~zhyan@180.166.196.218) Quit (Quit: This computer has gone to sleep)
[3:10] * masta (~masta@190.7.213.210) has joined #ceph
[3:10] * masta (~masta@190.7.213.210) has left #ceph
[3:17] * pentabular (~Adium@2601:9:4980:a01:78c5:c2e0:afdb:3610) has joined #ceph
[3:17] * yanzheng1 (~zhyan@180.166.196.218) has joined #ceph
[3:17] * pentabular is now known as Guest1307
[3:19] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) Quit (Quit: Leaving.)
[3:20] * lxo (~aoliva@lxo.user.oftc.net) Quit (Remote host closed the connection)
[3:21] * lxo (~aoliva@lxo.user.oftc.net) has joined #ceph
[3:27] * yanzheng1 (~zhyan@180.166.196.218) Quit (Quit: This computer has gone to sleep)
[3:27] * dneary (~dneary@96.237.180.105) has joined #ceph
[3:35] * Eco (~Eco@adsl-99-105-53-247.dsl.pltn13.sbcglobal.net) Quit (Ping timeout: 480 seconds)
[3:38] * kevinc (~kevinc__@ip174-65-68-134.sd.sd.cox.net) has joined #ceph
[3:39] <bens> dmick: errors in logs
[3:39] <bens> see my open ticket dmick
[3:40] <dmick> number?
[3:46] <bens> 1768
[3:47] <bens> 2014-08-14 15:29:31.878465 7f6eb0bc2700 10 cephx: verify_authorizer decrypted service osd secret_id=3740
[3:47] <bens> 2014-08-14 15:29:31.878496 7f6eb0bc2700 0 auth: could not find secret_id=3740
[3:52] * patcable (sid11336@id-11336.highgate.irccloud.com) Quit ()
[3:54] * zhaochao (~zhaochao@111.204.252.2) has joined #ceph
[3:59] <dmick> oh, those
[3:59] * evanjfraser (~quassel@122.252.188.1) has joined #ceph
[3:59] <evanjfraser> hello
[3:59] <evanjfraser> this looks more like it :)
[4:01] <evanjfraser> so... does anyone here use ceph with btrfs in production?
[4:01] <evanjfraser> I have... a historical dislike to XFS.
[4:01] <evanjfraser> having used it on IRIX it was fine, but it was adventurous on linux at times
[4:02] <evanjfraser> I'm looking to setup some object storage and architecturally ceph looks the best (to my mind anyway)
[4:02] <evanjfraser> but I have a few concerns
[4:03] <evanjfraser> is filesystem fragmentation an issue in production?
[4:03] <evanjfraser> I read somewhere that it could get that way looking at some XFS stats
[4:04] <evanjfraser> from: http://www.sebastien-han.fr/blog/2014/02/17/ceph-io-patterns-the-bad/ actually
[4:08] * kevinc (~kevinc__@ip174-65-68-134.sd.sd.cox.net) Quit (Quit: This computer has gone to sleep)
[4:08] * kevinc (~kevinc__@ip174-65-68-134.sd.sd.cox.net) has joined #ceph
[4:14] * yanzheng1 (~zhyan@180.166.196.218) has joined #ceph
[4:14] * Eco (~Eco@adsl-99-105-53-247.dsl.pltn13.sbcglobal.net) has joined #ceph
[4:26] * aknapp (~aknapp@ip68-99-237-112.ph.ph.cox.net) has joined #ceph
[4:28] * diegows (~diegows@190.190.5.238) Quit (Ping timeout: 480 seconds)
[4:45] * zerick (~eocrospom@190.187.21.53) Quit (Ping timeout: 480 seconds)
[4:55] * yanzheng2 (~zhyan@180.166.196.218) has joined #ceph
[5:02] * yanzheng1 (~zhyan@180.166.196.218) Quit (Ping timeout: 480 seconds)
[5:07] * Vacum_ (~vovo@88.130.215.111) has joined #ceph
[5:08] * sjusthm (~sam@24-205-54-233.dhcp.gldl.ca.charter.com) Quit (Quit: Leaving.)
[5:10] * haomaiwa_ (~haomaiwan@223.223.183.114) Quit (Remote host closed the connection)
[5:10] * zack_dolby (~textual@e0109-114-22-95-89.uqwimax.jp) Quit (Quit: My MacBook has gone to sleep. ZZZzzz???)
[5:10] * haomaiwang (~haomaiwan@124.248.208.2) has joined #ceph
[5:12] * Guest1307 (~Adium@2601:9:4980:a01:78c5:c2e0:afdb:3610) Quit (Quit: Leaving.)
[5:14] * haomaiwa_ (~haomaiwan@223.223.183.114) has joined #ceph
[5:14] * Lorne (Citadel@modemcable158.99-20-96.mc.videotron.ca) Quit (Read error: Connection reset by peer)
[5:14] * Lorne (Citadel@modemcable158.99-20-96.mc.videotron.ca) has joined #ceph
[5:14] * Vacum__ (~vovo@88.130.212.72) Quit (Ping timeout: 480 seconds)
[5:20] * pentabular (~Adium@c-50-148-128-107.hsd1.ca.comcast.net) has joined #ceph
[5:20] * haomaiwang (~haomaiwan@124.248.208.2) Quit (Ping timeout: 480 seconds)
[5:21] * pentabular is now known as Guest1320
[5:31] * haomaiwa_ (~haomaiwan@223.223.183.114) Quit (Remote host closed the connection)
[5:31] * haomaiwang (~haomaiwan@124.248.208.2) has joined #ceph
[5:37] * yanzheng1 (~zhyan@180.166.196.218) has joined #ceph
[5:40] * lupu (~lupu@86.107.101.214) Quit (Ping timeout: 480 seconds)
[5:42] * longguang (~chatzilla@123.126.33.253) Quit (Quit: ChatZilla 0.9.90.1 [Firefox 29.0.1/20140506152807])
[5:42] * adamcrume (~quassel@c-71-204-162-10.hsd1.ca.comcast.net) Quit (Remote host closed the connection)
[5:43] * yanzheng2 (~zhyan@180.166.196.218) Quit (Ping timeout: 480 seconds)
[5:45] * vbellur (~vijay@122.171.116.233) has joined #ceph
[5:45] * yanzheng1 (~zhyan@180.166.196.218) Quit (Quit: This computer has gone to sleep)
[5:51] * xarses (~andreww@c-76-126-112-92.hsd1.ca.comcast.net) has joined #ceph
[5:52] * haomaiwa_ (~haomaiwan@223.223.183.114) has joined #ceph
[5:53] * RameshN (~rnachimu@121.244.87.117) has joined #ceph
[5:58] * kanagaraj (~kanagaraj@117.197.190.62) has joined #ceph
[5:59] * haomaiwang (~haomaiwan@124.248.208.2) Quit (Ping timeout: 480 seconds)
[6:07] * dneary (~dneary@96.237.180.105) Quit (Ping timeout: 480 seconds)
[6:07] * yanzheng1 (~zhyan@180.166.196.218) has joined #ceph
[6:10] * kanagaraj (~kanagaraj@117.197.190.62) Quit (Read error: Connection reset by peer)
[6:10] * kanagaraj (~kanagaraj@117.197.190.62) has joined #ceph
[6:14] * rotbeard (~redbeard@2a02:908:df10:d300:76f0:6dff:fe3b:994d) Quit (Quit: Verlassend)
[6:18] * lalatenduM (~lalatendu@121.244.87.117) has joined #ceph
[6:19] * kevinc (~kevinc__@ip174-65-68-134.sd.sd.cox.net) Quit (Quit: Leaving)
[6:21] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) has joined #ceph
[6:24] * Guest1320 (~Adium@c-50-148-128-107.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[6:29] * aknapp (~aknapp@ip68-99-237-112.ph.ph.cox.net) Quit (Remote host closed the connection)
[6:30] * aknapp (~aknapp@ip68-99-237-112.ph.ph.cox.net) has joined #ceph
[6:32] * zack_dolby (~textual@e0109-114-22-95-89.uqwimax.jp) has joined #ceph
[6:38] * aknapp (~aknapp@ip68-99-237-112.ph.ph.cox.net) Quit (Read error: Operation timed out)
[6:41] * vbellur (~vijay@122.171.116.233) Quit (Quit: Leaving.)
[6:50] * kanagaraj (~kanagaraj@117.197.190.62) Quit (Ping timeout: 480 seconds)
[6:52] * michaelg (~michaelg@sfw-pao475.tango.me) Quit (Ping timeout: 480 seconds)
[6:52] * kanagaraj (~kanagaraj@117.197.190.109) has joined #ceph
[6:58] * kanagaraj (~kanagaraj@117.197.190.109) Quit (Read error: Operation timed out)
[6:59] * romxero (~romxero@c-50-131-25-97.hsd1.ca.comcast.net) has joined #ceph
[6:59] * kanagaraj (~kanagaraj@117.213.103.3) has joined #ceph
[7:01] * aknapp (~aknapp@ip68-99-237-112.ph.ph.cox.net) has joined #ceph
[7:04] * yanzheng1 (~zhyan@180.166.196.218) Quit (Quit: This computer has gone to sleep)
[7:09] * aknapp (~aknapp@ip68-99-237-112.ph.ph.cox.net) Quit (Ping timeout: 480 seconds)
[7:13] <romxero> can I mix 64-bit and 32-bit computers in a ceph cluster?
[7:13] * rdas (~rdas@121.244.87.115) has joined #ceph
[7:13] <romxero> or is that undefined?
[7:15] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) Quit (Quit: Leaving.)
[7:19] * bandrus (~oddo@216.57.72.205) Quit (Remote host closed the connection)
[7:24] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) has joined #ceph
[7:25] * vbellur (~vijay@121.244.87.117) has joined #ceph
[7:26] * Eco (~Eco@adsl-99-105-53-247.dsl.pltn13.sbcglobal.net) Quit (Ping timeout: 480 seconds)
[7:28] * ashishchandra (~ashish@49.32.0.151) has joined #ceph
[7:29] * yanzheng1 (~zhyan@180.166.196.218) has joined #ceph
[7:30] <romxero> anyone up?
[7:32] <evanjfraser> hey romxero, I'm up, but no use to you really as I haven't actually used ceph yet.
[7:32] <evanjfraser> just researching it atm
[7:35] * haomaiwa_ (~haomaiwan@223.223.183.114) Quit (Remote host closed the connection)
[7:35] * haomaiwang (~haomaiwan@124.248.208.2) has joined #ceph
[7:36] <romxero> ah I see
[7:36] <romxero> I'm researching too, I've used gluster in the past for personal research, but I'm putting together a big project for my laboratory utilizing a beowulf cluster
[7:36] <evanjfraser> how was your gluster experience?
[7:37] <evanjfraser> I'm weighing up between the two atm
[7:37] <evanjfraser> but my reads + writes are expected to be very random in behaviour
[7:37] <evanjfraser> and small
[7:37] <evanjfraser> so gluster concerns me on the metadata side
[7:38] <evanjfraser> as far as ceph goes, i'm not so sure why you would mix 32bit and 64bit within the cluster
[7:38] <evanjfraser> clustered filesystems like to be as balanced as possible usually
[7:40] <romxero> true, I was just wondering. I have 64-bit systems but also 32-bit systems, I'm weighing in on using the 32-bits for computational purposes, at the moment they are being used as support machines- they run openbsd for networking
[7:40] <romxero> gluster wasn't bad
[7:41] <romxero> though there was high cpu usage, I think it had to do with spinlocks and asynchronous file descriptors and atomic operations
[7:41] <romxero> I haven't looked into it much, just a hunch, though I think both are needed with the type of application
[7:42] * yanzheng1 (~zhyan@180.166.196.218) Quit (Quit: This computer has gone to sleep)
[7:51] * michaelg (~michaelg@208.87.60.83) has joined #ceph
[7:56] * karnan (~karnan@121.244.87.117) has joined #ceph
[7:58] * karnan (~karnan@121.244.87.117) Quit ()
[7:58] * karnan (~karnan@121.244.87.117) has joined #ceph
[8:00] * Pedras (~Adium@50.185.218.255) Quit (Quit: Leaving.)
[8:01] * yanzheng (~zhyan@134.134.139.72) Quit (Remote host closed the connection)
[8:01] * aknapp (~aknapp@ip68-99-237-112.ph.ph.cox.net) has joined #ceph
[8:02] <evanjfraser> yeah romxero, it looks like ceph might do better in that regard
[8:02] <romxero> I think so too
[8:03] <evanjfraser> but... I haven't even used it, let alone seen where it bottlenecks
[8:03] <evanjfraser> gonna setup a virtual cluster of ceph tomorrow to see how the bits interoperate
[8:04] <romxero> that would be ideal
[8:04] <evanjfraser> I'm also wondering how ceph deals with adding extra disks into the cluster
[8:04] <romxero> hmm
[8:04] <evanjfraser> and i'm still a bit confused with the OSD : disk ratio
[8:04] * bkopilov (~bkopilov@nat-pool-tlv-t.redhat.com) has joined #ceph
[8:05] <evanjfraser> some inktank benchmarks show a much better concurrent thread performance with a larger number of OSD's
[8:05] <evanjfraser> http://ceph.com/community/ceph-performance-part-1-disk-controller-write-throughput/
[8:05] <evanjfraser> but that is within a single node
[8:06] <evanjfraser> so it doesn't really show much of anything in a real world sense
[8:09] * aknapp (~aknapp@ip68-99-237-112.ph.ph.cox.net) Quit (Ping timeout: 480 seconds)
[8:10] * kanagaraj_ (~kanagaraj@117.213.129.176) has joined #ceph
[8:10] * blahnana (~bman@us1.blahnana.com) Quit (Ping timeout: 480 seconds)
[8:11] * kanagaraj_ (~kanagaraj@117.213.129.176) Quit ()
[8:11] <romxero> that is true with distributed systems in general, more nodes usually means better performance depending on the algorithm
[8:11] <Nats> you just want one OSD per disk
[8:11] <romxero> ah
[8:11] <romxero> ok
[8:12] <Nats> no hardware RAID
[8:12] * kanagaraj_ (~kanagaraj@117.213.129.176) has joined #ceph
[8:13] <Nats> get a decent card with JBOD / it mode, run each disk individually
[8:13] <Nats> i'm happy with lsi-9211 performance
[8:15] * yanzheng (~zhyan@jfdmzpr03-ext.jf.intel.com) has joined #ceph
[8:16] * kanagaraj (~kanagaraj@117.213.103.3) Quit (Ping timeout: 480 seconds)
[8:16] * Jakey (uid1475@id-1475.uxbridge.irccloud.com) Quit (Quit: Connection closed for inactivity)
[8:20] * yanzheng1 (~zhyan@180.166.196.218) has joined #ceph
[8:24] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) Quit (Quit: Leaving.)
[8:25] * ade (~abradshaw@193.202.255.218) has joined #ceph
[8:27] * Nacer (~Nacer@2001:41d0:fe82:7200:d94e:dc4d:382:af7c) has joined #ceph
[8:29] <evanjfraser> nice Nats.
[8:29] <evanjfraser> good to hear your experiences
[8:30] * yanzheng1 (~zhyan@180.166.196.218) Quit (Quit: This computer has gone to sleep)
[8:30] <evanjfraser> I'm actually going to be using a netapp as the backend disks... because we have it
[8:31] <evanjfraser> and I want some experience on object store vs filesystem behaviour
[8:31] <evanjfraser> but in theory things can be migrated to separate disk systems live right? :)
[8:31] * kanagaraj (~kanagaraj@101.63.224.25) has joined #ceph
[8:33] * kanagaraj (~kanagaraj@101.63.224.25) Quit ()
[8:35] * kanagaraj (~kanagaraj@117.213.132.235) has joined #ceph
[8:36] * kanagaraj_ (~kanagaraj@117.213.129.176) Quit (Read error: Operation timed out)
[8:37] * mourgaya (~kvirc@80.124.164.139) has joined #ceph
[8:37] * thb (~me@port-50424.pppoe.wtnet.de) has joined #ceph
[8:37] * thb is now known as Guest1333
[8:38] * Guest1333 is now known as thb
[8:39] * kanagaraj_ (~kanagaraj@117.216.98.60) has joined #ceph
[8:39] * kanagaraj_ (~kanagaraj@117.216.98.60) Quit (Read error: Connection reset by peer)
[8:39] * kanagaraj_ (~kanagaraj@117.216.98.60) has joined #ceph
[8:42] <Nats> yes
[8:43] <jerker> I would love to run OSDs on Netapp, but it is a bit like cutting butter with a golden sword. Or something :)
[8:43] <Nats> depending on your cluster size you get some performance reduction when adding new disks, because of all the copying that gets done to redistribute the data
[8:44] * lucas1 (~Thunderbi@222.247.57.50) has joined #ceph
[8:44] <Nats> temporary reduction i mean
[8:45] * kanagaraj__ (~kanagaraj@117.197.188.144) has joined #ceph
[8:45] * kanagaraj (~kanagaraj@117.213.132.235) Quit (Ping timeout: 480 seconds)
[8:47] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) has joined #ceph
[8:49] * kanagaraj_ (~kanagaraj@117.216.98.60) Quit (Read error: Operation timed out)
[8:50] <evanjfraser> yeah I'd expect that Nats
[8:50] * Nacer (~Nacer@2001:41d0:fe82:7200:d94e:dc4d:382:af7c) Quit (Remote host closed the connection)
[8:50] * blahnana (~bman@104-97-248-162-static.reverse.queryfoundry.net) has joined #ceph
[8:54] * gaveen (~gaveen@175.157.99.112) has joined #ceph
[8:55] * kanagaraj__ (~kanagaraj@117.197.188.144) Quit (Ping timeout: 480 seconds)
[8:59] * narurien (~henning@trh.betradar.com) Quit (Ping timeout: 480 seconds)
[9:02] * runfromn1where (~runfromno@pool-108-29-25-203.nycmny.fios.verizon.net) has joined #ceph
[9:02] * aknapp (~aknapp@ip68-99-237-112.ph.ph.cox.net) has joined #ceph
[9:03] * runfromnowhere (~runfromno@pool-108-29-25-203.nycmny.fios.verizon.net) Quit (Ping timeout: 480 seconds)
[9:04] * rotbeard (~redbeard@b2b-94-79-138-170.unitymedia.biz) has joined #ceph
[9:05] * Concubidated (~Adium@66-87-66-197.pools.spcsdns.net) has joined #ceph
[9:07] * hyperbaba (~hyperbaba@private.neobee.net) has joined #ceph
[9:08] * narurien (~henning@trh.betradar.com) has joined #ceph
[9:10] * aknapp (~aknapp@ip68-99-237-112.ph.ph.cox.net) Quit (Ping timeout: 480 seconds)
[9:10] * Concubidated (~Adium@66-87-66-197.pools.spcsdns.net) Quit (Read error: Connection reset by peer)
[9:11] * cok (~chk@2a02:2350:18:1012:acb3:8062:7ce1:4177) has joined #ceph
[9:12] * Concubidated (~Adium@66.87.66.197) has joined #ceph
[9:12] * rendar (~I@host98-9-dynamic.2-79-r.retail.telecomitalia.it) has joined #ceph
[9:13] * kanagaraj (~kanagaraj@117.216.100.49) has joined #ceph
[9:14] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) Quit (Quit: Leaving.)
[9:16] <mourgaya> hi, I don't find how to reset or remove grants on a bucket. how can I do that?
[9:16] <mourgaya> thanks
[9:18] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) has joined #ceph
[9:18] * haomaiwa_ (~haomaiwan@223.223.183.114) has joined #ceph
[9:19] * BManojlovic (~steki@91.195.39.5) has joined #ceph
[9:20] * Concubidated (~Adium@66.87.66.197) Quit (Ping timeout: 480 seconds)
[9:21] * vbellur (~vijay@121.244.87.117) Quit (Ping timeout: 480 seconds)
[9:21] * lcavassa (~lcavassa@2-229-47-79.ip195.fastwebnet.it) has joined #ceph
[9:21] * haomaiwang (~haomaiwan@124.248.208.2) Quit (Ping timeout: 480 seconds)
[9:23] * Concubidated (~Adium@66-87-66-197.pools.spcsdns.net) has joined #ceph
[9:25] * JC (~JC@AMontpellier-651-1-27-181.w90-57.abo.wanadoo.fr) has joined #ceph
[9:28] * andreask (~andreask@h081217017238.dyn.cm.kabsi.at) has joined #ceph
[9:28] * ChanServ sets mode +v andreask
[9:30] * linjan (~linjan@82.102.126.145) has joined #ceph
[9:30] * Lorne (Citadel@modemcable158.99-20-96.mc.videotron.ca) Quit (Read error: Connection reset by peer)
[9:30] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) Quit (Quit: Leaving.)
[9:30] * Lorne (Citadel@modemcable158.99-20-96.mc.videotron.ca) has joined #ceph
[9:35] * djh-work (~daniel@141.52.73.152) Quit (Remote host closed the connection)
[9:36] * djh-work (~daniel@141.52.73.152) has joined #ceph
[9:36] * vbellur (~vijay@209.132.188.8) has joined #ceph
[9:38] * Nacer (~Nacer@252-87-190-213.intermediasud.com) has joined #ceph
[9:39] * garphy`aw is now known as garphy
[9:42] * RameshN (~rnachimu@121.244.87.117) Quit (Ping timeout: 480 seconds)
[9:42] * karnan (~karnan@121.244.87.117) Quit (Ping timeout: 480 seconds)
[9:42] * danieagle (~Daniel@179.184.165.184.static.gvt.net.br) Quit (Ping timeout: 480 seconds)
[9:42] * Concubidated1 (~Adium@66.87.66.197) has joined #ceph
[9:42] * lalatenduM (~lalatendu@121.244.87.117) Quit (Quit: Leaving)
[9:45] * jordanP (~jordan@185.23.92.11) has joined #ceph
[9:45] * i_m (~ivan.miro@nat-5-carp.hcn-strela.ru) has joined #ceph
[9:46] * MACscr (~Adium@c-50-158-183-38.hsd1.il.comcast.net) Quit (Read error: Operation timed out)
[9:47] * Concubidated (~Adium@66-87-66-197.pools.spcsdns.net) Quit (Ping timeout: 480 seconds)
[9:50] * danieagle (~Daniel@179.184.165.184.static.gvt.net.br) has joined #ceph
[9:53] * Concubidated1 (~Adium@66.87.66.197) Quit (Quit: Leaving.)
[9:53] * lalatenduM (~lalatendu@121.244.87.117) has joined #ceph
[9:58] * cookednoodles (~eoin@eoin.clanslots.com) has joined #ceph
[10:03] * aknapp (~aknapp@ip68-99-237-112.ph.ph.cox.net) has joined #ceph
[10:04] * vbellur (~vijay@209.132.188.8) Quit (Ping timeout: 480 seconds)
[10:04] * analbeard (~shw@support.memset.com) has joined #ceph
[10:04] * ashishchandra (~ashish@49.32.0.151) Quit (Ping timeout: 480 seconds)
[10:04] * mourgaya (~kvirc@80.124.164.139) Quit (Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/)
[10:05] * yanzheng1 (~zhyan@180.166.196.218) has joined #ceph
[10:08] * cookednoodles (~eoin@eoin.clanslots.com) Quit (Remote host closed the connection)
[10:08] * cookednoodles (~eoin@eoin.clanslots.com) has joined #ceph
[10:09] * ashishchandra (~ashish@49.32.0.151) has joined #ceph
[10:11] * aknapp (~aknapp@ip68-99-237-112.ph.ph.cox.net) Quit (Ping timeout: 480 seconds)
[10:12] * Midnightmyth (~quassel@93-167-84-102-static.dk.customer.tdc.net) has joined #ceph
[10:13] * cok (~chk@2a02:2350:18:1012:acb3:8062:7ce1:4177) Quit (Quit: Leaving.)
[10:15] * vbellur (~vijay@121.244.87.117) has joined #ceph
[10:17] * romxero (~romxero@c-50-131-25-97.hsd1.ca.comcast.net) Quit (Quit: Leaving)
[10:19] * gurubert (~gurubert@2001:67c:2050:4064:a179:5f14:45d:8c47) has joined #ceph
[10:21] * cookednoodles (~eoin@eoin.clanslots.com) Quit (Remote host closed the connection)
[10:24] * RameshN (~rnachimu@121.244.87.117) has joined #ceph
[10:24] * Nacer (~Nacer@252-87-190-213.intermediasud.com) Quit (Remote host closed the connection)
[10:32] * MACscr (~Adium@c-98-214-170-53.hsd1.il.comcast.net) has joined #ceph
[10:34] * cookednoodles (~eoin@eoin.clanslots.com) has joined #ceph
[10:36] * karnan (~karnan@117.96.0.202) has joined #ceph
[10:38] <kfei> Hi all, how long it would take for an OSD status from 'down' to 'out'?
[10:39] <kfei> The documentation says it's 300 seconds by default
[10:39] <kfei> But I have an OSD in state 'down' for ~1 hour
[10:40] * darkling (~hrm@2001:630:53:b74:227:10ff:fe02:5c8) has joined #ceph
[10:42] <kfei> I mean in the report from `ceph osd tree`
[10:43] * linjan (~linjan@82.102.126.145) Quit (Remote host closed the connection)
[10:46] * Nacer (~Nacer@252-87-190-213.intermediasud.com) has joined #ceph
[10:49] * apolloJess (~Thunderbi@202.60.8.252) has joined #ceph
[10:51] * yanzheng (~zhyan@jfdmzpr03-ext.jf.intel.com) Quit (Remote host closed the connection)
[10:51] * yanzheng2 (~zhyan@180.166.196.218) has joined #ceph
[10:52] <gurubert> have you set 'noout' by any chance for your cluster?
[10:53] <apolloJess> hi all. we are experiencing DMA timeouts in our virtual machines backed by a ceph v0.80.5 cluster. It started when one of the OSDs died. Is this expected?
[10:56] <kfei> gurubert: how can I verify this?
[10:57] <kfei> gurubert: or 'noout' is a default setting?
[10:58] * yanzheng1 (~zhyan@180.166.196.218) Quit (Ping timeout: 480 seconds)
[11:00] * yanzheng2 (~zhyan@180.166.196.218) Quit (Ping timeout: 480 seconds)
[11:01] <gurubert> it should show up in "ceph -s"
[11:01] <gurubert> http://www.sebastien-han.fr/blog/2012/08/17/ceph-storage-node-maintenance/
[11:03] * branto (~borix@ip-213-220-214-245.net.upcbroadband.cz) has joined #ceph
[11:04] * aknapp (~aknapp@ip68-99-237-112.ph.ph.cox.net) has joined #ceph
[11:04] <kfei> there is no 'noout' flag in `ceph -s`
[11:05] <kfei> but can see some remapping and backfilling state info
[11:06] <kfei> seems that OSD has been kicked out by CRUSH map, but the 'down' flag in `ceph osd tree` just not changed
[11:10] * JC1 (~JC@AMontpellier-651-1-27-181.w90-57.abo.wanadoo.fr) has joined #ceph
[11:10] * arcimboldo (~antonio@130.60.144.205) has joined #ceph
[11:12] * aknapp (~aknapp@ip68-99-237-112.ph.ph.cox.net) Quit (Ping timeout: 480 seconds)
[11:12] <kfei> maybe there is no 'out' state in `ceph osd tree`?
[11:13] * shang (~ShangWu@42-68-17-110.EMOME-IP.hinet.net) has joined #ceph
[11:15] * JC (~JC@AMontpellier-651-1-27-181.w90-57.abo.wanadoo.fr) Quit (Ping timeout: 480 seconds)
[11:28] * zack_dolby (~textual@e0109-114-22-95-89.uqwimax.jp) Quit (Quit: My MacBook has gone to sleep. ZZZzzz???)
[11:41] * danieljh (~daniel@0001b4e9.user.oftc.net) Quit (Quit: leaving)
[11:43] * vmx (~vmx@dslb-084-056-040-080.084.056.pools.vodafone-ip.de) has joined #ceph
[11:49] * LeaChim (~LeaChim@host86-159-115-162.range86-159.btcentralplus.com) has joined #ceph
[11:49] * rotbeard (~redbeard@b2b-94-79-138-170.unitymedia.biz) Quit (Read error: Operation timed out)
[11:50] * i_m (~ivan.miro@nat-5-carp.hcn-strela.ru) Quit (Quit: Leaving.)
[11:53] * JC (~JC@AMontpellier-651-1-27-181.w90-57.abo.wanadoo.fr) has joined #ceph
[11:56] * kanagaraj (~kanagaraj@117.216.100.49) Quit (Ping timeout: 480 seconds)
[11:59] * JC1 (~JC@AMontpellier-651-1-27-181.w90-57.abo.wanadoo.fr) Quit (Ping timeout: 480 seconds)
[12:04] * rotbeard (~redbeard@b2b-94-79-138-170.unitymedia.biz) has joined #ceph
[12:04] * shang (~ShangWu@42-68-17-110.EMOME-IP.hinet.net) Quit (Read error: Connection reset by peer)
[12:05] * aknapp (~aknapp@ip68-99-237-112.ph.ph.cox.net) has joined #ceph
[12:07] * kanagaraj (~kanagaraj@117.213.99.225) has joined #ceph
[12:12] * yanzheng (~zhyan@222.64.171.72) has joined #ceph
[12:13] * aknapp (~aknapp@ip68-99-237-112.ph.ph.cox.net) Quit (Ping timeout: 480 seconds)
[12:15] * kanagaraj (~kanagaraj@117.213.99.225) Quit (Quit: Leaving)
[12:15] * kanagaraj (~kanagaraj@117.213.99.225) has joined #ceph
[12:18] * vbellur (~vijay@121.244.87.117) Quit (Read error: Operation timed out)
[12:20] * i_m (~ivan.miro@gbibp9ph1--blueice2n1.emea.ibm.com) has joined #ceph
[12:26] * kalleh (~kalleh@37-46-175-162.customers.ownit.se) has joined #ceph
[12:34] * vbellur (~vijay@209.132.188.8) has joined #ceph
[12:41] * zack_dolby (~textual@p8505b4.tokynt01.ap.so-net.ne.jp) has joined #ceph
[13:06] * aknapp (~aknapp@ip68-99-237-112.ph.ph.cox.net) has joined #ceph
[13:14] * aknapp (~aknapp@ip68-99-237-112.ph.ph.cox.net) Quit (Ping timeout: 480 seconds)
[13:24] * oro (~oro@2001:620:20:16:4910:1ec0:f90a:2816) has joined #ceph
[13:26] * oro_ (~oro@2001:620:20:16:4910:1ec0:f90a:2816) has joined #ceph
[13:34] * lupu (~lupu@86.107.101.214) has joined #ceph
[13:38] * RameshN (~rnachimu@121.244.87.117) Quit (Quit: Quit)
[13:51] * diegows (~diegows@190.190.5.238) has joined #ceph
[13:54] * linuxkidd_ (~linuxkidd@rtp-isp-nat-pool1-1.cisco.com) Quit (Remote host closed the connection)
[13:55] * zhaochao (~zhaochao@111.204.252.2) Quit (Quit: ChatZilla 0.9.90.1 [Iceweasel 31.0/20140803010241])
[13:58] * kanagaraj (~kanagaraj@117.213.99.225) Quit (Ping timeout: 480 seconds)
[13:58] * marrusl (~mark@209-150-43-182.c3-0.wsd-ubr2.qens-wsd.ny.cable.rcn.com) Quit (Remote host closed the connection)
[14:00] * marrusl (~mark@209-150-43-182.c3-0.wsd-ubr2.qens-wsd.ny.cable.rcn.com) has joined #ceph
[14:01] * michaelg (~michaelg@208.87.60.83) Quit (Quit: michaelg)
[14:06] * lofejndif (~lsqavnbok@balo.jager.io) has joined #ceph
[14:06] * aknapp (~aknapp@ip68-99-237-112.ph.ph.cox.net) has joined #ceph
[14:09] * Midnightmyth (~quassel@93-167-84-102-static.dk.customer.tdc.net) Quit (Ping timeout: 480 seconds)
[14:10] * linuxkidd_ (~linuxkidd@cpe-066-057-017-151.nc.res.rr.com) has joined #ceph
[14:15] * aknapp (~aknapp@ip68-99-237-112.ph.ph.cox.net) Quit (Ping timeout: 480 seconds)
[14:15] * aknapp (~aknapp@ip68-99-237-112.ph.ph.cox.net) has joined #ceph
[14:17] * danieljh (~daniel@0001b4e9.user.oftc.net) has joined #ceph
[14:17] * kanagaraj (~kanagaraj@61.3.147.3) has joined #ceph
[14:31] * t0rn (~ssullivan@2607:fad0:32:a02:d227:88ff:fe02:9896) has joined #ceph
[14:31] * aknapp (~aknapp@ip68-99-237-112.ph.ph.cox.net) Quit (Remote host closed the connection)
[14:32] * aknapp (~aknapp@64.202.160.233) has joined #ceph
[14:34] * oms101 (~oms101@p20030057EA09DD00EEF4BBFFFE0F7062.dip0.t-ipconnect.de) Quit (Quit: Leaving)
[14:36] * oms101 (~oms101@p20030057EA09DD00EEF4BBFFFE0F7062.dip0.t-ipconnect.de) has joined #ceph
[14:37] * bkopilov (~bkopilov@nat-pool-tlv-t.redhat.com) Quit (Read error: Operation timed out)
[14:41] * kanagaraj (~kanagaraj@61.3.147.3) Quit (Ping timeout: 480 seconds)
[14:43] * apolloJess (~Thunderbi@202.60.8.252) Quit (Quit: apolloJess)
[14:44] * kanagaraj (~kanagaraj@117.197.194.255) has joined #ceph
[14:47] * linuxkidd (~linuxkidd@cpe-066-057-017-151.nc.res.rr.com) Quit (Quit: Leaving)
[14:49] * KevinPerks (~Adium@2606:a000:80a1:1b00:886a:dda7:30aa:4119) has joined #ceph
[14:51] * linuxkidd_ (~linuxkidd@cpe-066-057-017-151.nc.res.rr.com) Quit (Quit: Leaving)
[14:55] * kanagaraj (~kanagaraj@117.197.194.255) Quit (Read error: Operation timed out)
[14:55] * dneary (~dneary@nat-pool-bos-u.redhat.com) has joined #ceph
[14:56] * lofejndif (~lsqavnbok@1E5AAAOJJ.tor-irc.dnsbl.oftc.net) Quit (Quit: gone)
[14:57] * linuxkidd (~linuxkidd@cpe-066-057-017-151.nc.res.rr.com) has joined #ceph
[15:02] * tupper_ (~tupper@108-83-203-37.lightspeed.rlghnc.sbcglobal.net) has joined #ceph
[15:03] * theanalyst (~abhi@0001c1e3.user.oftc.net) has joined #ceph
[15:04] * theanalyst (~abhi@0001c1e3.user.oftc.net) has left #ceph
[15:04] * xarses (~andreww@c-76-126-112-92.hsd1.ca.comcast.net) Quit (Read error: Operation timed out)
[15:05] * tupper_ (~tupper@108-83-203-37.lightspeed.rlghnc.sbcglobal.net) Quit ()
[15:09] * aknapp (~aknapp@64.202.160.233) Quit (Remote host closed the connection)
[15:10] * aknapp (~aknapp@64.202.160.233) has joined #ceph
[15:14] * xarses (~andreww@c-76-126-112-92.hsd1.ca.comcast.net) has joined #ceph
[15:17] * brad_mssw (~brad@shop.monetra.com) has joined #ceph
[15:18] * dmsimard_away is now known as dmsimard
[15:18] * aknapp (~aknapp@64.202.160.233) Quit (Ping timeout: 480 seconds)
[15:22] * defcon102_kali_linux (~defcon102@94.41.10.240.dynamic.ufanet.ru) has joined #ceph
[15:25] * wschulze (~wschulze@cpe-69-206-251-158.nyc.res.rr.com) has joined #ceph
[15:26] * karnan (~karnan@117.96.0.202) Quit (Quit: Leaving)
[15:27] * theanalyst (theanalyst@0001c1e3.user.oftc.net) has joined #ceph
[15:27] * defcon102_kali_linux (~defcon102@94.41.10.240.dynamic.ufanet.ru) Quit (Remote host closed the connection)
[15:27] * lupu (~lupu@86.107.101.214) Quit (Ping timeout: 480 seconds)
[15:28] * vbellur (~vijay@209.132.188.8) Quit (Quit: Leaving.)
[15:34] * rotbeard (~redbeard@b2b-94-79-138-170.unitymedia.biz) Quit (Quit: Leaving)
[15:35] * DV_ (~veillard@2001:41d0:1:d478::1) has joined #ceph
[15:41] * DV (~veillard@2001:41d0:1:d478::1) Quit (Ping timeout: 480 seconds)
[15:42] * PerlStalker (~PerlStalk@2620:d3:8000:192::70) has joined #ceph
[15:45] * allig8rz (~allig8r@128.135.219.116) has joined #ceph
[15:45] * allig8r (~allig8r@128.135.219.116) Quit (Read error: Connection reset by peer)
[15:47] * analbeard (~shw@support.memset.com) Quit (Remote host closed the connection)
[15:53] * andreask (~andreask@h081217017238.dyn.cm.kabsi.at) has left #ceph
[15:56] * kanagaraj (~kanagaraj@115.241.97.81) has joined #ceph
[16:01] <dmsimard> Is there a way to set an erasure coding profile to an existing, replicated, pool ?
[16:02] <dmsimard> Or is it only possible to create new pools with the EC profile ?
[16:04] * blynch (~blynch@vm-nat.msi.umn.edu) has joined #ceph
[16:04] <beardo> dmsimard, you cannot change an existing pool from replicated to EC. What you would probably want to do is create a new EC pool and use the replicated pool as a cache layer for the EC pool
[16:04] <beardo> http://ceph.com/docs/master/rados/operations/cache-tiering/
[16:06] <dmsimard> beardo: thanks for confirming :)
[16:07] <beardo> dmsimard, no problem
[16:08] * hyperbaba (~hyperbaba@private.neobee.net) Quit (Ping timeout: 480 seconds)
[16:10] * markbby (~Adium@168.94.245.3) has joined #ceph
[16:13] * iggy (~iggy@theiggy.com) Quit (Quit: leaving)
[16:14] * analbeard (~shw@support.memset.com) has joined #ceph
[16:17] * ashishchandra (~ashish@49.32.0.151) Quit (Quit: Leaving)
[16:21] * andreask (~andreask@zid-vpnn018.uibk.ac.at) has joined #ceph
[16:21] * ChanServ sets mode +v andreask
[16:22] * andreask (~andreask@zid-vpnn018.uibk.ac.at) has left #ceph
[16:33] * kanagaraj (~kanagaraj@115.241.97.81) Quit (Quit: Leaving)
[16:34] <nizedk> does anyone know if it will be supported - at some point in time - to dynamically change an EC pool from one profile to another, the same way you can change 'size' and 'min_size'? I foresee that I'd like to change the EC profile as my storage cluster increases in size, in order to get a lower storage overhead. I'd like to go from say 2:2 to 4:2 (or something like that).
[16:42] * joef1 (~Adium@2620:79:0:131:1d7e:ea82:ae83:a22e) has joined #ceph
[16:42] * joef (~Adium@2620:79:0:131:983a:5bde:4455:9f3a) Quit (Read error: Connection reset by peer)
[16:42] * danieljh (~daniel@0001b4e9.user.oftc.net) Quit (Quit: leaving)
[16:47] * Concubidated (~Adium@66-87-66-197.pools.spcsdns.net) has joined #ceph
[16:50] <beardo> nizedk, I would assume no. EC objects are immutable, so if you changed the profile, all the existing objects would need to be rewritten. To get something like this to work, I expect you would need to do something like pull all the EC objects into a replicated cache layer, make the change to the empty pool, and rewrite the objects back to the EC pool. This is effectively the same as creating a new EC, rather than a dynamic change of the existing pool.
[16:58] * rdas (~rdas@121.244.87.115) Quit (Quit: Leaving)
[17:00] * jobewan (~jobewan@snapp.centurylink.net) has joined #ceph
[17:02] <_Tassadar> nizedk: loicd should be able to provide you with a conclusive answer, since he mostly created EC
[17:03] <loicd> to be fair Samuel Just did most of the work ;-) I can answer questions though.
[17:03] * loicd reading
[17:04] * rotbeard (~redbeard@2a02:908:df10:d300:76f0:6dff:fe3b:994d) has joined #ceph
[17:05] <loicd> it is indeed impossible to change the profile for an existing pool
[17:06] <loicd> exporting and reimporting placement groups is a solution and davidz1 is working on this :-)
[17:06] * garphy is now known as garphy`aw
[17:06] * masta (~masta@190.7.213.210) has joined #ceph
[17:13] * garphy`aw is now known as garphy
[17:22] * yanzheng (~zhyan@222.64.171.72) Quit (Quit: This computer has gone to sleep)
[17:22] * yanzheng (~zhyan@222.64.171.72) has joined #ceph
[17:24] <dmsimard> So my use case is not so far fetched then :p
[17:24] <dmsimard> ^
[17:31] * sjm (~Adium@108.53.250.33) has joined #ceph
[17:31] * gurubert (~gurubert@2001:67c:2050:4064:a179:5f14:45d:8c47) Quit (Quit: Leaving)
[17:37] * Teduardo (~Teduardo@57.0.be.static.xlhost.com) Quit ()
[17:40] * steveeJ (~junky@HSI-KBW-085-216-022-246.hsi.kabelbw.de) has joined #ceph
[17:40] * hasues (~hasues@kwfw01.scrippsnetworksinteractive.com) has joined #ceph
[17:41] * yanzheng (~zhyan@222.64.171.72) Quit (Quit: This computer has gone to sleep)
[17:42] * yanzheng (~zhyan@222.64.171.72) has joined #ceph
[17:46] * aknapp (~aknapp@fw125-01-outside-active.ent.mgmt.glbt1.secureserver.net) has joined #ceph
[17:47] * yanzheng (~zhyan@222.64.171.72) Quit (Quit: This computer has gone to sleep)
[17:47] * yanzheng (~zhyan@222.64.171.72) has joined #ceph
[17:48] * ade (~abradshaw@193.202.255.218) Quit (Ping timeout: 480 seconds)
[17:48] <kitz> has anyone written up a "Scrubbing Schedules Explained" article or something like that?
[17:49] * adamcrume (~quassel@50.247.81.99) has joined #ceph
[17:50] * i_m (~ivan.miro@gbibp9ph1--blueice2n1.emea.ibm.com) Quit (Quit: Leaving.)
[17:51] * bandrus (~oddo@216.57.72.205) has joined #ceph
[17:53] * sputnik13 (~sputnik13@207.8.121.241) has joined #ceph
[17:53] * kalleh (~kalleh@37-46-175-162.customers.ownit.se) Quit (Ping timeout: 480 seconds)
[17:57] * vbellur (~vijay@122.172.26.19) has joined #ceph
[17:58] * analbeard (~shw@support.memset.com) Quit (Quit: Leaving.)
[17:59] * ikrstic (~ikrstic@93-87-197-83.dynamic.isp.telekom.rs) has joined #ceph
[18:00] * garphy is now known as garphy`aw
[18:04] * rmoe (~quassel@173-228-89-134.dsl.static.sonic.net) Quit (Ping timeout: 480 seconds)
[18:05] * baylight (~tbayly@204.15.85.169) has joined #ceph
[18:12] * rmoe (~quassel@12.164.168.117) has joined #ceph
[18:13] * darkling (~hrm@00012bd0.user.oftc.net) Quit (Ping timeout: 480 seconds)
[18:13] * rweeks (~rweeks@pat.hitachigst.com) has joined #ceph
[18:14] * garphy`aw is now known as garphy
[18:21] * yanzheng (~zhyan@222.64.171.72) Quit (Quit: This computer has gone to sleep)
[18:21] <bens> kitz: I nominate you
[18:21] * BManojlovic (~steki@91.195.39.5) Quit (Ping timeout: 480 seconds)
[18:21] * i_m (~ivan.miro@nat-5-carp.hcn-strela.ru) has joined #ceph
[18:23] * gregsfortytwo1 (~Adium@cpe-107-184-64-126.socal.res.rr.com) has joined #ceph
[18:25] * ade (~abradshaw@80-72-52-55.cmts.powersurf.li) has joined #ceph
[18:25] * Lorne (Citadel@modemcable158.99-20-96.mc.videotron.ca) Quit (Read error: Connection reset by peer)
[18:25] * Lorne (Citadel@modemcable158.99-20-96.mc.videotron.ca) has joined #ceph
[18:29] * arcimboldo (~antonio@130.60.144.205) Quit (Quit: Ex-Chat)
[18:31] * Midnightmyth (~quassel@93-167-84-102-static.dk.customer.tdc.net) has joined #ceph
[18:31] * markbby (~Adium@168.94.245.3) Quit (Quit: Leaving.)
[18:33] * jordanP (~jordan@185.23.92.11) Quit (Quit: Leaving)
[18:35] * branto (~borix@ip-213-220-214-245.net.upcbroadband.cz) has left #ceph
[18:36] * tws_ (~traviss@rrcs-24-123-86-154.central.biz.rr.com) Quit (Ping timeout: 480 seconds)
[18:38] * lalatenduM (~lalatendu@121.244.87.117) Quit (Quit: Leaving)
[18:41] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) has joined #ceph
[18:52] * zerick (~eocrospom@190.187.21.53) has joined #ceph
[18:53] * oro_ (~oro@2001:620:20:16:4910:1ec0:f90a:2816) Quit (Ping timeout: 480 seconds)
[18:53] * oro (~oro@2001:620:20:16:4910:1ec0:f90a:2816) Quit (Ping timeout: 480 seconds)
[18:54] * Nacer (~Nacer@252-87-190-213.intermediasud.com) Quit (Read error: Operation timed out)
[18:58] * KevinPerks1 (~Adium@2606:a000:80a1:1b00:164:b27:57d4:d66e) has joined #ceph
[19:00] * garphy is now known as garphy`aw
[19:01] * vbellur (~vijay@122.172.26.19) Quit (Quit: Leaving.)
[19:01] <nizedk> loicd thank you for the answer :-)
[19:03] <loicd> nizedk: yw :-)
[19:03] * adamcrume (~quassel@50.247.81.99) Quit (Remote host closed the connection)
[19:03] * KevinPerks (~Adium@2606:a000:80a1:1b00:886a:dda7:30aa:4119) Quit (Ping timeout: 480 seconds)
[19:03] * smiley_ (~smiley@pool-173-66-4-176.washdc.fios.verizon.net) has joined #ceph
[19:06] <smiley_> Hello all???I upgraded from emperor to firefly this afternoon???things look good at this point???however I see 'health HEALTH_WARN crush map has legacy tunables' in my ceph -s status???and I am wondering???what if anything I should do about it
[19:07] * ade (~abradshaw@80-72-52-55.cmts.powersurf.li) Quit (Quit: Too sexy for his shirt)
[19:07] <smiley_> is there any good reason to leave it as is???or should I update the tunables?
[19:08] * gaveen (~gaveen@175.157.99.112) Quit (Ping timeout: 480 seconds)
[19:08] * Sysadmin88 (~IceChat77@054533bc.skybroadband.com) Quit (Quit: On the other hand, you have different fingers.)
[19:13] * romxero (~romxero@c-50-131-25-97.hsd1.ca.comcast.net) has joined #ceph
[19:14] <romxero> ok I hope everyone is up, I'm going to leave an open question; can I mix and match 32-bit and 64-bit servers in a ceph cluster?
[19:15] * angdraug (~angdraug@12.164.168.117) has joined #ceph
[19:22] * xarses (~andreww@c-76-126-112-92.hsd1.ca.comcast.net) Quit (Ping timeout: 480 seconds)
[19:25] <gregsfortytwo1> you should be able to mix 32 and 64 bit servers, yes
[19:31] <steveeJ> what happens on write-back if i have an OSD in the cache pool and in the backend pool?
[19:34] * michaelg (~michaelg@sfw-pao475.tango.me) has joined #ceph
[19:35] * lcavassa (~lcavassa@2-229-47-79.ip195.fastwebnet.it) Quit (Quit: Leaving)
[19:37] * markbby (~Adium@168.94.245.1) has joined #ceph
[19:37] * markbby (~Adium@168.94.245.1) Quit ()
[19:40] * markbby (~Adium@168.94.245.4) has joined #ceph
[19:44] * iggy_ (~iggy@70.52.236.23.bc.googleusercontent.com) has joined #ceph
[19:46] * iggy_ (~iggy@70.52.236.23.bc.googleusercontent.com) Quit (Remote host closed the connection)
[19:46] * iggy_ (~iggy@70.52.236.23.bc.googleusercontent.com) has joined #ceph
[19:47] * iggy_ (~iggy@70.52.236.23.bc.googleusercontent.com) Quit (Remote host closed the connection)
[19:48] * iggy_ (~iggy@70.52.236.23.bc.googleusercontent.com) has joined #ceph
[19:49] * iggy_ (~iggy@70.52.236.23.bc.googleusercontent.com) Quit (Remote host closed the connection)
[19:49] * Pedras (~Adium@216.207.42.129) has joined #ceph
[19:49] * iggy_ (~iggy@70.52.236.23.bc.googleusercontent.com) has joined #ceph
[19:50] * iggy_ is now known as iggy
[19:51] * iggy (~iggy@70.52.236.23.bc.googleusercontent.com) Quit (Remote host closed the connection)
[19:52] * reed (~reed@75-101-54-131.dsl.static.sonic.net) has joined #ceph
[19:55] * Sysadmin88 (~IceChat77@054533bc.skybroadband.com) has joined #ceph
[19:59] * capri (~capri@212.218.127.222) has joined #ceph
[19:59] * iggy (~iggy@70.52.236.23.bc.googleusercontent.com) has joined #ceph
[20:00] * xarses (~andreww@12.164.168.117) has joined #ceph
[20:01] * iggy (~iggy@70.52.236.23.bc.googleusercontent.com) Quit (Remote host closed the connection)
[20:02] * aknapp_ (~aknapp@fw125-01-outside-active.ent.mgmt.glbt1.secureserver.net) has joined #ceph
[20:06] * aknapp_ (~aknapp@fw125-01-outside-active.ent.mgmt.glbt1.secureserver.net) Quit (Read error: Connection reset by peer)
[20:06] * aknapp_ (~aknapp@fw125-01-outside-active.ent.mgmt.glbt1.secureserver.net) has joined #ceph
[20:09] * aknapp (~aknapp@fw125-01-outside-active.ent.mgmt.glbt1.secureserver.net) Quit (Ping timeout: 480 seconds)
[20:12] * Nacer (~Nacer@203-206-190-109.dsl.ovh.fr) has joined #ceph
[20:12] <sironside> hi all . . . does anyone know if it's possible to force unmap an rbd block device?
[20:13] <sironside> I'm not sure why it thinks the device is busy - it isn't
[20:13] <sironside> at a total guess, the only culprit I can think of is udev if it's what noticed the partitions on the block device and created /dev/rbd0p1 and /dev/rbd0p2
[20:14] <sironside> (this is firefly 0.80.5 on RHEL7
[20:16] * bkopilov (~bkopilov@213.57.65.180) has joined #ceph
[20:16] * adamcrume (~quassel@eduroam-242-18.ucsc.edu) has joined #ceph
[20:20] <romxero> awesome, I'm going to mix and match servers then
[20:25] * mnc (~mnc@c-24-245-27-162.hsd1.mn.comcast.net) has joined #ceph
[20:26] * mnc (~mnc@c-24-245-27-162.hsd1.mn.comcast.net) Quit ()
[20:32] * Sysadmin88 (~IceChat77@054533bc.skybroadband.com) Quit (Quit: Say What?)
[20:33] <steveeJ> sironside: i've had that problem when i used a btrfs on an rbd device. even after unmounting the block device was in use
[20:33] <sironside> were you able to get it unmounted eventually somehow?
[20:33] * jobewan (~jobewan@snapp.centurylink.net) Quit (Quit: Leaving)
[20:33] <sironside> or did you reboot it?
[20:33] * jobewan (~jobewan@snapp.centurylink.net) has joined #ceph
[20:33] * andreask (~andreask@h081217017238.dyn.cm.kabsi.at) has joined #ceph
[20:33] * ChanServ sets mode +v andreask
[20:36] <seapasul1i> probably a bad idea but have you tried forcefully removing the kernel mod?
[20:37] <steveeJ> sironside: it was already unmounted, but i couldn't be unmapped
[20:37] <sironside> I have not, I'll give that a try
[20:37] * aknapp_ (~aknapp@fw125-01-outside-active.ent.mgmt.glbt1.secureserver.net) Quit (Remote host closed the connection)
[20:37] <sironside> steveeJ: same issue here
[20:37] <steveeJ> seapasul1i: that is a very bad idea if you have multiple mapped images :D
[20:37] <sironside> it's a KVM virtual machine in this case
[20:37] * aknapp (~aknapp@fw125-01-outside-active.ent.mgmt.glbt1.secureserver.net) has joined #ceph
[20:37] <sironside> I only use the RBD device initially
[20:38] <sironside> then I let qemu/KVM talk directly to ceph
[20:38] <sironside> so actually I could try force removing the module
[20:38] <seapasul1i> steveeJ: yeah assuming only one mapped image. Probably a bad assumption
[20:39] * aknapp (~aknapp@fw125-01-outside-active.ent.mgmt.glbt1.secureserver.net) Quit (Read error: Operation timed out)
[20:39] <sironside> I'll try it in a minute, just dd'ing that device first
[20:42] * aknapp (~aknapp@fw125-01-outside-active.ent.mgmt.glbt1.secureserver.net) has joined #ceph
[20:44] * iggy (~iggy@70.52.236.23.bc.googleusercontent.com) has joined #ceph
[20:44] * iggy (~iggy@70.52.236.23.bc.googleusercontent.com) Quit (Remote host closed the connection)
[20:44] * iggy (~iggy@70.52.236.23.bc.googleusercontent.com) has joined #ceph
[20:44] * iggy (~iggy@70.52.236.23.bc.googleusercontent.com) Quit (Remote host closed the connection)
[20:45] * iggy (~iggy@70.52.236.23.bc.googleusercontent.com) has joined #ceph
[20:45] * iggy (~iggy@70.52.236.23.bc.googleusercontent.com) Quit (Remote host closed the connection)
[20:46] * iggy (~iggy@70.52.236.23.bc.googleusercontent.com) has joined #ceph
[20:46] * iggy (~iggy@70.52.236.23.bc.googleusercontent.com) Quit (Remote host closed the connection)
[20:46] * iggy (~iggy@70.52.236.23.bc.googleusercontent.com) has joined #ceph
[20:47] * andreask (~andreask@h081217017238.dyn.cm.kabsi.at) has left #ceph
[20:57] * Venturi (~Venturi@93-103-91-169.dynamic.t-2.net) has joined #ceph
[20:58] <sironside> modprobe -rf rbd
[20:59] <sironside> no luck :( "modprobe: FATAL: Module rbd is in use."
[20:59] * adamcrume (~quassel@eduroam-242-18.ucsc.edu) Quit (Read error: Operation timed out)
[21:00] <sironside> didn't do any harm though
[21:02] * lxo (~aoliva@lxo.user.oftc.net) Quit (Remote host closed the connection)
[21:06] * iggy (~iggy@70.52.236.23.bc.googleusercontent.com) Quit (Remote host closed the connection)
[21:07] * lxo (~aoliva@lxo.user.oftc.net) has joined #ceph
[21:08] * iggy (~iggy@70.52.236.23.bc.googleusercontent.com) has joined #ceph
[21:09] * Nacer (~Nacer@203-206-190-109.dsl.ovh.fr) Quit (Remote host closed the connection)
[21:09] * iggy (~iggy@70.52.236.23.bc.googleusercontent.com) Quit (Remote host closed the connection)
[21:11] * thb (~me@0001bd58.user.oftc.net) Quit (Ping timeout: 480 seconds)
[21:14] * rendar (~I@host98-9-dynamic.2-79-r.retail.telecomitalia.it) Quit (Ping timeout: 480 seconds)
[21:14] * garphy`aw is now known as garphy
[21:16] * Nacer (~Nacer@2001:41d0:fe82:7200:7a31:c1ff:febd:ec2e) has joined #ceph
[21:16] * rendar (~I@host98-9-dynamic.2-79-r.retail.telecomitalia.it) has joined #ceph
[21:18] * iggy (~iggy@70.52.236.23.bc.googleusercontent.com) has joined #ceph
[21:20] * iggy (~iggy@70.52.236.23.bc.googleusercontent.com) Quit (Remote host closed the connection)
[21:21] * iggy (~iggy@70.52.236.23.bc.googleusercontent.com) has joined #ceph
[21:24] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) Quit (Read error: Connection reset by peer)
[21:24] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) has joined #ceph
[21:24] * iggy (~iggy@70.52.236.23.bc.googleusercontent.com) Quit (Remote host closed the connection)
[21:25] * iggy (~iggy@70.52.236.23.bc.googleusercontent.com) has joined #ceph
[21:29] * xarses (~andreww@12.164.168.117) Quit (Ping timeout: 480 seconds)
[21:30] * iggy (~iggy@70.52.236.23.bc.googleusercontent.com) Quit (Remote host closed the connection)
[21:30] <steveeJ> sironside: you could try to remove the device using "dmsetup remove -f .."
[21:32] * jobewan (~jobewan@snapp.centurylink.net) Quit (Quit: Leaving)
[21:32] * codice_ (~toodles@97-94-175-73.static.mtpk.ca.charter.com) has joined #ceph
[21:34] * codice (~toodles@97-94-175-73.static.mtpk.ca.charter.com) Quit (Ping timeout: 480 seconds)
[21:41] * xarses (~andreww@12.164.168.117) has joined #ceph
[21:41] * xarses (~andreww@12.164.168.117) Quit (Remote host closed the connection)
[21:44] * pentabular (~Adium@c-50-184-59-146.hsd1.ca.comcast.net) has joined #ceph
[21:44] * Nacer (~Nacer@2001:41d0:fe82:7200:7a31:c1ff:febd:ec2e) Quit (Remote host closed the connection)
[21:46] <sironside> dmsetup remove -f /dev/rbd0 ?
[21:47] <sironside> "Device rbd0 not found" (three times)
[21:47] <sironside> Command failed
[21:48] <steveeJ> looks like the dm doesn't have anything to do with the rbd. my bad
[21:52] <sironside> no worries, did no harm either
[21:53] <steveeJ> what happens if you try to unmap the device?
[21:53] <steveeJ> using regular "rbd unmap"
[21:56] * blSnoopy (~snoopy@miram.persei.mw.lg.virgo.supercluster.net) Quit (Remote host closed the connection)
[22:01] <Venturi> How does CEPH solve problem on rack power failure. Let's say that we do 3-way replica, 2 copys of data is situated somewhere in the first rack and the third copy is somewhere in the second rack. Power of the second rack suddenly goes down. Does it immediately, when it founds out that OSD is no longer present, start replicating data within first rack to assure 3-copies of data? In case of the
[22:01] <Venturi> second rack does not come back for some time, does that mean that the amount of data (all third copies of data) stored on the second rack could potentional be created on the first rack?
[22:03] <steveeJ> Venturi: yes, if you lose OSDs the remaining OSDs will need to compensate for the loss
[22:04] * sputnik13 (~sputnik13@207.8.121.241) Quit (Quit: My MacBook has gone to sleep. ZZZzzz???)
[22:04] * sputnik13 (~sputnik13@207.8.121.241) has joined #ceph
[22:05] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) Quit (Read error: Connection reset by peer)
[22:05] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) has joined #ceph
[22:11] <Venturi> so it is smarter to do 3-replicas within the same rack and than mirroring data to the second rack?
[22:11] <Venturi> when looking from the perspective of power-loss situation
[22:13] * jtang_ (~jtang@80.111.83.231) Quit (Remote host closed the connection)
[22:13] * jtang_ (~jtang@80.111.83.231) has joined #ceph
[22:17] * sjm (~Adium@108.53.250.33) has left #ceph
[22:20] <steveeJ> Venturi: i don't see more sense in that than letting ceph do it's job in case of a power failure
[22:20] * rturk|afk is now known as rturk
[22:21] * markbby (~Adium@168.94.245.4) Quit (Quit: Leaving.)
[22:23] * Nacer (~Nacer@203-206-190-109.dsl.ovh.fr) has joined #ceph
[22:24] * BManojlovic (~steki@95.180.4.243) has joined #ceph
[22:25] <Venturi> steveeJ: yes, i understand. I am just thinking about possible scenarios, also what could happen if you had to replicate all data to the first rack...in case large storage rack, there could be a lot of data activity to the first rack going on. Also, what could happen if you can not recreate the third copy of data to the first rack due to lack of space etc.
[22:27] * andreask (~andreask@h081217017238.dyn.cm.kabsi.at) has joined #ceph
[22:27] * ChanServ sets mode +v andreask
[22:29] * andreask (~andreask@h081217017238.dyn.cm.kabsi.at) has left #ceph
[22:29] <sironside> steveeJ: sorry, didn't realise you were talking. Normal rbd unmap says the device is busy
[22:29] <sironside> I can verify with listwatchers that there are two clients holding it open
[22:29] <sironside> both have the IP address of the same client server
[22:29] <steveeJ> Venturi: the activity in case of losing OSDs is something that bothers me too. there are ways to reduce the performance impact on backfilling
[22:29] * kfei (~root@114-27-82-57.dynamic.hinet.net) Quit (Ping timeout: 480 seconds)
[22:29] <sironside> one I can explain as the qemu connection
[22:30] <sironside> the other shouldn't be there
[22:36] <steveeJ> Venturi: one other idea i have is set a generous replica limit and adjust the timeouts to allow very long downtimes (http://ceph.com/docs/master/rados/configuration/mon-osd-interaction/) before anything will be copied
[22:41] * adamcrume (~quassel@c-71-204-162-10.hsd1.ca.comcast.net) has joined #ceph
[22:42] * kfei (~root@114-27-53-218.dynamic.hinet.net) has joined #ceph
[22:42] <Venturi> steveeJ: thx for the link. But what happens than with consistency model on which CEPH is based on. In case of second rack down, could the existing data be updated, although the third copy could not be?
[22:43] * alop (~abelopez@72-163-0-129.cisco.com) has joined #ceph
[22:44] * JC (~JC@AMontpellier-651-1-27-181.w90-57.abo.wanadoo.fr) Quit (Quit: Leaving.)
[22:44] <Venturi> I guess, for CEPH consistency model, it is beset that monitors found out about the problem as soon as possible?
[22:44] <steveeJ> Venturi: it could. when the OSDs come back online they will receive the changes
[22:44] * KevinPerks (~Adium@2606:a000:80a1:1b00:de0:9f99:4018:19e6) has joined #ceph
[22:46] <steveeJ> Venturi: if writes will be possible to the degraded PGs depends on the min_size you can set for the pools. if the available OSDs >= min_size, writes will still be acknowledged
[22:47] * BManojlovic (~steki@95.180.4.243) Quit (Ping timeout: 480 seconds)
[22:47] <Venturi> steveeJ: aha ok. thx. so basicly this kind of possible problem is solved by good configuration.
[22:48] <steveeJ> yeah. ceph is very flexible
[22:49] * KevinPerks1 (~Adium@2606:a000:80a1:1b00:164:b27:57d4:d66e) Quit (Ping timeout: 480 seconds)
[22:50] <Venturi> One more general question. Does there exist some web-managment GUI for CEPH or is all CLI based? I know that Calamari is now open-sourced, but it is just for monitoring purposes, right?
[22:51] <Venturi> I men web-management GUI for conifguring clusters
[22:51] <dmick> Calamari supports some management functionality
[22:51] <dmick> osd mark in/out, pool create/delete, etc.
[22:51] <dmick> not everything
[22:52] <Venturi> aha ok. thx
[22:56] * BManojlovic (~steki@95.180.4.243) has joined #ceph
[23:00] * rturk is now known as rturk|afk
[23:10] * danieljh (~daniel@0001b4e9.user.oftc.net) has joined #ceph
[23:10] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) Quit (Read error: Connection reset by peer)
[23:11] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) has joined #ceph
[23:12] <Venturi> Is erasure coding within Firefly based on SW computation or does it already know how to use some INTEL SIMD instructions for computation on the HW level?
[23:14] * ikrstic (~ikrstic@93-87-197-83.dynamic.isp.telekom.rs) Quit (Quit: Konversation terminated!)
[23:16] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) Quit (Quit: Leaving.)
[23:17] * Tamil (~Adium@cpe-108-184-74-11.socal.res.rr.com) has joined #ceph
[23:23] * rweeks (~rweeks@pat.hitachigst.com) Quit (Quit: Leaving)
[23:24] * markbby (~Adium@168.94.245.4) has joined #ceph
[23:29] <smiley_> I upgraded to ceph-deploy 1.5.11 today before trying to add a couple of nodes???and now???it seems like 'osd create' works???.the osd is not added/mounted/etc.
[23:29] <smiley_> here is the link to the zap and the create: http://pastebin.com/kVakkGqc
[23:30] <smiley_> this is all on ubuntu 12.04.5
[23:31] <smiley_> any suggestions would be appreciated.
[23:31] * t0rn (~ssullivan@2607:fad0:32:a02:d227:88ff:fe02:9896) Quit (Quit: Leaving.)
[23:32] * markbby (~Adium@168.94.245.4) Quit (Quit: Leaving.)
[23:35] * sz0 (~sz0@94.55.197.185) has joined #ceph
[23:36] * markbby (~Adium@168.94.245.3) has joined #ceph
[23:43] * markbby (~Adium@168.94.245.3) Quit (Quit: Leaving.)
[23:45] * jtang_ (~jtang@80.111.83.231) Quit (Remote host closed the connection)
[23:46] * Venturi (~Venturi@93-103-91-169.dynamic.t-2.net) Quit ()
[23:46] * brad_mssw (~brad@shop.monetra.com) Quit (Quit: Leaving)
[23:52] * sputnik13 (~sputnik13@207.8.121.241) Quit (Quit: My MacBook has gone to sleep. ZZZzzz???)
[23:54] * Concubidated1 (~Adium@66-87-66-197.pools.spcsdns.net) has joined #ceph
[23:54] * Concubidated (~Adium@66-87-66-197.pools.spcsdns.net) Quit (Read error: Connection reset by peer)
[23:56] * Concubidated1 (~Adium@66-87-66-197.pools.spcsdns.net) Quit (Read error: Connection reset by peer)
[23:56] * dneary (~dneary@nat-pool-bos-u.redhat.com) Quit (Ping timeout: 480 seconds)
[23:57] * jtang_ (~jtang@80.111.83.231) has joined #ceph
[23:59] * sputnik13 (~sputnik13@207.8.121.241) has joined #ceph

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