#ceph IRC Log


IRC Log for 2011-12-05

Timestamps are in GMT/BST.

[0:18] * aa (~aa@r190-135-4-48.dialup.adsl.anteldata.net.uy) Quit (Quit: Konversation terminated!)
[0:18] * andresambrois (~aa@r190-135-4-48.dialup.adsl.anteldata.net.uy) has joined #ceph
[0:31] * Bircoph (~Bircoph@nat0.campus.mephi.ru) Quit (Remote host closed the connection)
[0:32] * Bircoph (~Bircoph@nat0.campus.mephi.ru) has joined #ceph
[0:48] * Volley (~worf@chello080109200187.3.sc-graz.chello.at) Quit (Quit: Leaving)
[1:18] * andresambrois (~aa@r190-135-4-48.dialup.adsl.anteldata.net.uy) Quit (Quit: Konversation terminated!)
[1:19] * andresambrois (~aa@r190-135-4-48.dialup.adsl.anteldata.net.uy) has joined #ceph
[1:39] * andresambrois (~aa@r190-135-4-48.dialup.adsl.anteldata.net.uy) Quit (Quit: Konversation terminated!)
[1:39] * andresambrois (~aa@r190-135-4-48.dialup.adsl.anteldata.net.uy) has joined #ceph
[1:53] * tsuzuki (~tsuzuki@p9224-ipngn1601marunouchi.tokyo.ocn.ne.jp) has joined #ceph
[1:58] * andresambrois (~aa@r190-135-4-48.dialup.adsl.anteldata.net.uy) Quit (Ping timeout: 480 seconds)
[5:36] * bchrisman (~Adium@c-76-103-130-94.hsd1.ca.comcast.net) has joined #ceph
[7:45] * bchrisman (~Adium@c-76-103-130-94.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[7:47] * MarkDude (~MT@c-71-198-138-155.hsd1.ca.comcast.net) has joined #ceph
[8:16] * jantje (~jan@paranoid.nl) Quit (Ping timeout: 480 seconds)
[8:21] * jantje (~jan@paranoid.nl) has joined #ceph
[8:43] * gregorg (~Greg@ has joined #ceph
[10:08] * MarkDude (~MT@c-71-198-138-155.hsd1.ca.comcast.net) Quit (Read error: Connection reset by peer)
[12:38] * fronlius (~fronlius@g231173033.adsl.alicedsl.de) has joined #ceph
[12:44] * fronlius (~fronlius@g231173033.adsl.alicedsl.de) Quit (Quit: fronlius)
[12:57] * andresambrois (~aa@r190-135-43-161.dialup.adsl.anteldata.net.uy) has joined #ceph
[12:58] * fronlius (~fronlius@g231173033.adsl.alicedsl.de) has joined #ceph
[12:59] * tsuzuki (~tsuzuki@p9224-ipngn1601marunouchi.tokyo.ocn.ne.jp) Quit (Remote host closed the connection)
[13:56] * RupS| (~rups@panoramix.m0z.net) Quit (Read error: Connection reset by peer)
[13:59] * RupS (~rups@panoramix.m0z.net) has joined #ceph
[13:59] * fronlius (~fronlius@g231173033.adsl.alicedsl.de) Quit (Quit: fronlius)
[14:47] * morse_ (~morse@supercomputing.univpm.it) has joined #ceph
[14:49] * Bircoph (~Bircoph@nat0.campus.mephi.ru) Quit (Quit: leaving)
[14:50] * morse (~morse@supercomputing.univpm.it) Quit (Ping timeout: 480 seconds)
[14:55] * morse_ (~morse@supercomputing.univpm.it) Quit (Ping timeout: 480 seconds)
[15:00] * morse (~morse@supercomputing.univpm.it) has joined #ceph
[15:20] * morse (~morse@supercomputing.univpm.it) Quit (Ping timeout: 480 seconds)
[15:20] * morse_ (~morse@supercomputing.univpm.it) has joined #ceph
[15:51] * andresambrois (~aa@r190-135-43-161.dialup.adsl.anteldata.net.uy) Quit (Remote host closed the connection)
[16:06] * lxo (~aoliva@lxo.user.oftc.net) Quit (Quit: later)
[16:21] * bchrisman (~Adium@c-76-103-130-94.hsd1.ca.comcast.net) has joined #ceph
[17:00] * adjohn (~adjohn@70-36-139-247.dsl.dynamic.sonic.net) has joined #ceph
[17:18] * adjohn is now known as Guest19407
[17:18] * _adjohn (~adjohn@70-36-139-247.dsl.dynamic.sonic.net) has joined #ceph
[17:18] * _adjohn is now known as adjohn
[17:18] * Guest19407 (~adjohn@70-36-139-247.dsl.dynamic.sonic.net) Quit (Quit: Guest19407)
[17:18] * aa (~aa@r200-40-114-26.ae-static.anteldata.net.uy) has joined #ceph
[17:21] * bchrisman (~Adium@c-76-103-130-94.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[17:57] * fronlius (~fronlius@e176041138.adsl.alicedsl.de) has joined #ceph
[18:08] <sagewk> http://octo-pus.de/
[18:09] <grape> sagewk: that's slick
[18:11] <sagewk> i miss thedraw
[18:17] * adjohn (~adjohn@70-36-139-247.dsl.dynamic.sonic.net) Quit (Quit: adjohn)
[18:18] * Tv (~Tv|work@aon.hq.newdream.net) has joined #ceph
[18:37] * NightDog (~karl@52.84-48-58.nextgentel.com) has joined #ceph
[18:39] * bchrisman (~Adium@ has joined #ceph
[18:46] * joshd (~joshd@aon.hq.newdream.net) has joined #ceph
[18:59] * buck (~buck@bender.soe.ucsc.edu) Quit (Quit: Leaving)
[18:59] * buck (~buck@bender.soe.ucsc.edu) has joined #ceph
[19:04] * fronlius (~fronlius@e176041138.adsl.alicedsl.de) Quit (Quit: fronlius)
[19:05] * cp (~cp@76-220-17-197.lightspeed.sntcca.sbcglobal.net) has joined #ceph
[19:06] * lxo (~aoliva@lxo.user.oftc.net) has joined #ceph
[19:08] <chaos_> sagewk, have you seen my questions about adding mons from sunday?
[19:08] <sagewk> chaos_: nope..
[19:08] <sagewk> on ML or irc?
[19:08] <chaos_> irc
[19:08] <chaos_> sec
[19:08] <chaos_> I'll paste this
[19:08] <lxo> hmm, ceph -w (0.39) becomes quiet forever after a while reporting cluster information. known problem?
[19:08] <chaos_> 11:55 [ chaos_] sagewk, in 0.39 rls notes i see something like thos "mon: vastly simplified mon cluster expansion", How it has changed? Now when
[19:08] <chaos_> I'm trying to add new mon "ceph mon add" enters some loop, earlier it was easier, process exited with proper code and I could
[19:08] <chaos_> proceed and I still have to copy mon snapshot to new mon.
[19:08] <chaos_> 11:56 [ chaos_] Furthermore, if u add new mon to config before running 'ceph mon add', 'ceph mon add' crashes with some assertion failed error.
[19:08] <sagewk> der, brb, fire drill
[19:08] <chaos_> ups
[19:09] <chaos_> lxo, lxo ceph -w should be 'forever'
[19:09] <chaos_> lxo, ceph -s is one time info
[19:09] <sagewk> nm, wrong floor
[19:10] <sagewk> iirc. teh process is basically: 1) add to your ceph.conf (so ceph-mon starts up nicely), 2) ceph-mon --mkfs ... , 3) start ceph-mon. it'll join the cluster on it's own.. no explicit 'ceph mon add ..' is needed.
[19:10] <sagewk> will update that documentation today.
[19:10] <chaos_> hmm but it works
[19:11] <chaos_> when you add new mon to ceph.conf at existing mon 'ceph' fails with assertion failed
[19:11] <chaos_> please disable ceph mon add then ;p
[19:12] <Tv> lxo: somebody mentioned that on the mailing list
[19:13] <chaos_> sagewk, removing mon daemon is also easy?
[19:15] <sagewk> it works without the ceph mon add you m ean?
[19:15] <chaos_> sagewk, 'ceph mon add' still works ;p
[19:15] <sagewk> need to test that.. but i think it's 1) kill daemon, 2) ceph mon rm ...
[19:15] <sagewk> brb
[19:15] <sagewk> ah
[19:15] <sagewk> k
[19:37] <todin> sagewk: hi, the wip-journal branch did not fix it :-(
[19:39] <sagewk> todin: can you send the replay log with the new code?
[19:39] <todin> sagewk: i did that already
[19:40] <sagewk> oh i see
[19:51] <sagewk> todin: sweet, reproduced it.
[19:52] <todin> sagewk: ok, so you do not need more infos form me?
[19:52] <sagewk> todin: nope. thanks
[19:58] <sagewk> todin: can you retry wip-journal now?
[19:59] <todin> sagewk: ok, needs a few minutes
[20:07] <lxo> chaos_, I know ceph -w was supposed to be forever. the problem is it goes silent after a while. the amount of time in which it reports cluster info appears to be exactly 5 minutes in 0.39. in earlier versions, it kept on reporting forever
[20:11] * cp (~cp@76-220-17-197.lightspeed.sntcca.sbcglobal.net) Quit (Quit: cp)
[20:13] <sagewk> lxo, chaos_: will look in this today
[20:13] <chaos_> oh didn't know that, sorry ;p i haven't tried keep 'ceph -w', such long
[20:13] <lxo> btw, is there any way to get rid of a couple of mds oneshot replays that I killed before they completed? they've been there in the mds status line for days, in laggy-crashed state, but never going away
[20:14] * cp (~cp@76-220-17-197.lightspeed.sntcca.sbcglobal.net) has joined #ceph
[20:15] <lxo> sagewk, thanks. I'll assume you'll beat me to it, then, and look into the mds large-dirs-disappearing-files problem, that I'd planned to look into last night but ended up upgrading the cluster to 0.39 instead
[20:15] <sagewk> lxo: can you pastebin the mds dump?
[20:16] <lxo> err, regarding which problem?
[20:16] <lxo> if it's large-dirs-disappearing-files, it's supposed to be in the bug report. lemme look up the number
[20:17] <sagewk> the laggy-crashed thing. 'ceph mds dump'
[20:18] <lxo> is it this line that you want? 2011-12-05 17:18:00.963858 mds e43447: 1/1/1 up {0=1=up:active}, 2 up:standby-replay, 2 up:oneshot-replay(laggy or crashed)
[20:18] <lxo> I haven't kept mds dumps for those oneshot runs that I killed before they were completed
[20:19] <sagewk> lxo: the output from 'ceph mds dump'.
[20:19] <lxo> IIRC I wanted to clear the mds journal
[20:20] <sagewk> lxo: (it's just a dump of the mds map which that line is summarizing)
[20:21] <lxo> sage, http://pastebin.com/m0tk8D1R
[20:21] <cp> Is anyone else having trouble installing ceph today?
[20:21] <lxo> ^sagewk too
[20:21] <cp> looks like libcrypto++8 is not libcrypto++9
[20:21] <cp> s/not/now
[20:22] <sagewk> lxo: ok, 'ceph mds rm 37315' (and 37321) will clean those out. and opening a bug
[20:23] <sagewk> cp: debian sid?
[20:23] <lxo> sagewk, thanks. you mean you're opening the bug, or asking me to? (just to avoid a dupe)
[20:23] <sagewk> lxo: i opened it. :) thanks!
[20:25] <todin> sagewk: now it seems to be fixed
[20:25] <sagewk> todin: great, thanks!
[20:39] <sagewk> cp: let us know which deb are you using, and which platform are you installing on. i had to kludge around sid brokenness for 0.39 (happens about 30% of the time) so that may have been it.
[20:40] <cp> sagewk: I'm starting over to see if that helps. Ubuntu Oneric
[20:40] <sagewk> cp: those packages are built in an oneiric pbuilder.. the deps should be accurate. maybe double check your apt source line is right?
[20:45] <cp> sagewk: this is basically what I'm doing:
[20:45] <cp> wget http://newdream.net/~sage/pubkey.asc -q -O - | sudo apt-key add -
[20:45] <cp> echo 'deb http://ceph.newdream.net/debian/ oneiric main' >> /etc/apt/sources.list
[20:45] <cp> echo 'deb-src http://ceph.newdream.net/debian/ oneiric main' >> /etc/apt/sources.list
[20:45] <cp> apt-get update
[20:45] <cp> apt-get -y install ceph
[20:47] <cp> sagewk: Ah, I figured it out. I had some natty stuff still around
[20:47] <cp> THanks
[20:47] <Tv> cp: fwiw you can avoid yet another man-in-the-middle attack and grab the key with https.. http://ceph.newdream.net/docs/latest/ops/install/mkcephfs/#debian-ubuntu
[20:49] * lero (~lero@2001:470:1f06:b7c::2) has joined #ceph
[20:49] <cp> tv: Thanks
[21:14] * MarkDude (~MT@c-71-198-138-155.hsd1.ca.comcast.net) has joined #ceph
[21:21] * The_Bishop (~bishop@port-92-206-76-165.dynamic.qsc.de) Quit (Quit: Wer zum Teufel ist dieser Peer? Wenn ich den erwische dann werde ich ihm mal die Verbindung resetten!)
[21:25] * cp (~cp@76-220-17-197.lightspeed.sntcca.sbcglobal.net) Quit (Quit: cp)
[21:30] * cp (~cp@adsl-76-194-112-188.dsl.pltn13.sbcglobal.net) has joined #ceph
[21:55] * lero (~lero@2001:470:1f06:b7c::2) has left #ceph
[21:58] * cp (~cp@adsl-76-194-112-188.dsl.pltn13.sbcglobal.net) Quit (Quit: cp)
[21:58] * verwilst (~verwilst@dD576F878.access.telenet.be) has joined #ceph
[22:05] * cp (~cp@76-220-17-197.lightspeed.sntcca.sbcglobal.net) has joined #ceph
[22:05] * cp (~cp@76-220-17-197.lightspeed.sntcca.sbcglobal.net) Quit ()
[22:10] <Tv> grape: are you there?
[22:20] * The_Bishop (~bishop@p4FCDF3AF.dip.t-dialin.net) has joined #ceph
[22:23] * bchrisman (~Adium@ Quit (Quit: Leaving.)
[22:24] * bchrisman (~Adium@ has joined #ceph
[22:42] <sagewk> todin, lxo: pushed fix for the ceph -w thing
[22:44] * cp (~cp@c-67-180-21-185.hsd1.ca.comcast.net) has joined #ceph
[22:52] * mgalkiewicz (~maciej.ga@ has joined #ceph
[22:52] <mgalkiewicz> hello I have noticed that with new ceph version adding monitors is simplified. Do you have any documentation how to perform it?
[22:58] * stingray (~stingray@stingr.net) has joined #ceph
[23:06] * yehuda_hm (~yehuda@99-48-179-68.lightspeed.irvnca.sbcglobal.net) has joined #ceph
[23:13] <Tv> mgalkiewicz: not yet :(
[23:25] <sagewk> mgalkiewicz, tv: just pushed doc to master branch
[23:32] * nwatkins (~nwatkins@kyoto.soe.ucsc.edu) has joined #ceph
[23:36] <mgalkiewicz> so it is available in ceph repository?
[23:38] * verwilst (~verwilst@dD576F878.access.telenet.be) Quit (Quit: Ex-Chat)

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