#ceph IRC Log

Index

IRC Log for 2012-12-24

Timestamps are in GMT/BST.

[0:09] * ScOut3R (~ScOut3R@1F2EA078.dsl.pool.telekom.hu) Quit (Remote host closed the connection)
[0:25] * scuttlemonkey (~scuttlemo@96-42-146-5.dhcp.trcy.mi.charter.com) Quit (Quit: This computer has gone to sleep)
[0:45] * CloudGuy (~CloudGuy@5356416B.cm-6-7b.dynamic.ziggo.nl) has joined #ceph
[0:51] * ivan` (~ivan`@000130ca.user.oftc.net) Quit (Read error: Operation timed out)
[0:51] * ivan` (~ivan`@000130ca.user.oftc.net) has joined #ceph
[0:58] * loicd (~loic@magenta.dachary.org) Quit (Quit: Leaving.)
[1:14] * raisin (5da05682@ircip2.mibbit.com) has joined #ceph
[1:27] * raisin (5da05682@ircip2.mibbit.com) Quit (Quit: http://www.mibbit.com ajax IRC Client)
[1:28] * scuttlemonkey (~scuttlemo@96-42-146-5.dhcp.trcy.mi.charter.com) has joined #ceph
[1:28] * ChanServ sets mode +o scuttlemonkey
[1:32] <Etherael> Anyone know if there are packages for raspbian?
[1:40] * scuttlemonkey (~scuttlemo@96-42-146-5.dhcp.trcy.mi.charter.com) Quit (Quit: This computer has gone to sleep)
[1:54] * ScOut3R (~ScOut3R@1F2EA078.dsl.pool.telekom.hu) has joined #ceph
[1:59] * rektide_ (~rektide@deneb.eldergods.com) has joined #ceph
[1:59] * rektide (~rektide@deneb.eldergods.com) Quit (Read error: Connection reset by peer)
[2:03] * dosaboy (~user1@host86-163-127-16.range86-163.btcentralplus.com) Quit (Quit: Leaving.)
[2:03] * ScOut3R (~ScOut3R@1F2EA078.dsl.pool.telekom.hu) Quit (Remote host closed the connection)
[2:09] * Steki (~steki@bojanka.net) has joined #ceph
[2:13] * BManojlovic (~steki@85.222.183.165) Quit (Ping timeout: 480 seconds)
[2:25] * Steki (~steki@bojanka.net) Quit (Quit: Ja odoh a vi sta 'ocete...)
[2:27] * lx0 (~aoliva@lxo.user.oftc.net) has joined #ceph
[2:29] * scuttlemonkey (~scuttlemo@96-42-146-5.dhcp.trcy.mi.charter.com) has joined #ceph
[2:29] * ChanServ sets mode +o scuttlemonkey
[2:31] * tryggvil (~tryggvil@17-80-126-149.ftth.simafelagid.is) has joined #ceph
[2:32] * lxo (~aoliva@lxo.user.oftc.net) Quit (Ping timeout: 480 seconds)
[2:33] * scuttlemonkey (~scuttlemo@96-42-146-5.dhcp.trcy.mi.charter.com) Quit ()
[2:53] * KindOne (~KindOne@h138.181.130.174.dynamic.ip.windstream.net) has joined #ceph
[2:57] * drokita_ (~drokita@24-107-180-86.dhcp.stls.mo.charter.com) Quit (Quit: drokita_)
[3:01] * Ryan_Lane1 (~Adium@120.sub-70-196-132.myvzw.com) Quit (Quit: Leaving.)
[3:02] * Ryan_Lane (~Adium@140.sub-70-196-128.myvzw.com) has joined #ceph
[3:03] * Ryan_Lane (~Adium@140.sub-70-196-128.myvzw.com) Quit ()
[3:04] * Ryan_Lane (~Adium@140.sub-70-196-128.myvzw.com) has joined #ceph
[3:26] * scuttlemonkey (~scuttlemo@96-42-146-5.dhcp.trcy.mi.charter.com) has joined #ceph
[3:26] * ChanServ sets mode +o scuttlemonkey
[3:34] * ScOut3R (~ScOut3R@1F2EA078.dsl.pool.telekom.hu) has joined #ceph
[4:12] * ScOut3R (~ScOut3R@1F2EA078.dsl.pool.telekom.hu) Quit (Remote host closed the connection)
[4:17] * The_Bishop__ (~bishop@e179015021.adsl.alicedsl.de) Quit (Quit: Wer zum Teufel ist dieser Peer? Wenn ich den erwische dann werde ich ihm mal die Verbindung resetten!)
[4:46] * scuttlemonkey (~scuttlemo@96-42-146-5.dhcp.trcy.mi.charter.com) Quit (Quit: This computer has gone to sleep)
[5:00] * scuttlemonkey (~scuttlemo@96-42-146-5.dhcp.trcy.mi.charter.com) has joined #ceph
[5:00] * ChanServ sets mode +o scuttlemonkey
[5:42] * scuttlemonkey (~scuttlemo@96-42-146-5.dhcp.trcy.mi.charter.com) Quit (Quit: This computer has gone to sleep)
[5:47] * stp (~stp@dslb-084-056-011-102.pools.arcor-ip.net) has joined #ceph
[5:54] * scalability-junk (~stp@dslb-084-056-040-011.pools.arcor-ip.net) Quit (Ping timeout: 480 seconds)
[6:06] * calebamiles1 (~caleb@c-107-3-1-145.hsd1.vt.comcast.net) has joined #ceph
[6:09] * calebamiles (~caleb@c-107-3-1-145.hsd1.vt.comcast.net) Quit (Ping timeout: 480 seconds)
[6:19] * guigouz (~guigouz@177.33.243.196) Quit (Quit: Computer has gone to sleep.)
[6:39] * CloudGuy (~CloudGuy@5356416B.cm-6-7b.dynamic.ziggo.nl) Quit (Remote host closed the connection)
[7:09] * Etherael1 (~eric@node-fim.pool-101-108.dynamic.totbb.net) has joined #ceph
[7:14] * Etherael (~eric@node-8n1.pool-125-24.dynamic.totbb.net) Quit (Read error: Operation timed out)
[8:21] * illuminatis (~illuminat@0001adba.user.oftc.net) Quit (Ping timeout: 480 seconds)
[8:40] * Cube (~Cube@107.38.71.134) has joined #ceph
[8:51] * joao (~JL@89.181.158.122) Quit (Ping timeout: 480 seconds)
[9:15] * illuminatis (~illuminat@0001adba.user.oftc.net) has joined #ceph
[9:20] * madkiss (~madkiss@178.188.60.118) has joined #ceph
[9:26] * agh (~2ee79308@2600:3c00::2:2424) has joined #ceph
[9:26] <agh> Hello to all
[9:26] <agh> I've a problem. It'es maybe a bug
[9:26] <madkiss> hello agh
[9:26] <madkiss> oh
[9:27] <agh> I've OSD nodes
[9:27] <agh> with 2 NICS gigabit
[9:28] <agh> one for private network, the other one for the public
[9:28] <agh> and i do my tests on a 10Giga client
[9:28] <agh> a simple dd)
[9:28] <agh> But... doing that make the cluster going wrong. I've a lot of : [WRN] slow requests
[9:31] <agh> Any idea ?
[9:36] <madkiss> hm
[9:38] <agh> When I do this test on a 1gigabit client, it's okay
[9:38] <agh> But... seems that Ceph does not deal with this fact : a client faster than the cluster...
[9:38] <madkiss> well, do your OSD nodes have one network interface only?
[9:39] <madkiss> ah, no, sorry
[9:39] <agh> no, 2 NICs
[9:39] <madkiss> "private" and "public"
[9:39] <madkiss> what interface do they use for inter-OSD traffic?
[9:39] <agh> both interfaces are gigabit on OSDs
[9:39] <agh> only the client is 10G
[9:40] <madkiss> again, what interface do the OSDs use for inter-OSD traffic? the private or the public one?
[9:40] <agh> private one
[9:41] <madkiss> and for the traffic you produce with your client?
[9:41] <agh> on the public one
[9:41] <agh> I use CephFS, not rdb
[9:41] <madkiss> are you sure?
[9:42] <agh> Yes, I am :)
[9:42] <madkiss> What I suspect that's happening is that your client- and your inter-osd-traffic go through the same interface.
[9:42] <madkiss> How did you determine?
[9:43] <agh> 10sec, I do a pastbin of my conf and of my bench
[9:45] <agh> http://pastebin.com/JU8LvjUH
[9:47] <madkiss> Ah. Well, what I would recommend do is finding out which interfaces OSDs and public traffic really gothrough.
[9:47] <agh> I now, i use iftop on each interface
[9:47] <agh> and the result is coherent :
[9:48] <agh> client send data on the public interface to all the OSDs hosts.
[9:48] <agh> Then OSD speaks together on the private NIC
[9:50] <agh> When my tests are done with 1G everywhere, it's OK. But as soon as the client is faster than the cluster, it sucks
[9:58] <madkiss> Disk I/O problems on the OSD nodes?
[10:01] <agh> No... very good machine, with SAS 10K disks.
[10:01] <agh> The fact is that everything is working fine when the client is gigabit
[10:02] <madkiss> bonding or something?
[10:02] <agh> => dd if=/dev/zero of=/mnt/cephfs/toto bs=4k count=50M gives 125 MB/s ==> the gigabit network is the bottleneck
[10:02] <madkiss> that was sort of obvious.
[10:03] <madkiss> what'S the load on your OSDs when doing the test?
[10:03] * jbd_ (~jbd_@34322hpv162162.ikoula.com) has joined #ceph
[10:04] <agh> load with XFS is quite low.
[10:10] <agh> So ? No idea ? I'm sad!
[10:13] <madkiss> sorry, no idea.
[10:13] <madkiss> I'll have to sod off now anyway. Merry Christmas :)
[10:14] <styx-tdo> agh: can you bond some GigE interfaces for higher thruput?
[10:14] <styx-tdo> agh: i have a setup where the file servers are quad-GigE and the clients are 2xGigE.. no ceph, yet, but that setup works w/ iSCSI and nfs
[10:15] <agh> styx-tdo: no I cant
[10:15] <styx-tdo> 125 MB/s should be fine w/ 2 GigEs; if you change the MTU it'd be better even
[10:15] * idnc_sk (~idnc_sk@92.251.255.5.threembb.ie) has joined #ceph
[10:16] <idnc_sk> btw > the problem I had yesterday
[10:16] <idnc_sk> the ret err 95 > HTML 500 for all write ops
[10:16] <idnc_sk> you have to have radosgw support compiled-in in the whole cluster to use it
[10:17] <idnc_sk> I had a 4node cluster 2x osd + mons [+] 2x mds + mons + rgw; after I compiled rgw for the 2 storage nodes, all worked fine
[10:18] <idnc_sk> + its working withouth problems(for files up to 1gig, starting a test for larger) behind a nginx loadbalancer
[10:32] * madkiss (~madkiss@178.188.60.118) Quit (Quit: Leaving.)
[10:43] * tryggvil (~tryggvil@17-80-126-149.ftth.simafelagid.is) Quit (Quit: tryggvil)
[10:52] * idnc_sk (~idnc_sk@92.251.255.5.threembb.ie) Quit (Quit: leaving)
[11:25] * pixel (~pixel@188.72.65.190) has joined #ceph
[11:27] * pixel (~pixel@188.72.65.190) Quit ()
[11:41] * Machske (~bram@d5152D8A3.static.telenet.be) has joined #ceph
[11:44] * joao (~JL@89-181-150-243.net.novis.pt) has joined #ceph
[11:44] * ChanServ sets mode +o joao
[11:57] * loicd (~loic@2a01:e35:2eba:db10:1464:adbf:1822:372e) has joined #ceph
[12:08] <alexxy> hi all!
[12:08] <alexxy> why cephfs reports bogus data on df?
[12:09] <Machske> bogus is what way ?
[12:09] <Machske> for me it's not bogus :)
[12:10] <alexxy> Filesystem Size Used Avail Use% Mounted on
[12:10] <alexxy> 10.0.0.254:/ 66G 5.2G 61G 8% /home
[12:10] <alexxy> while fs size is ~16T
[12:16] <Machske> mounted via the kernel rbd module ?
[12:17] <alexxy> no
[12:17] <Machske> ceph-fuse ?
[12:17] <alexxy> mounted via cephfs module
[12:17] <Machske> I use ceph-fuse because the client is on a osd
[12:17] <Machske> that reports the size correctly
[12:18] <Machske> you could try to also mount it with ceph-fuse to see it reports correctly in that manner
[12:19] * lx0 (~aoliva@lxo.user.oftc.net) Quit (Remote host closed the connection)
[12:20] * lx0 (~aoliva@lxo.user.oftc.net) has joined #ceph
[12:23] * ScOut3R (~ScOut3R@1F2EA078.dsl.pool.telekom.hu) has joined #ceph
[12:24] * Leseb (~Leseb@bea13-1-82-228-104-16.fbx.proxad.net) has joined #ceph
[12:34] * Leseb (~Leseb@bea13-1-82-228-104-16.fbx.proxad.net) Quit (Quit: Leseb)
[12:54] * loicd (~loic@2a01:e35:2eba:db10:1464:adbf:1822:372e) Quit (Quit: Leaving.)
[12:54] * loicd (~loic@magenta.dachary.org) has joined #ceph
[13:01] * Leseb (~Leseb@bea13-1-82-228-104-16.fbx.proxad.net) has joined #ceph
[13:03] * Leseb (~Leseb@bea13-1-82-228-104-16.fbx.proxad.net) Quit ()
[13:13] * tryggvil (~tryggvil@17-80-126-149.ftth.simafelagid.is) has joined #ceph
[13:16] * madkiss (~madkiss@178.188.60.118) has joined #ceph
[13:17] * Vjarjadian (~IceChat77@5ad6d005.bb.sky.com) has joined #ceph
[13:18] * sagewk (~sage@2607:f298:a:607:6df9:7a80:af99:5918) Quit (Read error: Operation timed out)
[13:18] * madkiss (~madkiss@178.188.60.118) Quit ()
[13:28] * roald (~roaldvanl@87.209.150.214) has joined #ceph
[13:34] * sagewk (~sage@2607:f298:a:607:3c31:7432:41de:a8ca) has joined #ceph
[13:49] * CloudGuy (~CloudGuy@5356416B.cm-6-7b.dynamic.ziggo.nl) has joined #ceph
[13:53] * joao (~JL@89-181-150-243.net.novis.pt) Quit (Read error: Connection reset by peer)
[14:26] * vhasi_ (vhasi@vha.si) has joined #ceph
[14:28] * tryggvil (~tryggvil@17-80-126-149.ftth.simafelagid.is) Quit (Quit: tryggvil)
[14:28] * vhasi (vhasi@vha.si) Quit (Ping timeout: 480 seconds)
[14:29] * vhasi_ is now known as vhasi
[14:36] * ScOut3R (~ScOut3R@1F2EA078.dsl.pool.telekom.hu) Quit (Remote host closed the connection)
[14:54] * yehudasa (~yehudasa@2607:f298:a:607:f417:6a39:eebd:1d71) Quit (Read error: Operation timed out)
[14:56] * darkfader (~floh@188.40.175.2) has joined #ceph
[14:56] * madkiss (~madkiss@178.188.60.118) has joined #ceph
[14:57] * roald (~roaldvanl@87.209.150.214) Quit (Ping timeout: 480 seconds)
[15:06] * noob2 (~noob2@ext.cscinfo.com) has joined #ceph
[15:10] * yehudasa (~yehudasa@2607:f298:a:607:b149:30d3:e282:cd01) has joined #ceph
[15:15] * agh (~2ee79308@2600:3c00::2:2424) Quit (Quit: TheGrebs.com CGI:IRC (Session timeout))
[15:38] * tsygrl (~tsygrl314@c-75-68-140-25.hsd1.vt.comcast.net) has joined #ceph
[16:31] <Machske> is there a way to speed up recovert ? It seems slow sometimes, but there's plenty of disk io and network bandwidth unused during recovery
[16:46] * Machske (~bram@d5152D8A3.static.telenet.be) Quit (Quit: Ik ga weg)
[16:51] * scuttlemonkey (~scuttlemo@96-42-146-5.dhcp.trcy.mi.charter.com) has joined #ceph
[16:51] * ChanServ sets mode +o scuttlemonkey
[17:05] * ScOut3R (~ScOut3R@catv-89-133-42-156.catv.broadband.hu) has joined #ceph
[17:22] * ScOut3R (~ScOut3R@catv-89-133-42-156.catv.broadband.hu) Quit (Remote host closed the connection)
[17:25] * Leseb (~Leseb@bea13-1-82-228-104-16.fbx.proxad.net) has joined #ceph
[17:31] * ScOut3R (~ScOut3R@catv-89-133-42-156.catv.broadband.hu) has joined #ceph
[17:35] * uebassss (~enriquegq@65.Red-81-35-69.dynamicIP.rima-tde.net) has joined #ceph
[17:35] <uebassss> http://www.carolinaherrera.com/212/es/areyouonthelist?share=MbM1LOcZQJIqFu0FDvOJkJeFxDGt60lD65Nzz-7UPmPkz4rz3EUUdzs6j6FXsjB4447F-isvxjqkXd4Qey2GHw#episodio-6
[17:35] * uebassss (~enriquegq@65.Red-81-35-69.dynamicIP.rima-tde.net) Quit (autokilled: Do not spam other people. Mail support@oftc.net if you feel this is in error. (2012-12-24 16:35:18))
[17:42] * madkiss (~madkiss@178.188.60.118) Quit (Quit: Leaving.)
[17:46] * ScOut3R (~ScOut3R@catv-89-133-42-156.catv.broadband.hu) Quit (Remote host closed the connection)
[17:49] * scuttlemonkey (~scuttlemo@96-42-146-5.dhcp.trcy.mi.charter.com) Quit (Quit: This computer has gone to sleep)
[18:32] * CloudGuy (~CloudGuy@5356416B.cm-6-7b.dynamic.ziggo.nl) Quit (Remote host closed the connection)
[18:33] * scuttlemonkey (~scuttlemo@96-42-146-5.dhcp.trcy.mi.charter.com) has joined #ceph
[18:33] * ChanServ sets mode +o scuttlemonkey
[18:37] * loicd (~loic@magenta.dachary.org) Quit (Quit: Leaving.)
[18:37] * loicd (~loic@magenta.dachary.org) has joined #ceph
[18:56] * loicd (~loic@magenta.dachary.org) Quit (Quit: Leaving.)
[18:56] * loicd (~loic@magenta.dachary.org) has joined #ceph
[19:20] * ScOut3R (~ScOut3R@1F2EA078.dsl.pool.telekom.hu) has joined #ceph
[19:23] * ScOut3R (~ScOut3R@1F2EA078.dsl.pool.telekom.hu) Quit (Remote host closed the connection)
[19:24] * ScOut3R (~ScOut3R@1F2EA078.dsl.pool.telekom.hu) has joined #ceph
[19:50] * BManojlovic (~steki@85.222.183.165) has joined #ceph
[20:06] * Leseb (~Leseb@bea13-1-82-228-104-16.fbx.proxad.net) Quit (Quit: Leseb)
[20:36] * loicd (~loic@magenta.dachary.org) Quit (Quit: Leaving.)
[20:36] * loicd (~loic@magenta.dachary.org) has joined #ceph
[21:02] * ScOut3R (~ScOut3R@1F2EA078.dsl.pool.telekom.hu) Quit (Remote host closed the connection)
[21:03] * ScOut3R (~ScOut3R@1F2EA078.dsl.pool.telekom.hu) has joined #ceph
[21:21] * ScOut3R (~ScOut3R@1F2EA078.dsl.pool.telekom.hu) Quit (Remote host closed the connection)
[21:34] * BManojlovic (~steki@85.222.183.165) Quit (Quit: Ja odoh a vi sta 'ocete...)
[21:41] * loicd (~loic@magenta.dachary.org) Quit (Quit: Leaving.)
[21:41] * loicd (~loic@magenta.dachary.org) has joined #ceph
[21:45] * joao (~JL@89-181-150-243.net.novis.pt) has joined #ceph
[21:45] * ChanServ sets mode +o joao
[21:46] * roald (~Roald@87.209.150.214) has joined #ceph
[21:47] * loicd (~loic@magenta.dachary.org) Quit (Quit: Leaving.)
[21:53] * ScOut3R (~ScOut3R@1F2EA078.dsl.pool.telekom.hu) has joined #ceph
[22:19] * scuttlemonkey (~scuttlemo@96-42-146-5.dhcp.trcy.mi.charter.com) Quit (Quit: This computer has gone to sleep)
[22:27] * ScOut3R (~ScOut3R@1F2EA078.dsl.pool.telekom.hu) Quit (Remote host closed the connection)
[22:28] * scuttlemonkey (~scuttlemo@96-42-146-5.dhcp.trcy.mi.charter.com) has joined #ceph
[22:28] * ChanServ sets mode +o scuttlemonkey
[22:38] * CloudGuy (~CloudGuy@5356416B.cm-6-7b.dynamic.ziggo.nl) has joined #ceph
[23:15] * CloudGuy (~CloudGuy@5356416B.cm-6-7b.dynamic.ziggo.nl) Quit (Remote host closed the connection)
[23:31] * loicd (~loic@magenta.dachary.org) has joined #ceph

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