#ceph IRC Log

Index

IRC Log for 2011-09-24

Timestamps are in GMT/BST.

[0:03] * cp (~cp@adsl-76-194-112-188.dsl.pltn13.sbcglobal.net) Quit (Quit: cp)
[0:07] <Tv|work> fyi i'm fixing a bunch of the c* -> ceph-* changes, push coming as soon as make distcheck completes
[0:10] * cp (~cp@adsl-76-194-112-188.dsl.pltn13.sbcglobal.net) has joined #ceph
[0:11] <gregaf1> sagewk: you have any idea about this rbd OOM thing?
[0:21] <sagewk> what's the workload?
[0:22] <sagewk> how long does it take before the rsync makes it oom?
[0:22] <gregaf1> wonko_be said said 1TB rsync inside of a Xen dom0 (I think), dunno how long
[0:23] * bugoff (bram@november.openminds.be) has joined #ceph
[0:30] <Tv|work> so "make distcheck" has apparently been broken for a while.. fixing.
[0:36] * jojy (~jojyvargh@70-35-37-146.static.wiline.com) Quit (Read error: Operation timed out)
[0:38] <Tv|work> gregaf1: wouldn't a better criteria for a radosgw PUT being an acl put be the "?acl" query string?
[0:39] <Tv|work> http://docs.amazonwebservices.com/AmazonS3/latest/API/RESTObjectPUTacl.html
[0:45] * cp (~cp@adsl-76-194-112-188.dsl.pltn13.sbcglobal.net) Quit (Quit: cp)
[0:45] * cp (~cp@adsl-76-194-112-188.dsl.pltn13.sbcglobal.net) has joined #ceph
[0:46] * cp (~cp@adsl-76-194-112-188.dsl.pltn13.sbcglobal.net) Quit ()
[0:55] <Tv|work> rename cleanup, distcheck cleanup, etc pushed
[0:56] <Tv|work> also fixes for make dist, rpm & deb builds
[1:05] * bchrisman (~Adium@64.164.138.146) Quit (Quit: Leaving.)
[1:07] * cp (~cp@m892036d0.tmodns.net) has joined #ceph
[1:10] * cp (~cp@m892036d0.tmodns.net) Quit ()
[1:14] * cp (~cp@m892036d0.tmodns.net) has joined #ceph
[1:15] * The_Bishop (~bishop@sama32.de) Quit (Remote host closed the connection)
[1:26] * cp (~cp@m892036d0.tmodns.net) Quit (Quit: cp)
[1:44] * Tv|work (~Tv|work@aon.hq.newdream.net) Quit (Ping timeout: 480 seconds)
[1:51] * bchrisman (~Adium@c-98-207-207-62.hsd1.ca.comcast.net) has joined #ceph
[2:09] * sjustlaptop (~sam@aon.hq.newdream.net) Quit (Quit: Leaving.)
[3:04] * joshd (~joshd@aon.hq.newdream.net) Quit (Quit: Leaving.)
[3:20] * sjustlaptop (~sam@96-41-121-194.dhcp.mtpk.ca.charter.com) has joined #ceph
[3:26] * sjustlaptop (~sam@96-41-121-194.dhcp.mtpk.ca.charter.com) Quit (Quit: Leaving.)
[3:30] * jojy (~jojyvargh@67-207-96-234.static.wiline.com) has joined #ceph
[4:07] * jojy (~jojyvargh@67-207-96-234.static.wiline.com) Quit (Quit: jojy)
[4:38] * adjohn (~adjohn@50-0-92-177.dsl.dynamic.sonic.net) Quit (Quit: adjohn)
[4:52] * jojy (~jojyvargh@75-54-231-2.lightspeed.sntcca.sbcglobal.net) has joined #ceph
[4:54] * jojy (~jojyvargh@75-54-231-2.lightspeed.sntcca.sbcglobal.net) has left #ceph
[5:50] * adjohn (~adjohn@50-0-92-177.dsl.dynamic.sonic.net) has joined #ceph
[5:58] * votz (~votz@pool-108-52-121-23.phlapa.fios.verizon.net) Quit (Quit: Leaving)
[6:04] * adjohn (~adjohn@50-0-92-177.dsl.dynamic.sonic.net) Quit (Quit: adjohn)
[8:18] * sjustlaptop (~sam@96-41-121-194.dhcp.mtpk.ca.charter.com) has joined #ceph
[8:21] * sjustlaptop (~sam@96-41-121-194.dhcp.mtpk.ca.charter.com) Quit ()
[10:17] * tjikkun (~tjikkun@2001:7b8:356:0:225:22ff:fed2:9f1f) has joined #ceph
[10:39] <wonko_be> gregaf1 / sagewk: it went OOM after about 3 hours of rsync'ing
[10:41] <wonko_be> the data is mixed: a lot of documents, source code and mails, and a fairly large set of *ahum* backups of mp3 and avi files ;)
[14:15] * lxo (~aoliva@lxo.user.oftc.net) Quit (Remote host closed the connection)
[14:16] * lxo (~aoliva@lxo.user.oftc.net) has joined #ceph
[15:05] * jclendenan (~jclendena@204.244.194.20) Quit (Read error: Connection reset by peer)
[15:30] * darkfaded (~floh@188.40.175.2) has joined #ceph
[15:33] * darkfader (~floh@188.40.175.2) Quit (Remote host closed the connection)
[17:08] * Kioob1 (~Kioob@luuna.daevel.fr) has joined #ceph
[17:08] <Kioob1> hi
[17:09] * Kioob1 (~Kioob@luuna.daevel.fr) Quit ()
[17:09] * Kioob (~kioob@luuna.daevel.fr) has joined #ceph
[18:14] * Kioob (~kioob@luuna.daevel.fr) Quit (Quit: Leaving.)
[18:14] * Kioob (~kioob@luuna.daevel.fr) has joined #ceph
[19:43] * greglap (~Adium@166.205.141.87) has joined #ceph
[19:44] <greglap> hi Kioob
[20:27] * gregorg (~Greg@78.155.152.6) Quit (Ping timeout: 480 seconds)
[20:30] * greglap (~Adium@166.205.141.87) Quit (Quit: Leaving.)
[20:30] * greglap (~Adium@166.205.141.87) has joined #ceph
[21:13] <lxo> I'm observing a significant increase in memory footprint in cmon and cosd in 0.35. is that expected?
[21:27] <greglap> lxo: what levels? I wouldn't have thought so but we did widen a few basic data type
[21:27] <greglap> *types
[21:57] * Dantman (~dantman@S0106001731dfdb56.vs.shawcable.net) Quit (Read error: Connection reset by peer)
[21:57] * Dantman (~dantman@S0106001731dfdb56.vs.shawcable.net) has joined #ceph
[22:36] <Kioob> in the FAQ, it's said that Ceph is not ready for production use. Is it still the case ?
[22:55] * jclendenan (~jclendena@204.244.194.20) has joined #ceph
[23:01] <greglap> Kioob: depends on which parts of the system you're using, but basically yes: Ceph is not production-stable
[23:28] <lxo> greglap, it used to take some 1.3, 2 or 4 G per OSD, depending on 3, 2 or 1 active OSD. when I mentioned it, it was up to 6G on one and it got up to 14G on another
[23:29] <lxo> I single OSD, still recovering from a restart, is already past 8GB
[23:31] <lxo> now, since one got up to 14G, I think it's a safe bet that there's some leak, no?
[23:33] <greglap> lxo: oh, I think maybe there was a bit of badness that got in and is fixed in commit 5503d4501742fb6e8aee3e4096e75c933edc8e8f
[23:34] <greglap> due to some of the PG changes
[23:34] <greglap> can you try pulling that in? (it's on the stable branch)
[23:34] <greglap> otherwise I'll have to discuss with people on Monday
[23:36] <lxo> will do
[23:36] <lxo> thanks
[23:43] <lxo> greglap, 5503d450 is something else, unrelated to memory, and I already got that in
[23:51] <greglap> it's not a memory leak, but it was a problem the I think spun up a lot of memory-heavy ops
[23:51] <greglap> in any case, if you already have it then I'll need to discuss with people on Monday, sorry
[23:55] <lxo> hmm, actually, I'm not using that right now; the only machine on which I installed this patch is the one whose osd is inactive atm (it's trying to run my old ceph FS, processing the corrupt logs over and over on each restart). I'll try it on the others
[23:55] * greglap (~Adium@166.205.141.87) Quit (Read error: Connection reset by peer)
[23:55] <lxo> and then let you know
[23:55] <lxo> hmm, I guess he didn't see this
[23:55] <lxo> gregaf1, see above :-)
[23:56] * greglap (~Adium@166.205.141.87) has joined #ceph

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