#ceph IRC Log

Index

IRC Log for 2011-12-17

Timestamps are in GMT/BST.

[0:00] <todin> gregaf: http://85.214.49.87/ceph/osd.2.log.bz2
[0:02] <gregaf> todin: 403 Forbidden
[0:02] <todin> gregaf: fixed
[0:03] <guido> The news item for 0.39 says that for 0.40, you're going to work on rbd image cloning, but I cannot find anything like that in the redmine roadmap for 0.40. Am I just missing it?
[0:03] <gregaf> guido: I believe all the current tickets talk about "layering"
[0:05] <guido> I can't find anything about layering in the roadmap either...
[0:07] * morse (~morse@supercomputing.univpm.it) Quit (Ping timeout: 480 seconds)
[0:08] <gregaf> well, joshd knows the current status on that; maybe it got pushed back
[0:09] <todin> gregaf: another osd failed with loads of this failed decode messages http://pastebin.com/q75rzJje but this osd has an older ceph version
[0:10] <gregaf> you're running multiple versions simultaneously?
[0:11] <todin> gregaf: yep, how should I update with no down time?
[0:11] <todin> gregaf: multiple = 2
[0:11] <gregaf> well, nominally we try to make it work but I'm cautious about whether it actually does right now
[0:13] <todin> gregaf: so we should take no effort to debug this, I upgrade all osd with a downtime, an will see if it happends again?
[0:13] * mgalkiewicz (~maciej.ga@staticline18746.toya.net.pl) Quit (Remote host closed the connection)
[0:13] <gregaf> that's probably best, yeah
[0:14] <gregaf> my bet is their messengers got out of sync somehow
[0:14] <gregaf> although even if they are good on the same version we need to fix the new one running out of memory when the authorization failed (or whatever's actually broken)
[0:16] <todin> gregaf: that's true, but not today :-) I am out, have a nice weekend
[0:17] <gregaf> okay, you too!
[0:31] * fronlius (~fronlius@testing78.jimdo-server.com) Quit (Quit: fronlius)
[0:39] <Tv> guido: http://tracker.newdream.net/issues/1772 is relevant for that
[0:40] <joshd> guido: also http://tracker.newdream.net/issues/1773
[1:03] <guido> Thx
[1:03] <guido> I wouldn't have guessed it from the descriptions...
[1:04] <gregaf> those aren't the main ones
[1:04] <guido> Hm, if this involves a new header format for rbd images, will that make old ones unusable?
[1:05] <Tv> guido: i expect there will be backwards compat, but to enable the feature you need the new header.. we can probably convert them seamlessly
[1:07] <joshd> yup, tv's got the plan right
[1:08] <Tv> sjust: fyi wrt the mailing list: "Abgebrochen" = Abort
[1:08] * yhager (~yhager@173.180.85.48) Quit (Ping timeout: 480 seconds)
[1:08] <Tv> though that might mean user hit control-C
[1:09] <joshd> the new format will also have a notion of features supported, and the only interface dealing with it will be the rbd class on the osd, so future changes will be much easier
[1:10] <joshd> currently the rbd header on-disk format is read directly by the kernel and librbd
[1:12] * yhager (~yhager@173.180.85.48) has joined #ceph
[1:12] <guido> Huh? RBD class on the OSD? I thought RBD was implemented on the client side and would look like any other rados client to the osds...
[1:14] <joshd> the osds have an interface for defining custom operations on an object, like doing atomic read-modify-writes without any roundtrips to clients
[1:14] <joshd> rbd uses that for adding/removing snapshots from the rbd image header
[1:17] <joshd> the splitting of the block device into objects is all done client-side - there's just some header operations that are on the osd
[1:17] <joshd> any rados client could use the same methods rbd uses with rados_exec() in librados
[1:19] <guido> The header is just one object containing metadata information about the image and a list of the objects with the actual data?
[1:21] <joshd> yeah, but it doesn't store a list of objects
[1:22] <guido> So how are the objects that hold the actual data identified?
[1:22] <joshd> the objects all the same size, and are named deterministically, so clients can interpret 'no object found' as zeros
[1:23] <joshd> objects are only created when they're first written to
[1:30] * andresambrois (~aa@r190-135-27-29.dialup.adsl.anteldata.net.uy) has joined #ceph
[1:34] * aa (~aa@r190-135-26-243.dialup.adsl.anteldata.net.uy) Quit (Read error: Operation timed out)
[1:34] * fronlius (~fronlius@g231136055.adsl.alicedsl.de) has joined #ceph
[1:40] * yhager (~yhager@173.180.85.48) Quit (Ping timeout: 480 seconds)
[1:53] * Tv (~Tv|work@aon.hq.newdream.net) Quit (Ping timeout: 480 seconds)
[1:59] * fronlius (~fronlius@g231136055.adsl.alicedsl.de) Quit (Quit: fronlius)
[2:01] * fronlius (~fronlius@g231136055.adsl.alicedsl.de) has joined #ceph
[2:09] * bchrisman (~Adium@108.60.121.114) Quit (Quit: Leaving.)
[2:09] * bchrisman (~Adium@108.60.121.114) has joined #ceph
[2:10] * bchrisman (~Adium@108.60.121.114) Quit ()
[2:17] * MattBenjamin (~matt@aa2.linuxbox.com) has left #ceph
[2:19] * fronlius (~fronlius@g231136055.adsl.alicedsl.de) Quit (Quit: fronlius)
[2:36] * Tv__ (~Tv__@cpe-76-168-227-45.socal.res.rr.com) has joined #ceph
[3:00] * yhager (~yhager@173.180.85.48) has joined #ceph
[3:37] * yhager (~yhager@173.180.85.48) Quit (Ping timeout: 480 seconds)
[3:51] * joshd (~joshd@aon.hq.newdream.net) Quit (Quit: Leaving.)
[4:18] * bchrisman (~Adium@c-76-103-130-94.hsd1.ca.comcast.net) has joined #ceph
[4:26] * bchrisman (~Adium@c-76-103-130-94.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[4:40] * yhager (~yhager@173.180.85.48) has joined #ceph
[4:59] * yhager (~yhager@173.180.85.48) Quit (Ping timeout: 480 seconds)
[5:12] * adjohn (~adjohn@70-36-197-80.dsl.dynamic.sonic.net) has joined #ceph
[5:15] * Tv__ (~Tv__@cpe-76-168-227-45.socal.res.rr.com) Quit (Ping timeout: 480 seconds)
[5:26] * adjohn (~adjohn@70-36-197-80.dsl.dynamic.sonic.net) Quit (Quit: adjohn)
[5:34] * adjohn (~adjohn@70-36-197-80.dsl.dynamic.sonic.net) has joined #ceph
[5:35] * adjohn (~adjohn@70-36-197-80.dsl.dynamic.sonic.net) Quit ()
[5:37] * votz (~votz@pool-108-52-122-97.phlapa.fios.verizon.net) has joined #ceph
[5:45] * bchrisman (~Adium@c-76-103-130-94.hsd1.ca.comcast.net) has joined #ceph
[6:16] * andresambrois (~aa@r190-135-27-29.dialup.adsl.anteldata.net.uy) Quit (Read error: Operation timed out)
[8:44] * Monster_Rob (MonsterRob@ppp-70-246-84-38.dsl.okcyok.swbell.net) has joined #ceph
[9:28] * Monster_Rob (MonsterRob@ppp-70-246-84-38.dsl.okcyok.swbell.net) Quit (Ping timeout: 480 seconds)
[9:36] * xns (~xns@evul.net) Quit (Read error: Connection reset by peer)
[9:51] * yhager (~yhager@173.180.85.48) has joined #ceph
[11:00] * yhager (~yhager@173.180.85.48) Quit (Remote host closed the connection)
[11:31] <wido> sjust: thanks! bfbde5b18525406fc3b678751459e989ea5d4977 fixed it for me
[11:31] <wido> running now and waiting if the scrub problem comes back
[11:40] * NightDog (~karl@52.84-48-58.nextgentel.com) Quit (Read error: Connection reset by peer)
[11:40] * NightDog (~karl@52.84-48-58.nextgentel.com) has joined #ceph
[11:52] * NightDog__ (~karl@52.84-48-58.nextgentel.com) has joined #ceph
[11:52] * NightDog (~karl@52.84-48-58.nextgentel.com) Quit (Read error: Connection reset by peer)
[12:48] * alexxy (~alexxy@79.173.81.171) Quit (Remote host closed the connection)
[12:54] * alexxy (~alexxy@79.173.81.171) has joined #ceph
[12:55] * fronlius (~fronlius@g231136055.adsl.alicedsl.de) has joined #ceph
[15:18] * fronlius (~fronlius@g231136055.adsl.alicedsl.de) Quit (Remote host closed the connection)
[15:18] * fronlius (~fronlius@g231136055.adsl.alicedsl.de) has joined #ceph
[15:46] * fronlius_ (~fronlius@g231136055.adsl.alicedsl.de) has joined #ceph
[15:46] * NightDog__ (~karl@52.84-48-58.nextgentel.com) Quit (Read error: Connection reset by peer)
[15:46] * NightDog__ (~karl@52.84-48-58.nextgentel.com) has joined #ceph
[15:47] * fronlius_ (~fronlius@g231136055.adsl.alicedsl.de) Quit (Remote host closed the connection)
[15:47] * fronlius_ (~fronlius@g231136055.adsl.alicedsl.de) has joined #ceph
[15:47] * fronlius (~fronlius@g231136055.adsl.alicedsl.de) Quit (Read error: Connection reset by peer)
[15:47] * fronlius_ is now known as fronlius
[16:06] * edwardw (~edwardw@8.19.33.115) has joined #ceph
[16:06] * NightDog__ (~karl@52.84-48-58.nextgentel.com) Quit (Read error: Connection reset by peer)
[16:06] * NightDog__ (~karl@52.84-48-58.nextgentel.com) has joined #ceph
[16:35] <todin> gregaf: after updating the whole cluster to current master, I don't have the pthread memory problem any longer, so far it runs stable
[16:38] * edwardw (~edwardw@8.19.33.115) Quit (Ping timeout: 480 seconds)
[16:41] * edwardw (~edwardw@8.19.33.115) has joined #ceph
[16:53] * edwardw (~edwardw@8.19.33.115) Quit (Ping timeout: 480 seconds)
[16:56] * MarkDude (~MT@208.88.9.226) has joined #ceph
[16:58] * NightDog__ (~karl@52.84-48-58.nextgentel.com) Quit (Read error: Connection reset by peer)
[16:58] * NightDog__ (~karl@52.84-48-58.nextgentel.com) has joined #ceph
[17:04] * votz (~votz@pool-108-52-122-97.phlapa.fios.verizon.net) Quit (Ping timeout: 480 seconds)
[17:25] * votz (~votz@pool-108-52-122-97.phlapa.fios.verizon.net) has joined #ceph
[17:39] * NightDog__ (~karl@52.84-48-58.nextgentel.com) Quit (Read error: Connection reset by peer)
[17:39] * NightDog__ (~karl@52.84-48-58.nextgentel.com) has joined #ceph
[18:06] * fronlius_ (~fronlius@g231136055.adsl.alicedsl.de) has joined #ceph
[18:13] * fronlius (~fronlius@g231136055.adsl.alicedsl.de) Quit (Ping timeout: 480 seconds)
[18:13] * fronlius_ is now known as fronlius
[18:14] * fronlius_ (~fronlius@e182094218.adsl.alicedsl.de) has joined #ceph
[18:14] * NightDog__ (~karl@52.84-48-58.nextgentel.com) Quit (Read error: Connection reset by peer)
[18:15] * NightDog__ (~karl@52.84-48-58.nextgentel.com) has joined #ceph
[18:21] * fronlius (~fronlius@g231136055.adsl.alicedsl.de) Quit (Ping timeout: 480 seconds)
[18:21] * fronlius_ is now known as fronlius
[18:37] * MarkDude (~MT@208.88.9.226) Quit (Quit: Leaving)
[18:43] * yehuda_hm (~yehuda@99-48-179-68.lightspeed.irvnca.sbcglobal.net) Quit (Ping timeout: 480 seconds)
[18:44] * mtk (~mtk@ool-44c35967.dyn.optonline.net) Quit (Remote host closed the connection)
[18:55] * NightDog__ (~karl@52.84-48-58.nextgentel.com) Quit (Read error: Connection reset by peer)
[18:55] * NightDog__ (~karl@52.84-48-58.nextgentel.com) has joined #ceph
[19:02] * NightDog__ (~karl@52.84-48-58.nextgentel.com) Quit (Quit: Leaving)
[19:02] * Nightdog (~karl@52.84-48-58.nextgentel.com) has joined #ceph
[19:28] * Tv__ (~Tv__@cpe-76-168-227-45.socal.res.rr.com) has joined #ceph
[19:28] * Nightdog (~karl@52.84-48-58.nextgentel.com) Quit (Read error: Connection reset by peer)
[19:28] * Nightdog (~karl@52.84-48-58.nextgentel.com) has joined #ceph
[19:31] * Nightdog (~karl@52.84-48-58.nextgentel.com) Quit (Read error: Connection reset by peer)
[19:31] * Nightdog (~karl@52.84-48-58.nextgentel.com) has joined #ceph
[19:47] * Nightdog (~karl@52.84-48-58.nextgentel.com) Quit (Read error: Connection reset by peer)
[19:47] * Nightdog (~karl@52.84-48-58.nextgentel.com) has joined #ceph
[20:04] * Nightdog (~karl@52.84-48-58.nextgentel.com) Quit (Read error: Connection reset by peer)
[20:05] * Nightdog (~karl@52.84-48-58.nextgentel.com) has joined #ceph
[20:36] * Nightdog (~karl@52.84-48-58.nextgentel.com) Quit (Read error: Connection reset by peer)
[20:37] * Nightdog (~karl@52.84-48-58.nextgentel.com) has joined #ceph
[20:40] * fronlius (~fronlius@e182094218.adsl.alicedsl.de) Quit (Quit: fronlius)
[21:33] * sjust (~sam@aon.hq.newdream.net) Quit (Ping timeout: 480 seconds)
[22:26] * Nightdog (~karl@52.84-48-58.nextgentel.com) Quit (Read error: Connection reset by peer)
[22:26] * Nightdog (~karl@52.84-48-58.nextgentel.com) has joined #ceph
[22:55] * Nightdog (~karl@52.84-48-58.nextgentel.com) Quit (Read error: Connection reset by peer)
[22:55] * Nightdog (~karl@52.84-48-58.nextgentel.com) has joined #ceph
[23:40] * Nightdog (~karl@52.84-48-58.nextgentel.com) Quit (Read error: Connection reset by peer)
[23:41] * Nightdog (~karl@52.84-48-58.nextgentel.com) has joined #ceph
[23:56] * lxo (~aoliva@lxo.user.oftc.net) has joined #ceph

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