#ceph IRC Log

Index

IRC Log for 2012-02-17

Timestamps are in GMT/BST.

[0:16] * fronlius (~fronlius@f054111204.adsl.alicedsl.de) Quit (Quit: fronlius)
[0:26] * jmlowe (~Adium@c-98-223-195-84.hsd1.in.comcast.net) has joined #ceph
[0:51] <darkfader> sagewk1: i just met a big time fan of yours, thomas uhl
[0:52] * adjohn (~adjohn@rackspacesf.static.monkeybrains.net) Quit (Remote host closed the connection)
[0:52] * adjohn (~adjohn@50.56.129.169) has joined #ceph
[1:00] * BManojlovic (~steki@212.200.241.85) Quit (Remote host closed the connection)
[1:29] * yoshi (~yoshi@p8031-ipngn2701marunouchi.tokyo.ocn.ne.jp) has joined #ceph
[1:37] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) has left #ceph
[1:41] * adamw (~AdamW@S010600226b8aa2ea.vc.shawcable.net) has joined #ceph
[1:42] <adamw> hi, folks. so it looks like ceph has a lot of issues compiling with gcc 4.7 and i don't see fixes in master...is anyone working on that? don't want to be painfully duplicating stuff
[1:55] * joao (~joao@89.181.154.123) Quit (Quit: joao)
[1:57] * linuxuser1357 (~linuxuser@50-82-41-66.client.mchsi.com) Quit (Quit: Leaving)
[2:14] * lofejndif (~lsqavnbok@57.Red-88-19-214.staticIP.rima-tde.net) Quit (Quit: Leaving)
[2:32] * chutzpah (~chutz@216.174.109.254) Quit (Quit: Leaving)
[2:38] * The_Bishop (~bishop@cable-89-16-138-109.cust.telecolumbus.net) Quit (Quit: Wer zum Teufel ist dieser Peer? Wenn ich den erwische dann werde ich ihm mal die Verbindung resetten!)
[3:00] * MarkDude (~MT@c-71-198-138-155.hsd1.ca.comcast.net) Quit (Quit: Leaving)
[3:39] * elder (~elder@c-71-195-31-37.hsd1.mn.comcast.net) Quit (Quit: Leaving)
[4:03] * adjohn (~adjohn@50.56.129.169) Quit (Quit: adjohn)
[4:40] * dmick (~dmick@aon.hq.newdream.net) Quit (Quit: Leaving.)
[4:51] * yoshi (~yoshi@p8031-ipngn2701marunouchi.tokyo.ocn.ne.jp) Quit (Remote host closed the connection)
[4:53] * yoshi (~yoshi@p8031-ipngn2701marunouchi.tokyo.ocn.ne.jp) has joined #ceph
[8:19] * adamw (~AdamW@S010600226b8aa2ea.vc.shawcable.net) Quit (Ping timeout: 480 seconds)
[8:30] * joao (~joao@89.181.147.200) has joined #ceph
[8:40] * gregorg (~Greg@78.155.152.6) has joined #ceph
[9:04] * joao (~joao@89.181.147.200) Quit (Quit: joao)
[9:18] * verwilst (~verwilst@d51A5B5DF.access.telenet.be) has joined #ceph
[9:35] * fronlius (~fronlius@testing78.jimdo-server.com) has joined #ceph
[9:42] * joao (~joao@193.136.122.17) has joined #ceph
[9:51] * MarkDude (~MT@c-71-198-138-155.hsd1.ca.comcast.net) has joined #ceph
[10:13] * fghaas (~florian@85-127-86-65.dynamic.xdsl-line.inode.at) has joined #ceph
[10:15] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) has joined #ceph
[10:29] * fghaas (~florian@85-127-86-65.dynamic.xdsl-line.inode.at) Quit (Ping timeout: 480 seconds)
[10:45] * fghaas (~florian@85-127-86-65.dynamic.xdsl-line.inode.at) has joined #ceph
[11:11] * MarkDude (~MT@c-71-198-138-155.hsd1.ca.comcast.net) Quit (Quit: Leaving)
[11:49] * adjohn (~adjohn@50-0-92-115.dsl.dynamic.sonic.net) has joined #ceph
[11:49] * adjohn (~adjohn@50-0-92-115.dsl.dynamic.sonic.net) Quit ()
[12:25] * nhorman (~nhorman@99-127-245-201.lightspeed.rlghnc.sbcglobal.net) has joined #ceph
[12:37] * yoshi (~yoshi@p8031-ipngn2701marunouchi.tokyo.ocn.ne.jp) Quit (Remote host closed the connection)
[12:42] * lofejndif (~lsqavnbok@57.Red-88-19-214.staticIP.rima-tde.net) has joined #ceph
[12:47] * ogelbukh (~weechat@nat3.4c.ru) has joined #ceph
[13:34] * aliguori (~anthony@cpe-70-123-132-139.austin.res.rr.com) has joined #ceph
[13:34] * aliguori_ (~anthony@cpe-70-123-132-139.austin.res.rr.com) has joined #ceph
[13:34] * aliguori_ (~anthony@cpe-70-123-132-139.austin.res.rr.com) Quit ()
[13:50] * lofejndif (~lsqavnbok@57.Red-88-19-214.staticIP.rima-tde.net) Quit (Ping timeout: 480 seconds)
[13:59] * lofejndif (~lsqavnbok@154.Red-83-52-212.dynamicIP.rima-tde.net) has joined #ceph
[14:07] * lofejndif (~lsqavnbok@154.Red-83-52-212.dynamicIP.rima-tde.net) Quit (Quit: Leaving)
[14:07] * jmlowe (~Adium@c-98-223-195-84.hsd1.in.comcast.net) Quit (Quit: Leaving.)
[14:12] * gohko (~gohko@natter.interq.or.jp) Quit (Ping timeout: 480 seconds)
[14:37] * The_Bishop (~bishop@e179006005.adsl.alicedsl.de) has joined #ceph
[14:38] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) Quit (Quit: Leaving.)
[15:48] * f4m8 is now known as f4m8_
[15:53] * The_Bishop (~bishop@e179006005.adsl.alicedsl.de) Quit (Ping timeout: 480 seconds)
[15:57] * The_Bishop (~bishop@e179006005.adsl.alicedsl.de) has joined #ceph
[16:06] * jmlowe (~Adium@129-79-195-139.dhcp-bl.indiana.edu) has joined #ceph
[16:18] * mtk (~mtk@ool-44c35967.dyn.optonline.net) Quit (Remote host closed the connection)
[16:29] * mtk (~mtk@ool-44c35967.dyn.optonline.net) has joined #ceph
[16:58] * gregorg (~Greg@78.155.152.6) Quit (Quit: Quitte)
[17:11] * Kioob`Taff (~plug-oliv@local.plusdinfo.com) Quit (Quit: Leaving.)
[17:20] * The_Bishop (~bishop@e179006005.adsl.alicedsl.de) Quit (Quit: Wer zum Teufel ist dieser Peer? Wenn ich den erwische dann werde ich ihm mal die Verbindung resetten!)
[17:23] * bchrisman (~Adium@c-76-103-130-94.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[17:44] * jmlowe (~Adium@129-79-195-139.dhcp-bl.indiana.edu) Quit (Quit: Leaving.)
[17:45] * aa (~aa@r200-40-114-26.ae-static.anteldata.net.uy) has joined #ceph
[17:52] * adamw (~AdamW@S010600226b8aa2ea.vc.shawcable.net) has joined #ceph
[18:20] * Tv|work (~Tv__@aon.hq.newdream.net) has joined #ceph
[18:20] * BManojlovic (~steki@212.200.243.83) has joined #ceph
[18:21] * jmlowe (~Adium@c-98-223-195-84.hsd1.in.comcast.net) has joined #ceph
[18:25] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) has joined #ceph
[18:30] * bchrisman (~Adium@108.60.121.114) has joined #ceph
[18:37] * nhorman (~nhorman@99-127-245-201.lightspeed.rlghnc.sbcglobal.net) Quit (Ping timeout: 480 seconds)
[18:39] * joao (~joao@193.136.122.17) Quit (Quit: joao)
[18:43] * aa (~aa@r200-40-114-26.ae-static.anteldata.net.uy) Quit (Quit: Konversation terminated!)
[18:44] * aa (~aa@r200-40-114-26.ae-static.anteldata.net.uy) has joined #ceph
[18:48] * eightyeight (~88@pthree.org) has left #ceph
[19:01] * joshd (~joshd@aon.hq.newdream.net) has joined #ceph
[19:03] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) Quit (Quit: Leaving.)
[19:03] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) has joined #ceph
[19:05] * joao (~joao@89.181.147.200) has joined #ceph
[19:05] * chutzpah (~chutz@216.174.109.254) has joined #ceph
[19:12] * adjohn (~adjohn@rackspacesf.static.monkeybrains.net) has joined #ceph
[19:15] * nhorman (~nhorman@99-127-245-201.lightspeed.rlghnc.sbcglobal.net) has joined #ceph
[19:17] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) Quit (Quit: Leaving.)
[19:34] <fghaas> is the "rbd create pool/image@snap" syntax still expected to be working? I recently tested this on Debian wheezy (0.40), and it would not accept that ... whereas the "rbd snap create" syntax worked just fine. am I doing something wrong?
[19:38] <fghaas> sorry, actually squeeze with a 3.2.0 kernel from backports and the ceph source page from wheezy rebuilt on squeeze. but I don't think that should make a difference
[19:45] * MarkDude (~MT@c-71-198-138-155.hsd1.ca.comcast.net) has joined #ceph
[19:51] <joshd> fghaas: it should be working
[19:53] <joshd> looks like our tests with the command line tool need updating to use both syntaxes
[19:54] <fghaas> possibly :)
[19:55] * adamw (~AdamW@S010600226b8aa2ea.vc.shawcable.net) Quit (Quit: Ex-Chat)
[19:56] * dmick (~dmick@aon.hq.newdream.net) has joined #ceph
[19:58] <joshd> oh, I didn't read closely enough
[19:58] <joshd> creating snapshots only ever worked with 'rbd snap create'
[19:59] * bchrisman (~Adium@108.60.121.114) Quit (Quit: Leaving.)
[19:59] <fghaas> To create a new snapshot:
[19:59] <fghaas> rbd create mypool/myimage@mysnap
[19:59] <fghaas> ^^ from the rbd man page
[20:00] <fghaas> if that never worked joshd, then the man page needs fixing :)
[20:03] <joshd> yeah, thanks for pointing that out
[20:16] * bchrisman (~Adium@108.60.121.114) has joined #ceph
[20:22] * lofejndif (~lsqavnbok@154.Red-83-52-212.dynamicIP.rima-tde.net) has joined #ceph
[20:29] <fghaas> joshd: so, would it be "rbd snap create mypool/myimage", or only "rbd -p mypool snap create myimage"?
[20:31] <joshd> both, but you need to specify the snapshot name too, which could be using either syntax (--snap or @snapname)
[20:33] <fghaas> "rbd snap create --snap=mysnap mypool/myimage"?
[20:33] <joshd> yeah, that will work
[20:34] <fghaas> oh, so what you're saying is just "rbd create mypool/myimage@mysnap" won't work, but "rbd snap create mypool/myimage@mysnap" will?
[20:34] <joshd> right
[20:35] <joshd> "rbd create" is intended for creating new images, and all the snapshot handling subcommands are prefixed by "snap"
[20:35] <fghaas> ok, trivial patch on the way :)
[20:36] <joshd> similarly, "rbd rm pool/image@snap" won't work for deleting snapshots, you need to use "rbd snap rm pool/image@snap"
[20:37] <fghaas> roger that
[20:43] <fghaas> ... and there's your patch. thanks for the explanation!
[20:44] <joshd> great, thanks!
[20:46] * fronlius (~fronlius@testing78.jimdo-server.com) Quit (Quit: fronlius)
[20:50] <joshd> mind if I add your signed-off-by and copy the changes into the rst man pages?
[20:52] <lxo> been looking into snapshotted files that aren't properly replicated across all replicas
[20:53] <fghaas> joshd: I don't mind the signed-off-by at all; did I stick my changes into the wrong locations? if so please educate me and I'll do better next time :)
[20:53] <lxo> the files that are missing from the <pg>_<snapid> directories are not really missing, they're present in the <pg>_head tree, just not hard-linked to the _<snapid> dir
[20:55] * adjohn (~adjohn@rackspacesf.static.monkeybrains.net) Quit (Quit: adjohn)
[20:55] <lxo> now, the odd thing is that all such links are present in 2 or 3 out of 3 replicas, never less than that, and I couldn't find a pattern to explain why some PGs have all the hardlink fully replicated, while others have it missing in one of the replicas
[20:56] <lxo> now, my question is whether these hard links are relevant at all, i.e., whether they're being missing or present could possibly lead to any data loss should the replicas that are missing the hardlinks go belly up
[20:56] <lxo> I mentioned before that scrub doesn't complain about the links being missing, and repair doesn't fix it
[20:57] * MarkDude (~MT@c-71-198-138-155.hsd1.ca.comcast.net) Quit (Read error: Connection reset by peer)
[20:57] <lxo> also, isn't it odd that a file whose snapshot id is 4 is present in all of _head, _4 and _2?
[20:59] * jmlowe (~Adium@c-98-223-195-84.hsd1.in.comcast.net) Quit (Ping timeout: 480 seconds)
[21:00] <joshd> fghaas: looks like the process isn't documented outside of a commit message, but our man pages are in docs/man/<section>/<rst_file>. the doc building script at admin/build-doc generates actual man pages under build-doc/output/man, whose contents can be copied into the repo's top level man directory
[21:00] <fghaas> ah, yeah I totally don't mind _not_ having to deal with raw nroff :)
[21:07] <joshd> btw, if you have suggestions for improving command line tool usage info/options, I'd be happy to hear them. I know the rbd tool in particular can be rather confusing
[21:08] <fghaas> er yes. oh, while you're at it: "--test-map-object" in osdmaptool is undocumented
[21:09] * u3q (~ben@uranus.tspigot.net) Quit (resistance.oftc.net synthon.oftc.net)
[21:09] * chutzpah (~chutz@216.174.109.254) Quit (resistance.oftc.net synthon.oftc.net)
[21:09] * Meths (rift@2.25.214.237) Quit (resistance.oftc.net synthon.oftc.net)
[21:09] * grape (~grape@216.24.166.226) Quit (resistance.oftc.net synthon.oftc.net)
[21:09] * acaos (~zac@209-99-103-42.fwd.datafoundry.com) Quit (resistance.oftc.net synthon.oftc.net)
[21:09] * guido (~guido@mx1.hannover.ccc.de) Quit (resistance.oftc.net synthon.oftc.net)
[21:09] * Azrael (~azrael@terra.negativeblue.com) Quit (resistance.oftc.net synthon.oftc.net)
[21:09] * nolan (~nolan@phong.sigbus.net) Quit (resistance.oftc.net synthon.oftc.net)
[21:09] * eternaleye___ (~eternaley@195.215.30.181) Quit (resistance.oftc.net synthon.oftc.net)
[21:10] * u3q (~ben@uranus.tspigot.net) has joined #ceph
[21:11] * chutzpah (~chutz@216.174.109.254) has joined #ceph
[21:11] * Meths (rift@2.25.214.237) has joined #ceph
[21:11] * Azrael (~azrael@terra.negativeblue.com) has joined #ceph
[21:11] * guido (~guido@mx1.hannover.ccc.de) has joined #ceph
[21:11] * grape (~grape@216.24.166.226) has joined #ceph
[21:11] * acaos (~zac@209-99-103-42.fwd.datafoundry.com) has joined #ceph
[21:11] * eternaleye___ (~eternaley@195.215.30.181) has joined #ceph
[21:11] * nolan (~nolan@phong.sigbus.net) has joined #ceph
[21:13] <joshd> indeed
[21:14] <joshd> I should probably mention we have tests that rely on usage info in src/test/cli
[21:15] <joshd> those are run by 'make check'
[21:15] <fghaas> does that patch look better?
[21:16] <joshd> yeah, looks great
[21:19] <fghaas> also, it's not obvious from the various *maptool man pages and --help outputs whether there's any way to feed them a *map on stdin
[21:20] <fghaas> as in "ceph osd getmap | osdmaptool <some magic option>"?
[21:21] <joshd> I don't think they read from stdin right now
[21:22] <joshd> there's more cleanup to make sure "ceph *" doesn't write to stdout as well
[21:22] <joshd> that is, it can write extra info that's not the osdmap
[21:23] <fghaas> ok
[21:23] <joshd> so for any kind of binary data you'll want to use "ceph -o <file>" currently
[21:23] <fghaas> could you add support for "ceph -o -"?
[21:24] <fghaas> (bikeshedding and super low priority, but would be neat)
[21:24] <joshd> I think it might already support that, actually
[21:24] * verwilst (~verwilst@d51A5B5DF.access.telenet.be) Quit (Quit: Ex-Chat)
[21:25] <joshd> it goes to stdout by default, it's just that there's some cleanup to make sure any extraneous stuff goes to stderr instead
[21:28] <fghaas> k. and now that you already asked about command line tools and docs (regretting that yet? :) ), the ceph man page could really use an exhaustive list of subcommands
[21:31] <fghaas> or, this would be a really awesome touch, you could implement "ceph <subsystem> help" which you may then even combine with clever scripting trickery to generate your man pages
[21:31] <fghaas> so you don't have to maintain the same information in two places in the tree
[21:31] <joshd> yeah.... they should definitely be documented. the best there is right now is http://ceph.newdream.net/docs/latest/control/, but it's not complete or very descriptive yet
[21:34] <fghaas> yup, and the man pages still has the old url, http://ceph.newdream.net/wiki/Monitor_commands :)
[21:38] <joshd> we're in the process of hiring a tech writer, so the docs will improve (and be more up to date) soon
[21:39] <fghaas> good luck finding one :)
[21:42] <iggy> good luck finding one that doesn't walk in front of a bus 3 months in
[21:46] <fghaas> I was referring to the general truth that good tech writers are few and far between
[21:50] <iggy> because they all walk in front of busses
[21:54] * adjohn (~adjohn@rackspacesf.static.monkeybrains.net) has joined #ceph
[21:58] * adjohn (~adjohn@rackspacesf.static.monkeybrains.net) Quit ()
[22:03] <joao> lol
[22:05] * nhorman (~nhorman@99-127-245-201.lightspeed.rlghnc.sbcglobal.net) Quit (Quit: Leaving)
[22:09] <dmick> clearly we need anti-bus laws
[22:12] <Tv|work> dmick: you mean anti-wheel laws
[22:13] <Tv|work> as per Doctorow
[22:17] * verwilst (~verwilst@d51A5B5DF.access.telenet.be) has joined #ceph
[22:17] * adjohn (~adjohn@rackspacesf.static.monkeybrains.net) has joined #ceph
[22:18] <lxo> wouldn't it be much easier to just forbid tech writers from being good? :-D
[22:19] <Tv|work> there's nothing worse than having actually worked with a *good* technical writer, once in your life
[22:19] <Tv|work> because then you know what's possible
[22:43] * adjohn (~adjohn@rackspacesf.static.monkeybrains.net) Quit (Ping timeout: 480 seconds)
[22:43] * adjohn (~adjohn@rackspacesf.static.monkeybrains.net) has joined #ceph
[22:52] * joao (~joao@89.181.147.200) Quit (Quit: joao)
[23:11] * The_Bishop (~bishop@cable-89-16-138-109.cust.telecolumbus.net) has joined #ceph
[23:13] * fghaas (~florian@85-127-86-65.dynamic.xdsl-line.inode.at) Quit (Ping timeout: 480 seconds)
[23:38] * adjohn is now known as Guest2979
[23:38] * _adjohn (~adjohn@rackspacesf.static.monkeybrains.net) has joined #ceph
[23:38] * _adjohn is now known as adjohn
[23:39] * Guest2979 (~adjohn@rackspacesf.static.monkeybrains.net) Quit (Ping timeout: 480 seconds)
[23:41] * adjohn is now known as Guest2980
[23:41] * _adjohn (~adjohn@rackspacesf.static.monkeybrains.net) has joined #ceph
[23:41] * _adjohn is now known as adjohn
[23:46] * Guest2980 (~adjohn@rackspacesf.static.monkeybrains.net) Quit (Ping timeout: 480 seconds)
[23:55] * MarkDude (~MT@c-71-198-138-155.hsd1.ca.comcast.net) has joined #ceph

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