#ceph IRC Log

Index

IRC Log for 2011-06-01

Timestamps are in GMT/BST.

[0:33] * aliguori_ (~anthony@32.97.110.64) Quit (Quit: Ex-Chat)
[0:35] * MarkN (~nathan@59.167.240.178) has joined #ceph
[0:49] <damoxc> are there any docs on where to begin debugging mount errors?
[0:52] <joshd> damoxc: there are a couple things to check listed on the wiki: http://ceph.newdream.net/wiki/Mounting_the_file_system
[0:53] <damoxc> I have 8 crashed+peering pgs
[0:53] <damoxc> would that affect things?
[0:53] <joshd> I don't think that should prevent you from mounting
[0:54] <damoxc> maybe it's the client
[0:54] <joshd> is there anything in your monitor's log when you try to mount?
[0:54] * MarkN (~nathan@59.167.240.178) has left #ceph
[0:54] <gregaf> the PGs might be preventing the MDS from starting up, if it can't read from them
[0:55] <damoxc> mds e28213: 3/3/3 up {0=0=up:replay,1=2=up:replay,2=1=up:replay}
[0:55] <damoxc> that's my mds line
[0:57] <gregaf> damoxc: yeah, your MDSes aren't active so your client can't mount yet
[0:57] <gregaf> the actual problem is whatever's keeping your OSDs from going active
[0:58] <damoxc> is there a way to output the broken pgs?
[0:58] <gregaf> ceph pg dump -o -
[0:58] <gregaf> I believe
[0:58] <damoxc> ah yes, cool thanks
[0:59] <gregaf> but you'll want sjust or joshd to help you debug that
[0:59] <gregaf> (sjust sjust sjust! pay attention to the support channels while you're on support watch!)
[0:59] <sjust> yes, here
[0:59] <gregaf> :p
[1:00] <sjust> damoxc: what do you get if you run ceph pg dump -o - | grep -v 'active+clean'
[1:00] <damoxc> sjust: http://dpaste.com/548931/
[1:01] * joshd (~joshd@ip-66-33-206-8.dreamhost.com) Quit (Quit: Leaving.)
[1:01] <sjust> how about ceph osd dump -o -?
[1:02] <damoxc> http://dpaste.com/548932/
[1:03] <sjust> several pg's are crashed+peering
[1:03] <sjust> and stuck
[1:03] <sjust> hmm
[1:03] <sjust> can you post logs?
[1:03] <sjust> from osd4
[1:03] <damoxc> sure thing
[1:04] <damoxc> how much do you want?
[1:04] <sjust> do you know how long it has been misbehaving?
[1:05] <damoxc> ever since I upgraded and rebooted the cluster
[1:05] <damoxc> quite a few hours
[1:05] <damoxc> there's some tracebacks in there
[1:06] <sjust> oh, it restarted?
[1:07] <sjust> give me 8 hours?
[1:07] <sjust> I mean, 8 hours of log should work
[1:07] <damoxc> okay that's the whole log near enough, starts at 16:16, ends now at 00:07
[1:07] <damoxc> 136mb
[1:08] <sjust> ah, that'll work
[1:08] <sjust> I was afraid it would be many GB :)
[1:08] <damoxc> ah good :-)
[1:11] <damoxc> here it is, http://damoxc.net/osd.4.log
[1:14] <sjust> damoxc: lookin
[1:15] * greglap (~Adium@ip-66-33-206-8.dreamhost.com) Quit (Quit: Leaving.)
[1:16] <sjust> damoxc: what was the osd log level?
[1:16] <damoxc> sjust: default I think
[1:16] <damoxc> I have debug ms = 1 in [global], can't remember what that does though
[1:17] <damoxc> message traffic by the looks of things
[1:17] <sjust> ms is the messenger debugging
[1:18] <sjust> ah, osd debugging defaults to 0, I'm afraid
[1:18] <damoxc> ah okay
[1:18] <sjust> debug osd = 25
[1:18] <sjust> try adding that to the osd4 section of your ceph.conf
[1:18] <sjust> and restart osd4
[1:19] <sjust> unfortunately, that might just make the problem go away, but we may get lucky and catch it with logging on
[1:20] <damoxc> okay just restarted
[1:20] <damoxc> they are currently 8 active+clean+replay
[1:21] <damoxc> and now they are clean
[1:21] <sjust> none marked crashed+peering?
[1:21] <damoxc> no not anymore
[1:21] <sjust> :(, oh well
[1:22] <sjust> debug 25 on the osd is going to generate a log of logging, so you may want to turn it down to 10
[1:22] <damoxc> okay will do
[1:22] <sjust> are the MDS's going up?
[1:22] <damoxc> still in replay
[1:24] <sjust> all pg's are active+clean?
[1:24] <damoxc> 1 active+clean+scrubbing
[1:24] <damoxc> rest are all active+clea
[1:24] <sjust> that should be harmless
[1:24] <damoxc> *active+clean
[1:25] <sjust> hmm, the mds should be able to read now
[1:25] <sjust> greg should be back in the channel in a bit, I'll need to ask him about the MDS side of things
[1:26] <damoxc> okay, thanks for all your help, much appreciated
[1:26] <sjust> damoxc: thanks for your time and testing :)
[1:26] <damoxc> could it be anything to do with multi mds being enabled?
[1:26] <sjust> damoxc: yes
[1:26] <sjust> damoxc: you could try restarting the MDS daemons
[1:29] <damoxc> hmm that hasn't seemed to have helped
[1:33] <damoxc> will gregaf be long? I should probably be heading to bed soon so will have to leave it until tomorrow evening if so
[1:35] * verwilst (~verwilst@dD57671BC.access.telenet.be) Quit (Quit: Ex-Chat)
[1:48] * Dantman (~dantman@S0106001731dfdb56.vs.shawcable.net) Quit (Remote host closed the connection)
[1:49] <sjust> damoxc: sorry, got distracted
[1:49] <sjust> damoxc: I guess he is out for the day, sorry about that
[1:50] <sjust> damoxc: If you get a chance to post one/all of the MDS logs, I can show them to Greg or Sage tommorrow
[1:52] <damoxc> sjust: okay thanks, I'll set the logging up to 20/25 tomorrow morning and then post the logs up tomorrow evening (BST)
[1:52] * morse (~morse@supercomputing.univpm.it) Quit (Remote host closed the connection)
[1:52] * morse (~morse@supercomputing.univpm.it) has joined #ceph
[1:53] <sjust> damoxc: ok
[1:54] * nms (martin@sexyba.be) Quit (Remote host closed the connection)
[1:54] * nms (martin@sexyba.be) has joined #ceph
[2:05] * s15y (~s15y@sac91-2-88-163-166-69.fbx.proxad.net) Quit (Ping timeout: 480 seconds)
[2:15] * s15y (~s15y@sac91-2-88-163-166-69.fbx.proxad.net) has joined #ceph
[2:17] * yoshi (~yoshi@p24092-ipngn1301marunouchi.tokyo.ocn.ne.jp) has joined #ceph
[2:18] <yoshi> Hi. Anyone from the dev team around?
[2:25] * Tv (~Tv|work@ip-66-33-206-8.dreamhost.com) Quit (Ping timeout: 480 seconds)
[2:29] * bchrisman (~Adium@70-35-37-146.static.wiline.com) Quit (Quit: Leaving.)
[2:34] * pruby (~tim@leibniz.catalyst.net.nz) Quit (Ping timeout: 480 seconds)
[2:37] * joshd (~jdurgin@75.28.69.238) has joined #ceph
[2:40] * greglap (~Adium@cpe-76-170-84-245.socal.res.rr.com) has joined #ceph
[2:42] * alexxy (~alexxy@79.173.81.171) Quit (Ping timeout: 480 seconds)
[3:02] * Dantman (~dantman@S0106001731dfdb56.vs.shawcable.net) has joined #ceph
[3:14] * joshd (~jdurgin@75.28.69.238) Quit (Read error: Connection reset by peer)
[3:22] * joshd (~jdurgin@adsl-75-28-69-238.dsl.irvnca.sbcglobal.net) has joined #ceph
[3:31] <joshd> yoshi: thanks for filing a bug
[3:31] <yoshi> joshd: Sure. Hope that helps. The version we're testing could be a bit old.
[3:32] <joshd> yoshi: did anything in your environment change, or was the monitor suddenly not recognized by the osd?
[3:32] <yoshi> joshd: It happened suddenly to me. Not configuration change to hardware or software.
[3:33] <yoshi> joshd: Not -> No
[3:33] <joshd> yoshi: and all the monitors are still up?
[3:35] <joshd> yoshi: if you've still got it running, what's the output of 'ceph -s' and 'ceph osd dump -o -'?
[3:35] <yoshi> joshd: Yeah. That's something weird. The cmon and cosd look running fine, but when you command rbd list or similar commands, it doesn't respond. I've started producing debug messages, but the problem doesn't show up.
[3:35] <yoshi> joshd: Let me try. Just a sec.
[3:37] <yoshi> $ ceph -s
[3:37] <yoshi> 2011-06-01 10:36:15.728352 pg v117387: 264 pgs: 264 active+clean+degraded; 5752 MB data, 15188 MB used, 131 GB / 146 GB avail; 1561/3122 degraded (50.000%)
[3:37] <yoshi> 2011-06-01 10:36:15.728848 mds e10: 1/1/1 up {0=up:replay}
[3:37] <yoshi> 2011-06-01 10:36:15.728873 osd e22: 1 osds: 1 up, 1 in
[3:37] <yoshi> 2011-06-01 10:36:15.728906 log 2011-05-31 10:46:00.592870 mon0 192.168.100.7:6789/0 2 : [INF] osd0 192.168.100.7:6801/11271 boot
[3:37] <yoshi> 2011-06-01 10:36:15.728934 class rbd (v1.3 [x86-64])
[3:37] <yoshi> 2011-06-01 10:36:15.728946 mon e1: 1 mons at {0=192.168.100.7:6789/0}
[3:37] <yoshi> joshd: Oops. A bit messy.
[3:38] <joshd> np, some kind of pastebin's probably better for this stuff
[3:39] <joshd> yoshi: do you have anything in the monitor log?
[3:39] <yoshi> Right. how about this. https://gist.github.com/094501455e96675a75be
[3:40] <yoshi> joshd: For 'ceph osd dump -o -' https://gist.github.com/5f635e49f01d5ce75454
[3:40] <joshd> yoshi: ah, the osd is stuck in the boot state
[3:41] <yoshi> joshd: I see. One thing I noticed is that the journal could be the problem? I'm using on XFS, and using writeahead intentionally.
[3:43] <joshd> yoshi: I'm not aware of many people running it on XFS, but I don't think it should be a problem
[3:43] <yoshi> joshd: OK. Here is the log for mon just in case. https://gist.github.com/71b8b824a369ce6e24d3
[3:45] <joshd> yoshi: in your ceph.conf, could you add 'debug osd 25' to the [osd.0] section and restart the osd daemon?
[3:45] <yoshi> joshd: Sure. Hold on.
[3:46] <joshd> yoshi: unfortunately our logging is pretty verbose, so it's hard to get the relevant bits from just the last few lines
[3:47] <yoshi> joshd: Yeah. It's showing up tuns of messages. Do you have any keyword to grep?
[3:47] <joshd> yoshi: not really sure what the error will be in this case
[3:48] <yoshi> joshd: All right. Let me dump a bit larger.
[3:51] * pruby (~tim@leibniz.catalyst.net.nz) has joined #ceph
[3:51] <yoshi> joshd: There you go. https://gist.github.com/c9d36daf6f3a800ffce7
[3:54] <joshd> yoshi: it looks like the osd started fine, the problem may be with the client trying to use rbd
[3:55] <yoshi> joshd: Hmm. All the daemons and the client are in the same single node. What could be the problem?
[3:55] <joshd> yoshi: do you have authentication enabled?
[3:55] <yoshi> joshd: None. I disable authentication at ceph.conf.
[3:57] <joshd> yoshi: can you add 'debug ms = 5' and 'debug monc = 10' to the global section of your ceph.conf, and then try 'rbd ls'?
[3:57] * ethanr (~ethanr@66.220.144.74) has joined #ceph
[3:57] <yoshi> joshd: Cool. Just a sec.
[3:58] <yoshi> joshd: It seems I've already set, 'debug ms = 1' and 'debug mon = 20'. Should I still need to add 'debug monc'?
[3:59] <joshd> yoshi: yeah, it stands for monclient
[4:00] <yoshi> joshd: Got it.
[4:01] <ethanr> anyone using ceph for real data?
[4:02] <yoshi> joshd: Here is the log for mon. The client is logged. https://gist.github.com/6ef08542fcfc7cf173e2
[4:02] * bchrisman (~Adium@c-98-207-207-62.hsd1.ca.comcast.net) has joined #ceph
[4:07] * djlee_ (~dlee064@des152.esc.auckland.ac.nz) has joined #ceph
[4:13] <joshd> ethanr: ceph still isn't quite ready for production
[4:14] <yoshi> joshd: Looks a bit uninteresting. https://gist.github.com/2acecdb4d382dcb174a0
[4:14] <joshd> ethanr: sage gave a nice summary of the present status here: http://www.mail-archive.com/ceph-devel@vger.kernel.org/msg02635.html
[4:14] * djlee__ (~dlee064@des152.esc.auckland.ac.nz) Quit (Ping timeout: 480 seconds)
[4:15] * djlee (~dlee064@des152.esc.auckland.ac.nz) has joined #ceph
[4:15] <joshd> yoshi: grep -A 30 client1721501
[4:16] <ethanr> thanks joshd
[4:16] <ethanr> joshd how many ppl are workign on it atm ?
[4:16] <ethanr> oh 7 ppl
[4:16] <ethanr> =D
[4:17] <joshd> ethanr: yup
[4:17] <yoshi> joshd: Sure. https://gist.github.com/e7d9a2557e639ab99fa0
[4:19] <joshd> yoshi:hmm, didn't quite capture what I was looking for
[4:20] <joshd> yoshi: there's an interesting part between 11:00:29.000224 and 11:00:30.963324
[4:20] <yoshi> joshd: Let me take a look.
[4:21] * djlee_ (~dlee064@des152.esc.auckland.ac.nz) Quit (Ping timeout: 480 seconds)
[4:22] <joshd> yoshi: if you see any negative numbers, they're probably error codes - look for things after any lines with rbd in them
[4:26] <yoshi> joshd: Although I don't see negative numbers at a first glance, it seems 'pg 3.1c doesn't exist' could be the problem? https://gist.github.com/92dbfb454b7396a4e5d2
[4:30] <joshd> yoshi: I don't see pg 3.1c in that gist, is that where the rbd header is?
[4:30] <joshd> yoshi: if so, that could be the problem
[4:31] <joshd> yoshi: I've got to go, but hopefully we can figure this out tomorrow
[4:31] <yoshi> joshd: Sure. Last shot. How to read rbd header?
[4:32] <joshd> yoshi: you mean the log message, or the actual thing on-disk?
[4:32] <yoshi> joshd: On-dist. I though I can check it.
[4:32] <yoshi> joshd: cat file?
[4:33] <yoshi> joshd: though -> thought
[4:33] <joshd> yoshi: yeah, it's binary though, and we don't have a good tool for viewing it
[4:33] <joshd> other than rbd info :)
[4:33] <yoshi> joshd: I see. Then it's OK. I'll email the correct gist.
[4:34] <joshd> yoshi: if you could upload the whole log somewhere that'd be great, otherwise just get the parts after mentions of rbd
[4:35] <yoshi> joshd: Just in case if you're still around. https://gist.github.com/a94dc905822776034289
[4:36] <joshd> yoshi: that looks like it is the problem, but I'll have to check tomorrow
[4:36] <yoshi> joshd: NP. See you soon.
[4:36] <joshd> yoshi: see you
[4:36] * joshd (~jdurgin@adsl-75-28-69-238.dsl.irvnca.sbcglobal.net) Quit (Quit: Leaving.)
[4:57] <ethanr> joshd how far off do you think you guys are for prod data?
[4:57] <ethanr> and what is the instability
[5:02] <ethanr> meh
[5:07] <ajm> "http://www.mail-archive.com/ceph-devel@vger.kernel.org/msg02635.html" says "a few months"
[5:10] <ethanr> w00t
[5:13] * yehuda_hm (~yehuda@99-48-179-68.lightspeed.irvnca.sbcglobal.net) has joined #ceph
[5:15] * mtk (~mtk@ool-182c8e6c.dyn.optonline.net) Quit (Ping timeout: 480 seconds)
[5:19] * mtk (~mtk@ool-182c8e6c.dyn.optonline.net) has joined #ceph
[5:30] * djlee_ (~dlee064@des152.esc.auckland.ac.nz) has joined #ceph
[5:37] * djlee (~dlee064@des152.esc.auckland.ac.nz) Quit (Ping timeout: 480 seconds)
[6:10] * djlee__ (~dlee064@des152.esc.auckland.ac.nz) has joined #ceph
[6:18] * djlee_ (~dlee064@des152.esc.auckland.ac.nz) Quit (Ping timeout: 480 seconds)
[6:29] * votz (~votz@dhcp0020.grt.resnet.group.UPENN.EDU) Quit (Quit: Leaving)
[6:39] * votz (~votz@dhcp0020.grt.resnet.group.upenn.edu) has joined #ceph
[7:08] * neurodrone (~neurodron@cpe-76-180-162-12.buffalo.res.rr.com) Quit (Quit: zzZZZZzz)
[7:13] * ethanr (~ethanr@66.220.144.74) Quit (Quit: Leaving)
[7:18] * alexxy (~alexxy@79.173.81.171) has joined #ceph
[7:18] * Yulya_ (~Yu1ya_@ip-95-220-185-194.bb.netbynet.ru) has joined #ceph
[7:19] * DanielFriesen (~dantman@S0106001731dfdb56.vs.shawcable.net) has joined #ceph
[7:24] * Yulya__ (~Yu1ya_@ip-95-220-146-178.bb.netbynet.ru) Quit (Ping timeout: 480 seconds)
[7:26] * Dantman (~dantman@S0106001731dfdb56.vs.shawcable.net) Quit (Ping timeout: 480 seconds)
[8:55] * failbaitr (~innerheig@62.212.76.29) has joined #ceph
[9:22] * allsystemsarego (~allsystem@188.27.167.240) has joined #ceph
[10:34] * Juul_ (~Juul@c-76-21-88-119.hsd1.ca.comcast.net) has joined #ceph
[11:09] * Juul_ (~Juul@c-76-21-88-119.hsd1.ca.comcast.net) Quit (Quit: Leaving)
[11:33] * f4m8_ (~f4m8@lug-owl.de) Quit (Quit: leaving)
[11:33] * f4m8 (~f4m8@lug-owl.de) has joined #ceph
[12:14] * yoshi (~yoshi@p24092-ipngn1301marunouchi.tokyo.ocn.ne.jp) Quit (Remote host closed the connection)
[12:57] * jbd (~jbd@ks305592.kimsufi.com) has left #ceph
[13:01] * mtk (~mtk@ool-182c8e6c.dyn.optonline.net) Quit (Ping timeout: 482 seconds)
[13:06] * mtk (~mtk@ool-182c8e6c.dyn.optonline.net) has joined #ceph
[14:05] * jbd (~jbd@ks305592.kimsufi.com) has joined #ceph
[14:12] * aliguori (~anthony@cpe-70-123-132-139.austin.res.rr.com) has joined #ceph
[14:59] * artemgr (~chatzilla@178.140.133.5) has joined #ceph
[14:59] * neurodrone (~neurodron@cpe-76-180-162-12.buffalo.res.rr.com) has joined #ceph
[15:02] <artemgr> Hi. I have a weird issue here with librados IoCtx::read: "string value; bl.copy (0, bl.length(), value); return value;" throws /ceph::buffer::malformed_input, buffer::malformed_input: embedded NULL in string!/, whereas "return string (bl.c_str(), bl.length());" at the same place works okay.
[15:08] * artemgr (~chatzilla@178.140.133.5) Quit (Remote host closed the connection)
[15:11] * stingray (~stingray@stingr.net) has joined #ceph
[15:11] * stingray is now known as failboat
[15:11] * artemgr (~chatzilla@178.140.133.5) has joined #ceph
[15:31] * yoshi (~yoshi@KD027091032046.ppp-bb.dion.ne.jp) has joined #ceph
[15:48] * squig (~bendeluca@soho-94-143-249-50.sohonet.co.uk) has joined #ceph
[16:03] * iggy (~iggy@theiggy.com) Quit (Read error: Operation timed out)
[16:04] * eternaleye_ (~eternaley@195.215.30.181) has joined #ceph
[16:04] * darkfader (~floh@188.40.175.2) has joined #ceph
[16:05] * jeffhung_ (~jeffhung@60-250-103-120.HINET-IP.hinet.net) has joined #ceph
[16:06] * lidongyang_ (~lidongyan@222.126.194.154) has joined #ceph
[16:06] * jeffhung (~jeffhung@60-250-103-120.HINET-IP.hinet.net) Quit (Read error: Connection reset by peer)
[16:06] * darkfaded (~floh@188.40.175.2) Quit (Read error: Connection reset by peer)
[16:06] * eternaleye (~eternaley@195.215.30.181) Quit (Read error: Connection reset by peer)
[16:06] * lidongyang (~lidongyan@222.126.194.154) Quit (Read error: Connection reset by peer)
[16:07] * iggy (~iggy@theiggy.com) has joined #ceph
[16:18] * yoshi (~yoshi@KD027091032046.ppp-bb.dion.ne.jp) Quit (Remote host closed the connection)
[17:03] * aliguori (~anthony@cpe-70-123-132-139.austin.res.rr.com) Quit (Quit: Ex-Chat)
[17:05] * aliguori (~anthony@cpe-70-123-132-139.austin.res.rr.com) has joined #ceph
[17:35] * Tv (~Tv|work@ip-66-33-206-8.dreamhost.com) has joined #ceph
[17:38] * greglap (~Adium@cpe-76-170-84-245.socal.res.rr.com) Quit (Quit: Leaving.)
[17:45] * bchrisman (~Adium@c-98-207-207-62.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[17:53] * greglap (~Adium@mobile-198-228-209-203.mycingular.net) has joined #ceph
[18:03] <greglap> artemgr: I think the exception is thrown if you're trying to copy non-string data into a string
[18:03] <greglap> in this case your data contains 0-valued memory and strings can't deal with that
[18:03] * yehuda_hm (~yehuda@99-48-179-68.lightspeed.irvnca.sbcglobal.net) Quit (Ping timeout: 480 seconds)
[18:04] <greglap> sagewk1 would know better, though
[18:06] <artemgr> greglap: The thing is, I'm using std::string for 0-valued data for years. AFAIK, std::string should be safe.
[18:07] <artemgr> greglap: I used project Voldemort before moving to Ceph, and sure thing it uses std::string to handle keys and values, the same ones I now store in Ceph.
[18:08] <artemgr> greglap: (The C++ client of Project Voldemort, that is).
[18:09] * greglap1 (~Adium@mobile-198-228-210-059.mycingular.net) has joined #ceph
[18:10] <Tv> yeah afaik std::string is length-prefixed, not nil-terminated
[18:11] <Tv> artemgr: what was the exception (sorry i joined in the middle)
[18:12] <artemgr> Tv: let me repeat the original message..
[18:12] <artemgr> I have a weird issue here with librados IoCtx::read: "string value; bl.copy (0, bl.length(), value); return value;" throws /ceph::buffer::malformed_input, buffer::malformed_input: embedded NULL in string!/, whereas "return string (bl.c_str(), bl.length());" at the same place works okay.
[18:13] * greglap (~Adium@mobile-198-228-209-203.mycingular.net) Quit (Ping timeout: 480 seconds)
[18:14] <Tv> artemgr: yeah that's an explicit check in our "bufferlist" library, and it really doesn't seem necessary when dest is an std::string
[18:14] <Tv> but i'm afraid of bufferlist ;)
[18:15] <Tv> oh funny the variant where dest is a char* doesn't seem to have that check
[18:15] <greglap1> they might just be backwards
[18:18] <artemgr> tv: well, IoCtx::read only accepts a bufferlist, and its supposed to return binary data there. ; )
[18:18] <Tv> yeah
[18:18] <Tv> it's a bug fer sure
[18:19] <Tv> i have a branch with the obvious change, poking at it with a stick..
[18:21] <artemgr> thanks. i wonder why it didn't come up in your unit tests etc. bufferlist::c_str used more often than bufferlist::copy?
[18:22] <artemgr> i was thinking that this path (e.g. read, copy into std::string) should be pretty heavily used somewhere..
[18:23] <sagewk1> i think the check was added because we were getting null values in strings when we weren't supposed to. that bug hasn't resurfaced since, though. should be safe to remove the check
[18:23] <greglap1> we don't use strings too often internally, I only see that version of copy called in one place that probably doesn't use zero-valued data
[18:23] <artemgr> i see
[18:24] <Tv> sounds like a rados test for binary keys would be good..
[18:33] * bchrisman (~Adium@70-35-37-146.static.wiline.com) has joined #ceph
[18:39] <Tv> artemgr: oh wait what did you do to trigger that?
[18:40] <Tv> i find it hard to believe libradospp wouldn't support binary *data*
[18:40] <greglap1> Tv: I think he's explicitly doing the copy himself...
[18:41] <Tv> so ::read reads into bl, he tries to copy out of it? that would make sense
[18:41] <greglap1> yeah, with the bufferlist copy function
[18:41] <Tv> yeah
[18:41] <greglap1> the only other place it's used is in one of Sam's(?) newish tests
[18:41] <greglap1> that particular version, I mean
[18:41] <artemgr> tv: http://paste.pocoo.org/show/398991/
[18:41] <greglap1> train's in though, be back soon!
[18:42] <Tv> artemgr: ah yes; thanks
[18:42] * greglap1 (~Adium@mobile-198-228-210-059.mycingular.net) Quit (Read error: Connection reset by peer)
[18:43] <Tv> pushed branch bufferlist-copy-nil with the trivial bugfix
[18:43] <sagewk1> tv: looks ok to me, wanna put it in the next branch?
[18:44] <Tv> sure
[18:45] <Tv> done
[18:56] <artemgr> that reminds me... Ceph is not as exception-less as it is claimed there: http://marc.info/?l=ceph-devel&m=130690354923592&w=2
[18:57] * joshd (~joshd@ip-66-33-206-8.dreamhost.com) has joined #ceph
[18:58] <Tv> artemgr: that message was written after a conversation we had; the conversation made clear that bufferlist is the exception, and with a decent reason
[18:58] * greglap (~Adium@ip-66-33-206-8.dreamhost.com) has joined #ceph
[19:19] * artemgr (~chatzilla@178.140.133.5) Quit (Quit: ChatZilla 0.9.87 [Firefox 4.0.1/20110413222027])
[19:25] <Tv> so what's the story with 10:30 vs 11 meeting etc?
[19:25] <Tv> sagewk1: ?
[19:31] <gregaf> he's in a meeting again, he was hoping to be done but I guess he's not
[19:31] <gregaf> *grrr*
[19:53] <Tv> ok so now the 11am meeting is clearly defined as *hosting* dev
[19:57] <gregaf> oh, yeah, that got cleared up last week I thought
[19:57] <gregaf> although no idea what we're doing on friday
[20:01] <Tv> heh
[20:16] <sagewk1> tv: basically non-ceph.. so hosting+emerging, so for us just the dho overlap.
[22:07] <djlee__> guys, what's the exact benefit of the actual size of journal setup? e.g., 1GB, 2GB, other than the possible amount of recovery log..?
[22:08] * alexxy (~alexxy@79.173.81.171) Quit (Ping timeout: 480 seconds)
[22:09] * djlee__ (~dlee064@des152.esc.auckland.ac.nz) Quit (Quit: Ex-Chat)
[22:09] * eternaleye_ is now known as eternaleye
[22:10] * djlee (~dlee064@des152.esc.auckland.ac.nz) has joined #ceph
[22:13] * djlee (~dlee064@des152.esc.auckland.ac.nz) Quit ()
[22:14] * alexxy (~alexxy@79.173.81.171) has joined #ceph
[22:15] * djlee (~dlee064@des152.esc.auckland.ac.nz) has joined #ceph
[23:10] * ghaskins (~ghaskins@66-189-113-47.dhcp.oxfr.ma.charter.com) Quit (Quit: Leaving)
[23:45] * allsystemsarego (~allsystem@188.27.167.240) Quit (Quit: Leaving)

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