#ceph IRC Log


IRC Log for 2012-03-08

Timestamps are in GMT/BST.

[0:12] * lofejndif (~lsqavnbok@82VAAB9O9.tor-irc.dnsbl.oftc.net) Quit (Quit: Leaving)
[0:19] * lofejndif (~lsqavnbok@214.Red-83-43-124.dynamicIP.rima-tde.net) has joined #ceph
[0:22] * jmlowe (~Adium@c-71-201-31-207.hsd1.in.comcast.net) Quit (Quit: Leaving.)
[0:32] * The_Bishop (~bishop@178-17-163-220.static-host.net) has joined #ceph
[0:41] * Jaykra (~Jamie@64-126-89-248.dyn.everestkc.net) Quit (Quit: Leaving.)
[0:44] <Tv|work> why must booting PCs be this much of a pain
[0:45] * jluis (~JL@ace.ops.newdream.net) has joined #ceph
[0:45] * jluis (~JL@ace.ops.newdream.net) Quit (Remote host closed the connection)
[0:45] * tnt_ (~tnt@37.191-67-87.adsl-dyn.isp.belgacom.be) Quit (Ping timeout: 480 seconds)
[0:47] * joao (~JL@ace.ops.newdream.net) Quit (Quit: Leaving)
[0:51] <sagewk> sjusthm: that's strange.. i haven't seen it at all (w/ the yaml task i sent you yesterday)
[1:05] <sjusthm> sagewk: that one had btrfs enabled, right?
[1:07] * yoshi (~yoshi@p8031-ipngn2701marunouchi.tokyo.ocn.ne.jp) has joined #ceph
[1:08] <sagewk> oh, right. ;)
[1:13] * mtk (~mtk@ool-44c35967.dyn.optonline.net) Quit (Remote host closed the connection)
[1:16] * BManojlovic (~steki@ Quit (Remote host closed the connection)
[1:21] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) has left #ceph
[1:33] * Tv|work (~Tv_@aon.hq.newdream.net) Quit (Ping timeout: 480 seconds)
[1:41] <sagewk> sjusthm: can i check for omap key existance with get_omap_iterator() and then calling ->valid()?
[1:41] <sjusthm> yes
[1:42] <sjusthm> you also need to check for the header, I guess
[1:43] <sagewk> if the iterator has a lock, i have to delete it before syncing i assume?
[1:43] <sjusthm> yes
[1:43] <sjusthm> well, no, probably not, but as a general rule, don't hold onto objectmap iterators
[1:45] <sjusthm> also, you need to call seek_to_first() before checking valid
[1:45] <sagewk> k
[1:45] * sjustlaptop (~sam@24-205-39-1.dhcp.gldl.ca.charter.com) has joined #ceph
[1:54] <sagewk> yehudasa_: rgw encoders look good
[1:55] <sagewk> sjusthm: pushed wip-2098 with object_map sync. we don't set guards for any omap operations that have read-side effects, though.. if there are any!
[1:55] <yehudasa_> sagewk: thanks
[2:04] * Tv__ (~tv@cpe-24-24-131-250.socal.res.rr.com) has joined #ceph
[2:13] <sjustlaptop> sagewk: there aren't any omap operations with read-side effects :)
[2:17] <yehudasa_> sjustlaptop: prefix looks good
[2:17] <sjustlaptop> ok, I'll push to master
[2:20] * eightyeight (~atoponce@pthree.org) has joined #ceph
[2:23] * eightyeight (~atoponce@pthree.org) Quit ()
[2:24] * eightyeight (~atoponce@pthree.org) has joined #ceph
[2:36] * eightyeight (~atoponce@pthree.org) Quit (Ping timeout: 480 seconds)
[2:38] * eightyeight (~atoponce@pthree.org) has joined #ceph
[2:42] <sage> sjustlaptop: shouldn't we object_map->sync() in sync_entry() too?
[2:42] <sjustlaptop> sage: the leveldb directory is under current/
[2:43] <sage> doesn't leveldb have to flush too? or does every operation write into the page cache?
[2:43] <sjustlaptop> and it applies all operations to the fs immediately
[2:43] <sage> gotcha, ok.
[2:43] <sage> it might be worth an object_map->flush() operation that's a no-op for the leveldb implementation, just so it's clear
[2:45] <sjusthm> seems simple enough to worry about when/if we change backends
[2:45] <sage> yeah as long as we remember ;)
[2:46] <sjusthm> I can add it to the comments in the ObjectMap header file since it's a part of the interface contract
[2:58] * lofejndif (~lsqavnbok@214.Red-83-43-124.dynamicIP.rima-tde.net) Quit (Quit: Leaving)
[3:28] * adjohn (~adjohn@rackspacesf.static.monkeybrains.net) Quit (Quit: adjohn)
[3:29] * lxo (~aoliva@lxo.user.oftc.net) Quit (Ping timeout: 480 seconds)
[3:38] * lxo (~aoliva@lxo.user.oftc.net) has joined #ceph
[4:12] * chutzpah (~chutz@ Quit (Quit: Leaving)
[4:13] * sjustlaptop (~sam@24-205-39-1.dhcp.gldl.ca.charter.com) has left #ceph
[4:46] * Jaykra (~Jamie@64-126-89-248.dyn.everestkc.net) has joined #ceph
[5:01] * joshd (~joshd@aon.hq.newdream.net) Quit (Quit: Leaving.)
[5:35] * Jaykra (~Jamie@64-126-89-248.dyn.everestkc.net) Quit (Quit: Leaving.)
[5:35] * Jaykra (~Jamie@64-126-89-248.dyn.everestkc.net) has joined #ceph
[5:43] * sjusthm (~sam@24-205-39-1.dhcp.gldl.ca.charter.com) Quit (Read error: Connection reset by peer)
[6:01] * Jaykra (~Jamie@64-126-89-248.dyn.everestkc.net) has left #ceph
[6:09] * adjohn (~adjohn@50-0-92-115.dsl.dynamic.sonic.net) has joined #ceph
[6:55] * jpieper_ (~josh@209-6-86-62.c3-0.smr-ubr2.sbo-smr.ma.cable.rcn.com) has joined #ceph
[6:55] * timg (~tim@leibniz.catalyst.net.nz) has joined #ceph
[6:56] * guido (~guido@mx1.hannover.ccc.de) Quit (resistance.oftc.net oxygen.oftc.net)
[6:56] * aliguori (~anthony@cpe-70-123-132-139.austin.res.rr.com) Quit (resistance.oftc.net oxygen.oftc.net)
[6:56] * pruby (~tim@leibniz.catalyst.net.nz) Quit (resistance.oftc.net oxygen.oftc.net)
[6:56] * jpieper (~josh@209-6-86-62.c3-0.smr-ubr2.sbo-smr.ma.cable.rcn.com) Quit (resistance.oftc.net oxygen.oftc.net)
[6:56] * cattelan_away (~cattelan@c-66-41-26-220.hsd1.mn.comcast.net) Quit (resistance.oftc.net oxygen.oftc.net)
[6:58] * guido (~guido@mx1.hannover.ccc.de) has joined #ceph
[6:58] * aliguori (~anthony@cpe-70-123-132-139.austin.res.rr.com) has joined #ceph
[6:58] * cattelan_away (~cattelan@c-66-41-26-220.hsd1.mn.comcast.net) has joined #ceph
[7:01] * aliguori (~anthony@cpe-70-123-132-139.austin.res.rr.com) Quit (Read error: Operation timed out)
[7:12] * aliguori (~anthony@cpe-70-123-132-139.austin.res.rr.com) has joined #ceph
[7:45] * Tv__ (~tv@cpe-24-24-131-250.socal.res.rr.com) Quit (Ping timeout: 480 seconds)
[8:21] * tnt_ (~tnt@37.191-67-87.adsl-dyn.isp.belgacom.be) has joined #ceph
[9:40] * LarsFronius (~LarsFroni@testing78.jimdo-server.com) has joined #ceph
[9:44] * tnt_ (~tnt@37.191-67-87.adsl-dyn.isp.belgacom.be) Quit (Ping timeout: 480 seconds)
[9:52] * tnt_ (~tnt@212-166-48-236.win.be) has joined #ceph
[11:00] * adjohn (~adjohn@50-0-92-115.dsl.dynamic.sonic.net) Quit (Quit: adjohn)
[11:19] * stxShadow (~jens@p4FD0657D.dip.t-dialin.net) has joined #ceph
[11:30] * yoshi (~yoshi@p8031-ipngn2701marunouchi.tokyo.ocn.ne.jp) Quit (Remote host closed the connection)
[12:05] * tnt_ (~tnt@212-166-48-236.win.be) Quit (Ping timeout: 480 seconds)
[12:12] * tnt_ (~tnt@37.191-67-87.adsl-dyn.isp.belgacom.be) has joined #ceph
[13:17] * stxShadow (~jens@p4FD0657D.dip.t-dialin.net) Quit (Remote host closed the connection)
[13:23] * tnt_ (~tnt@37.191-67-87.adsl-dyn.isp.belgacom.be) Quit (Ping timeout: 480 seconds)
[13:30] * tnt_ (~tnt@212-166-48-236.win.be) has joined #ceph
[13:33] <nhm> good morning cephalopod enthusiasts!
[13:37] * andreask (~andreas@85-127-86-65.dynamic.xdsl-line.inode.at) has joined #ceph
[13:38] * andreask (~andreas@85-127-86-65.dynamic.xdsl-line.inode.at) has left #ceph
[14:05] * mtk (~mtk@ool-44c35967.dyn.optonline.net) has joined #ceph
[14:36] <lxo> sage, ping
[15:03] * BManojlovic (~steki@ has joined #ceph
[15:13] * Jaykra (~Jamie@64-126-89-248.dyn.everestkc.net) has joined #ceph
[15:14] * Jaykra (~Jamie@64-126-89-248.dyn.everestkc.net) has left #ceph
[15:47] <dwm__> nhm: Howdy.
[15:47] <nhm> dwm__: hello!
[15:48] <nhm> wow, double free or corruption from doing a git grep.
[15:49] * dwm__ tries to work out how to optimally size (multiple?) ceph OSDs in a box.
[15:49] <nhm> dwm__: size in terms of disk size?
[15:57] <dwm__> nhm: Number of disks.
[15:58] <dwm__> If I've, say, got a 2U box with 22-odd spinning disks in it and a chunk of RAM, trying to determine whether to simply have one big OSD across the whole set, one OSD per disk, or something imbetween.
[15:59] <nhm> dwm__: hopefully we'll have some examples soon.
[16:00] <dwm__> I recall someone talking on the mailing-list about the ideal number/size of PGs or similar, but I can't remember the specifics.
[16:00] <iggy> dwm__: there have been quite a few emails on the topic on the mailing list
[16:00] <dwm__> I'm about to go digging back in my mailing list history..
[16:00] <iggy> that's your best bet for now
[16:22] * BManojlovic (~steki@ Quit (Quit: Ja odoh a vi sta 'ocete...)
[16:33] * joao (~JL@ has joined #ceph
[16:33] <nhm> morning joao
[16:34] <joao> good morning nhm :)
[16:42] * jluis (~JL@ace.ops.newdream.net) has joined #ceph
[16:43] * joao (~JL@ Quit (Ping timeout: 480 seconds)
[16:59] <dwm__> Hmm, found some references to number-of-PGs/OSD; I thought I remembered seeing suggestions for bounds on the size of a single PG..
[17:03] <nhm> dwm__: I think I remember seeing something like that as well.
[17:05] * Jaykra (~Jamie@ has joined #ceph
[17:08] * jluis is now known as joao
[17:08] * Jaykra (~Jamie@ Quit (Read error: Connection reset by peer)
[17:08] <joao> nhm, have you ever build ceph's tests?
[17:08] <nhm> joao: I've built s3-tests...
[17:09] * tnt_ (~tnt@212-166-48-236.win.be) Quit (Ping timeout: 480 seconds)
[17:09] <joao> probably not talking about the same; nevermind, I'll figure it out, although it's not time sensitive :)
[17:11] <nhm> joao: yeah, probably not the same tests. I can try building whatever you are working on if you'd like.
[17:11] <joao> no need, thanks
[17:11] <joao> I'm just wondering what should I link the test to
[17:12] <joao> but I'll figure it out while I'm solving the missing symbols errors :p
[17:13] * jmlowe (~Adium@129-79-195-139.dhcp-bl.indiana.edu) has joined #ceph
[17:15] * Jaykra (~Jamie@ has joined #ceph
[17:16] <nhm> joao: hehe
[17:16] <joao> oh goodie
[17:16] <joao> I really love this time of day
[17:16] <nhm> joao: oh?
[17:16] <joao> people in the US start replying to mails
[17:16] <nhm> heh
[17:17] <joao> ;)
[17:20] * tnt_ (~tnt@37.191-67-87.adsl-dyn.isp.belgacom.be) has joined #ceph
[17:31] <joao> srsly, sometimes I'm baffled by my reduced attention span
[17:31] <joao> read ./configure's help a bunch of times, and somehow always missed the --with-debug key
[17:33] <nhm> joao: any luck this morning with sepia?
[17:33] <joao> no joy
[17:33] <joao> still using flak
[17:34] <sagewk> fwiw, guys, the old sepia nodes work, they just cant reach outside network. that means opening the ssh connections takes 30 seconds (reverse dns times out) but after that things behave
[17:34] <joao> right now, not even from flak am I able to access sepia16
[17:35] <sagewk> the main exception is that the workunit tasks pull things from github, so those won't work
[17:35] <sagewk> try from metropolis[.ceph.dreamhost.com]
[17:35] <sagewk> that machine is on the same subnet
[17:36] <joao> ah yes, it works from metropolis :)
[17:37] <joao> I guess I'll be installing the missing libs in sepia using metropolis as a proxy for .debs :p
[17:37] <joao> manual proxy though
[17:37] <nhm> sagewk: yeah, the thing that has been blocking me from flak is that no nodes are available right now. I do have a couple of previously locked ones I could take down and re-use. Didn't want to do that at first, but maybe it's time.
[17:37] <sagewk> joao: missing libs?
[17:38] <sagewk> you can build on metropolis, and push the code to sepia with a
[17:38] <joao> sage, I'm missing a couple of libs in sepia16 that prevents me from build the ceph repo
[17:38] <sagewk> - ceph:
[17:38] <sagewk> path: /path/to/ceph
[17:38] <sagewk> (it'll push out the code from your build dir)
[17:38] <joao> oh
[17:38] <joao> that sounds nice
[17:39] <joao> I'll look into it later tonight then :)
[17:41] <sagewk> nhm: the check would have been before lock.py was added. not sure it really matters, tho (and i may be misremembering).
[17:41] <joao> btw, sagewk, from the looks of it, the workload Ted described on fsdevel is pretty similar to ceph's
[17:41] <sagewk> nhm: i think we just need an assert len(roles) <= len(targets) sometime after the lock check.
[17:41] <sagewk> joao: yeah :)
[17:42] <joao> this thread sounds very promising
[17:42] <nhm> sagewk: ok, that's kind of what I was thinking, but I wanted to check in first.
[17:43] * bchrisman (~Adium@c-76-103-130-94.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[17:44] <sagewk> elder: around?
[17:44] * Jaykra (~Jamie@ Quit (Read error: Connection reset by peer)
[17:45] * Jaykra (~Jamie@ has joined #ceph
[17:45] <nhm> sage: ok, looks like lock.py was added back in Jul.
[18:01] * Tv|work (~Tv_@aon.hq.newdream.net) has joined #ceph
[18:01] * Jaykra (~Jamie@ Quit (Read error: Connection reset by peer)
[18:08] * aliguori (~anthony@cpe-70-123-132-139.austin.res.rr.com) Quit (Quit: Ex-Chat)
[18:10] * Jaykra (~Jamie@ has joined #ceph
[18:13] <nhm> joao: did you have to do anything to get to the sepia nodes from flak? I managed to get some nodes to lock on flak but still can't get to them.
[18:13] <joao> I wasn't able to reach them from flak
[18:13] <joao> I just did as sagewk suggested, and accessed them from metropolis :)
[18:14] <nhm> joao: ah, ok
[18:14] <joao> trying to access them from flak times out
[18:14] <joao> and by "them" I mean "sepia16" :p
[18:15] <nhm> heh, no home directory on metropolis...
[18:17] * Jaykra (~Jamie@ Quit (Quit: Leaving.)
[18:17] <nhm> joao: if you happen to have sudo access on metropolis would you mind making me a home directory?
[18:18] <Tv|work> crowdsourced adminning :(
[18:18] <joao> nhm, I dont :\
[18:18] <sagewk> nhm: got it
[18:18] <sagewk> try now
[18:18] <nhm> sagewk: cool, thanks
[18:22] <nhm> Tv|work: I suppose we could setup a formal ticketing system for IT requests. ;)
[18:22] <Tv|work> nhm: it's more about doing the work consistently, than knowing what there is out there to do
[18:23] <Tv|work> nhm: new stuff tries to be way better automated, but it's easy to slip off that track under pressure to get things done
[18:24] * LarsFronius (~LarsFroni@testing78.jimdo-server.com) Quit (Quit: LarsFronius)
[18:25] <nhm> Tv|work: I figure it'll come with time...
[18:26] <nhm> Tv|work: Honestly this is nice compared to my old job where developers and IT staff were constantly battling over admin rights and ticket response times.
[18:33] * joshd (~joshd@aon.hq.newdream.net) has joined #ceph
[18:34] * aliguori (~anthony@ has joined #ceph
[18:51] * chutzpah (~chutz@ has joined #ceph
[18:57] * bchrisman (~Adium@ has joined #ceph
[20:11] * nhorman (~nhorman@99-127-245-201.lightspeed.rlghnc.sbcglobal.net) has joined #ceph
[20:25] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) has joined #ceph
[21:14] * adjohn (~adjohn@rackspacesf.static.monkeybrains.net) has joined #ceph
[21:20] * adjohn (~adjohn@rackspacesf.static.monkeybrains.net) Quit (Quit: adjohn)
[21:21] * stxShadow (~jens@ip-88-153-224-220.unitymediagroup.de) has joined #ceph
[21:21] * adjohn (~adjohn@rackspacesf.static.monkeybrains.net) has joined #ceph
[21:22] <stxShadow> good evening all
[21:27] <Tv|work> sjust1, gregaf1: http://eagain.net/articles/books/
[21:29] <nhm> Tv|work: I really enjoy Stephenson's work.
[21:29] <Tv|work> nhm: Neal? Bleh.
[21:29] <nhm> Tv|work: To each his own. :P
[21:30] <Tv|work> Robert L. Forward, Fred Hoyle, Vernor Vinge, Charless Stross, etc for me
[21:30] <nhm> I see you like Niven, whom I like too though.
[21:31] <Tv|work> yeah Niven is good, though he keeps churning the same story for too long
[21:31] <Tv|work> sjust1: http://www.antipope.org/charlie/blog-static/fiction/accelerando/accelerando-intro.html
[21:32] <Tv|work> oh and Hal Clement ftw
[21:32] <nhm> Most of the recreational reading I do these days are children's stories though. :)
[21:32] <Tv|work> a lot of the old good books are only available as collector's reprints, these days, sadly
[21:33] <Tv|work> alright back to configuring openvpn
[21:34] * BManojlovic (~steki@ has joined #ceph
[21:35] <dmick> Snow Crash and Diamond Age were geek fun, man, how can you not dig those?.... sadface.
[21:35] <dmick> ok Zodiac kinda went off course for me
[21:36] <Tv|work> dmick: as i said over the lunch table, i don't like stories where you can s/computer/magic/g; s/spaceship/dragon/g; s/something/wizard/g; and have the story still make as much sense
[21:36] <dmick> heh
[21:36] <Tv|work> dmick: as far as i'm concerned, Neal Stephenson and William Gibson write fluff
[21:36] <dmick> but I like fluff! :)
[21:36] <Tv|work> and i like hard scifi
[21:37] <dmick> ok ok, opinions etc.
[21:37] <dmick> I like hard sf too
[21:37] <Tv|work> IMHO, Gibson's concept of cyberspace is as stupid as current day Hollywood stories.
[21:37] <nhm> dmick: I still need to pick up reamde
[21:37] <dmick> heh. looking at Accelerando: I have a t-shirt from Brouwerij 't IJ
[21:38] <Tv|work> Accelerando is awesome. Grab it for free online, if you like it *then* buy it ;)
[21:38] <Tv|work> it's a short read but.. well compressed.
[21:38] <Tv|work> there's like a between-the-lines reference at least every 5 lines or so, though the whole story
[21:41] * lofejndif (~lsqavnbok@214.Red-83-43-124.dynamicIP.rima-tde.net) has joined #ceph
[21:42] * lofejndif (~lsqavnbok@214.Red-83-43-124.dynamicIP.rima-tde.net) Quit (Max SendQ exceeded)
[21:43] <stxShadow> how may i rewight an osd -> that it totaly moves the data to another node
[21:43] <stxShadow> after that i want to remove it safly
[21:43] * lofejndif (~lsqavnbok@214.Red-83-43-124.dynamicIP.rima-tde.net) has joined #ceph
[21:44] <joao> Tv|work, in all justice, it's fun to read about someone being murdered by a couple KB's of RAM
[21:44] <joao> the world should have been much funnier in Gibson's mind
[21:46] <Tv|work> joao: Starfish/Maelstrom/Behemoth contains a subplot about a virus that realizes that messages that include the name of the antihero main character are more likely to pass through antivirus scanners, and starts spamming imaginary stories about sightings to help itself spread.
[21:46] <Tv|work> joao: thereby helping the main character evade surveillance via false positives, etc
[21:49] <joao> that looks like an interesting plot
[21:49] <Tv|work> unfortunately the writing gets worse and worse as the story goes along
[21:49] <joao> I must say I've never been much of a sci-fi novel kind of guy, except for the hitchhikers guide to the galaxy saga
[21:49] <Tv|work> the last book did not get a publisher interested, so it's not edited at all :(
[21:50] <Tv|work> way better from the same author is Blindsight
[21:50] <Tv|work> it ain't scifi unless it's got science :(
[21:50] <joao> *fi then
[21:50] <joao> :p
[21:50] <Tv|work> hitchhiker's is a very fun read, but it's more absurdist than scifi
[21:51] <joao> that's the one case I don't really care what category it falls into :x
[21:52] <joao> now that I think of it, aside from a couple of asimov's books and the necromancer, I don't think I remember reading much [sci-]fi
[21:53] <Tv|work> yeah then you might not want my reading list ;)
[21:55] <joao> oh no, I most certainly do
[21:55] <joao> I like to pile up books and read them when I get bored
[21:55] <joao> aside from Agatha Christie's books
[21:56] <joao> I just wished I'd get bored more often, since I'm yet to finish catch-22 and 1984
[22:00] <joao> Tv|work, from the top of your head, name a couple that would be an interesting read :)
[22:01] <sjust1> Tv|work: newer Stephenson stuff is a lot better than his older stuff
[22:06] <gregaf1> stxShadow: you can mark an OSD out without marking it down; that'll cause a full data export :)
[22:06] <stxShadow> only "ceph osd out x" ?
[22:08] <gregaf1> yep, I believe so!
[22:08] <Tv|work> joao: decent starters: Ringworld by Niven, The Legacy of Heorot, A Fire Upon the Deep
[22:08] <gregaf1> there might be some more steps involved in actually removing it from the map once data's migrated off, but that's a different thing
[22:08] <joao> great, thanks
[22:09] <stxShadow> thx ... btw ... i got rid of my inconsistencies by exporting the data and deleting the pools .... the reimport the data
[22:09] <Tv|work> joao: perhaps less to the Heorot thing, that's just a recent read for me, so fresh in my mind
[22:09] * joao adds them to his book wishlist
[22:10] <gregaf1> okay, glad it's working again
[22:10] <Tv|work> joao: but once you get a hang of "what I like", if you agree, dig through the 5/5 ratings and 4/5 ratings on that page
[22:11] <joao> sure
[22:11] <joao> I'm badly in need of a book that pulls me into the story
[22:12] <stxShadow> gregaf1 ..... yes ..... but one little problem left ..... if any of my 4 OSDs fail ...... there is no chance to access the data until the osd is restarted ...... any ideas ?
[22:12] <dmick> Tv|work: how do you feel about Ellison?
[22:12] <joao> the last I can remember that did it for me was Gaiman's American gods
[22:12] <joao> and it has been a while since
[22:12] <gregaf1> stxShadow: sounds like you created a pool that isn't replicated?
[22:12] <dmick> ooh, love Gaiman too. again not really sf at all
[22:12] <joao> yeah, I know :p
[22:13] <Tv|work> dmick: Harlan? haven't read much
[22:14] <stxShadow> gregaf1 ..... not .... thats not the case .... i just checked it:
[22:14] <Tv|work> dmick: oh hey i may have read his street gang book
[22:15] <stxShadow> root@fcmsnode3:~# ceph osd dump - | grep pool -c
[22:15] <stxShadow> 151
[22:15] <stxShadow> root@fcmsnode3:~# ceph osd dump - | grep pool | grep "rep size 2" -c
[22:15] <stxShadow> 151
[22:17] <gregaf1> huh, dunno then
[22:18] * joao (~JL@ace.ops.newdream.net) Quit (Read error: Operation timed out)
[22:18] <gregaf1> are they actually getting marked down at any point?
[22:20] <stxShadow> yes .... they are marked down normaly .... "ceph osd tree" report them down too ....
[22:20] <gregaf1> but you can't retrieve data?
[22:20] * nhorman (~nhorman@99-127-245-201.lightspeed.rlghnc.sbcglobal.net) Quit (Quit: Leaving)
[22:21] <gregaf1> I guess I'd get a pg dump and ask sjust or yehudasa_ to take a look at it
[22:22] <stxShadow> yes ..... neither by rbd nor by ceph client
[22:22] <stxShadow> it still freezes until the osd is up again ....
[22:22] <stxShadow> shall is send the dump by mail ?
[22:23] <stxShadow> and: with down osd or up ?
[22:24] <gregaf1> down will have more info about what's going wrong
[22:25] <stxShadow> ok .... i will send that when one osd is down next time ..... actually i get a sms if one is down ;)
[22:25] <stxShadow> thanks a lot again .... and good night
[22:27] * stxShadow (~jens@ip-88-153-224-220.unitymediagroup.de) Quit (Quit: bye bye !! )
[22:35] * joao (~JL@ has joined #ceph
[22:40] * lofejndif (~lsqavnbok@214.Red-83-43-124.dynamicIP.rima-tde.net) Quit (Ping timeout: 480 seconds)
[22:47] * lofejndif (~lsqavnbok@214.Red-83-43-124.dynamicIP.rima-tde.net) has joined #ceph
[22:48] * lofejndif (~lsqavnbok@214.Red-83-43-124.dynamicIP.rima-tde.net) Quit (Max SendQ exceeded)
[22:52] * lofejndif (~lsqavnbok@214.Red-83-43-124.dynamicIP.rima-tde.net) has joined #ceph
[22:59] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) has left #ceph
[23:08] * lofejndif (~lsqavnbok@214.Red-83-43-124.dynamicIP.rima-tde.net) Quit (Quit: Leaving)
[23:09] * lofejndif (~lsqavnbok@214.Red-83-43-124.dynamicIP.rima-tde.net) has joined #ceph
[23:18] <sagewk> sjust1: wip-975 and wip-2153 when you have some time?
[23:18] * aliguori (~anthony@ Quit (Remote host closed the connection)
[23:20] <sjust1> ok
[23:27] * LarsFronius (~LarsFroni@g231139098.adsl.alicedsl.de) has joined #ceph
[23:50] * mtk (~mtk@ool-44c35967.dyn.optonline.net) Quit (Remote host closed the connection)
[23:58] * jmlowe (~Adium@129-79-195-139.dhcp-bl.indiana.edu) Quit (Quit: Leaving.)

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