#ceph IRC Log


IRC Log for 2011-07-08

Timestamps are in GMT/BST.

[0:06] <cmccabe> sagewk: looks like I managed to reproduce our infinitely long pool listing
[0:07] <sagewk> cool. with logs?
[0:07] <cmccabe> yes
[0:07] <sagewk> where can i find them?
[0:08] <cmccabe> they're on metropolis
[0:09] <sagewk> path?
[0:10] <cmccabe> one sec, I didn't output it correctly in one runnable
[0:15] <gregaf> Tv: right now teuthology grabs core dumps but not binaries?
[0:15] <Tv> gregaf: the binaries are in the tarball..
[0:16] <gregaf> which tarball?
[0:16] <gregaf> oh, just off gitbuilder
[0:16] <gregaf> right
[0:17] <Tv> remember: if you want to have them stick around, there has to be a branch or tag pointing at the commit
[0:30] <cmccabe> sagewk: ok, haven't quite managed to reproduce with the logging I wanted yet
[0:37] <cmccabe> I'm going to crank up the number of objects and see if that makes a difference
[0:43] * mtk (~mtk@ool-182c8e6c.dyn.optonline.net) Quit (Remote host closed the connection)
[1:09] <sagewk> cmccabe: any luck?
[2:15] * Tv (~Tv|work@ip-64-111-111-107.dreamhost.com) Quit (Ping timeout: 480 seconds)
[2:36] * aliguori (~anthony@cpe-70-123-132-139.austin.res.rr.com) Quit (Quit: Ex-Chat)
[2:43] * yoshi_ (~yoshi@p15251-ipngn1601marunouchi.tokyo.ocn.ne.jp) has joined #ceph
[2:50] * bchrisman (~Adium@70-35-37-146.static.wiline.com) Quit (Quit: Leaving.)
[3:21] * joshd (~joshd@ip-64-111-111-107.dreamhost.com) Quit (Quit: Leaving.)
[3:30] * jim (~chatzilla@astound-69-42-16-6.ca.astound.net) Quit (Ping timeout: 480 seconds)
[3:41] * jim (~chatzilla@astound-69-42-16-6.ca.astound.net) has joined #ceph
[4:05] <cmccabe> test runs finished successfully, looks like
[4:07] * bchrisman (~Adium@c-98-207-207-62.hsd1.ca.comcast.net) has joined #ceph
[5:12] * yoshi_ (~yoshi@p15251-ipngn1601marunouchi.tokyo.ocn.ne.jp) Quit (Remote host closed the connection)
[5:24] * cmccabe (~cmccabe@c-24-23-254-199.hsd1.ca.comcast.net) has left #ceph
[5:38] * yoshi_ (~yoshi@p15251-ipngn1601marunouchi.tokyo.ocn.ne.jp) has joined #ceph
[9:47] * jim (~chatzilla@astound-69-42-16-6.ca.astound.net) Quit (Remote host closed the connection)
[9:54] * jim (~chatzilla@astound-69-42-16-6.ca.astound.net) has joined #ceph
[10:30] * DanielFriesen (~dantman@S0106001731dfdb56.vs.shawcable.net) Quit (Remote host closed the connection)
[10:36] * jiaju (~jjzhang@ has joined #ceph
[10:40] * Dantman (~dantman@S0106001731dfdb56.vs.shawcable.net) has joined #ceph
[11:52] * jiaju (~jjzhang@ Quit (Quit: ??????)
[12:08] * yoshi_ (~yoshi@p15251-ipngn1601marunouchi.tokyo.ocn.ne.jp) Quit (Remote host closed the connection)
[12:39] <damoxc> what's the best way to destroy and re-create a osd?
[12:39] <damoxc> it seems the btrfs partition has become corrupted
[12:55] * tjikkun (~tjikkun@2001:7b8:356:0:204:bff:fe80:8080) Quit (Ping timeout: 480 seconds)
[14:44] * aliguori (~anthony@cpe-70-123-132-139.austin.res.rr.com) has joined #ceph
[17:19] * morse (~morse@supercomputing.univpm.it) Quit (Remote host closed the connection)
[17:31] * Tv (~Tv|work@ip-64-111-111-107.dreamhost.com) has joined #ceph
[17:43] * bchrisman (~Adium@c-98-207-207-62.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[17:53] * greglap (~Adium@ has joined #ceph
[17:54] <sagewk> damoxc: something like kill `pidof cosd` ; cosd -i <id> --mkfs ; service ceph start
[17:54] <sagewk> make sure you don't clobber the osd keyring or its a few more steps to reextract that from the monitor
[18:03] * greglap (~Adium@ Quit (Quit: Leaving.)
[18:17] * morse (~morse@supercomputing.univpm.it) has joined #ceph
[18:17] * morse (~morse@supercomputing.univpm.it) Quit (Remote host closed the connection)
[18:19] * morse (~morse@supercomputing.univpm.it) has joined #ceph
[18:27] * morse (~morse@supercomputing.univpm.it) Quit (Ping timeout: 480 seconds)
[18:33] * morse (~morse@supercomputing.univpm.it) has joined #ceph
[18:36] * bchrisman (~Adium@70-35-37-146.static.wiline.com) has joined #ceph
[18:40] * joshd (~joshd@ip-64-111-111-107.dreamhost.com) has joined #ceph
[18:50] * jbd (~jbd@ks305592.kimsufi.com) has joined #ceph
[19:36] * aliguori (~anthony@cpe-70-123-132-139.austin.res.rr.com) Quit (Quit: Ex-Chat)
[20:26] * aliguori (~anthony@ has joined #ceph
[20:30] <sjust> sagewk: clone_range sources must have the same object locator as the target, right?
[20:32] <gregaf> sjust: sage is in a meeting until lunch, but I think you're correct
[20:32] <gregaf> yehudasa would know; he wrote it
[20:32] <sjust> ok, thought so
[20:32] <sjust> it's the only way to be sure that it's in the same pg, after all
[20:32] <bchrisman> uggg??? same build??? previously I could create an rbd image, load the rbd module, and /dev/rbd0 would show up??? not happening right now.. anything I'm missing?
[20:33] <bchrisman> nothign in dmesg or messages seems particular out of place.
[20:33] <bchrisman> rbd ls bar and rbd info bar give me what I expect
[20:34] <bchrisman> ahh forgot to add in sysfs :)
[21:19] * gregorg_taf (~Greg@ has joined #ceph
[21:27] * jbd (~jbd@ks305592.kimsufi.com) has left #ceph
[21:27] * jbd (~jbd@ks305592.kimsufi.com) has joined #ceph
[22:02] * jbd (~jbd@ks305592.kimsufi.com) has left #ceph
[22:03] * jmlowe (~Adium@140-182-129-35.dhcp-bl.indiana.edu) has joined #ceph
[22:03] <jmlowe> A little quick feedback on using /dev/rbdN to back a kvm machine
[22:04] <jmlowe> cache=writeback
[22:04] <jmlowe> time tar xjf linux-
[22:04] <jmlowe> real 10m17.685s
[22:04] <jmlowe> user 0m19.312s
[22:04] <jmlowe> sys 0m7.448s
[22:04] <jmlowe> cache=none
[22:04] <jmlowe> time tar xjf linux-
[22:04] <jmlowe> real 0m28.363s
[22:04] <jmlowe> user 0m20.761s
[22:04] <jmlowe> sys 0m6.054s
[22:07] <darkfaded> writeback is slower than uncached?
[22:07] * aliguori (~anthony@ Quit (Ping timeout: 480 seconds)
[22:07] <jmlowe> other way arround
[22:09] * lidongyang (~lidongyan@ Quit (Ping timeout: 480 seconds)
[22:10] <jmlowe> using writeback in qemu with /dev/rbd0 as a raw block store is 21x faster
[22:10] * jbd (~jbd@ks305592.kimsufi.com) has joined #ceph
[22:10] * lidongyang (~lidongyan@ has joined #ceph
[22:11] <jmlowe> you guys weren't kidding when you said the buffering in rbd was lacking
[22:16] * gregorg_taf (~Greg@ Quit (Quit: Quitte)
[22:17] * aliguori (~anthony@ has joined #ceph
[22:20] <jmlowe> elevator=noop in kvm host seems to be a bad idea
[22:26] * jbd (~jbd@ks305592.kimsufi.com) has left #ceph
[22:47] * aliguori (~anthony@ Quit (Quit: Ex-Chat)
[23:01] <gregaf> jmlowe: yeah, there just isn't any buffering at all :)
[23:03] <yehudasa> sjust: that's what we do obviously, but there's nothing inherent that actually requires it, you can clone_range random stuff, but the results will be random
[23:05] * jbd (~jbd@ks305592.kimsufi.com) has joined #ceph
[23:29] * jbd (~jbd@ks305592.kimsufi.com) has left #ceph
[23:48] <gregaf> bchrisman: vger doesn't accept html email, full stop ??? I don't think I've heard about anybody with a real email having problems otherwise, though
[23:49] <Tv> that's the likely reason, though at least back in the day vger also used a blacklist of words you must not say or you're spam, and that sometimes got triggered
[23:49] <Tv> as in, part of someone's name or something
[23:50] <bchrisman> hmm??? that was just strange because I've sent email from my gmail account to the list before.. maybe they consider iscsi to be a spamword.. :)
[23:50] <Tv> bchrisman: the plain vs rich text in gmail has been causing me a ton of trouble lately too.. it could be google is fiddling with the default settings or something
[23:50] <Tv> i keep accidentally replying in rich text
[23:50] <Tv> hey at least my "Reply to all" Labs feature works now!
[23:50] <Tv> (no idea what changed)
[23:51] <bchrisman> ahh yeah??? that was one thing I thought might be the culprit??? they keep 'fixing' things.
[23:51] <bchrisman> Tv: hmm.. that in gmail? something I can access? :)
[23:51] <Tv> bchrisman: yeah, in settings, under Labs
[23:51] <gregaf> I've never had such problems; I wonder if I just had my settings long enough that their system doesn't dare change anything on me and carried it over via account switching
[23:51] <Tv> the plain/rich text thing used to be "what ever you chose the last time you sent a message"
[23:52] <Tv> but now, if the message contains certain characters, it'll choose rich text by itself
[23:52] <Tv> replying to emails from china kept triggering it for me
[23:52] <Tv> gregaf: well it's also the whole a/b testing thing
[23:53] <gregaf> hmm, maybe if you switch your text encoding it won't resort to HTML?
[23:53] <Tv> i have enforced utf-8, which is the best i can do
[23:53] <bchrisman> Tv: cool??? will turn that on until it embarrasses me somehow.. :)
[23:57] * jmlowe (~Adium@140-182-129-35.dhcp-bl.indiana.edu) has left #ceph
[23:59] * NashTrash (~Adium@ has left #ceph

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