#ceph IRC Log


IRC Log for 2011-05-20

Timestamps are in GMT/BST.

[0:00] * lxo (~aoliva@ Quit (Read error: Connection reset by peer)
[0:03] * joshd (~joshd@ip-66-33-206-8.dreamhost.com) has joined #ceph
[0:06] * aliguori (~anthony@ Quit (Remote host closed the connection)
[0:10] * alexxy (~alexxy@ Quit (Quit: No Ping reply in 180 seconds.)
[0:14] * ghaskins_mobile (~ghaskins_@66-189-113-47.dhcp.oxfr.ma.charter.com) has joined #ceph
[0:15] * lxo (~aoliva@ has joined #ceph
[0:17] * lx0 (~aoliva@ has joined #ceph
[0:21] * bchrisman (~Adium@c-98-207-207-62.hsd1.ca.comcast.net) has joined #ceph
[0:23] * lxo (~aoliva@ Quit (Ping timeout: 480 seconds)
[0:28] * lx0 (~aoliva@ Quit (Ping timeout: 480 seconds)
[0:33] * lx0 (~aoliva@ has joined #ceph
[0:47] * lx0 (~aoliva@ Quit (Read error: Connection reset by peer)
[0:59] * lx0 (~aoliva@ has joined #ceph
[1:21] <sagewk> it's so quiet in here today!
[1:21] <bchrisman> that's what happens when greg's out? :)
[1:22] <sagewk> or tv takes a break from battling autotest :)
[1:22] <Tv> hah
[1:22] <Tv> yeah today is hunker down & write code day
[1:22] <Tv> bchrisman: then again, you don't see the cardboard robot we've bult
[1:22] <Tv> built
[1:23] <bchrisman> heh
[1:25] * ghaskins_mobile (~ghaskins_@66-189-113-47.dhcp.oxfr.ma.charter.com) Quit (Quit: This computer has gone to sleep)
[1:33] * lx0 (~aoliva@ Quit (Ping timeout: 480 seconds)
[1:52] * ghaskins_mobile (~ghaskins_@66-189-113-47.dhcp.oxfr.ma.charter.com) has joined #ceph
[1:52] * verwilst (~verwilst@dD5769611.access.telenet.be) Quit (Quit: Ex-Chat)
[2:03] * joshd (~joshd@ip-66-33-206-8.dreamhost.com) Quit (Quit: Leaving.)
[2:04] * joshd (~joshd@ip-66-33-206-8.dreamhost.com) has joined #ceph
[2:10] <Tv> "This is a convenience method, equivalent
[2:10] <Tv> to C{shutdown(1)}, for people who don't make it a habit to
[2:10] <Tv> memorize unix constants from the 1970s."
[2:10] <Tv> whee ;)
[2:24] * lxo (~aoliva@ has joined #ceph
[2:28] * ghaskins_mobile (~ghaskins_@66-189-113-47.dhcp.oxfr.ma.charter.com) Quit (Quit: This computer has gone to sleep)
[2:34] * lxo (~aoliva@ Quit (Ping timeout: 480 seconds)
[2:39] * Tv (~Tv|work@ip-66-33-206-8.dreamhost.com) Quit (Ping timeout: 480 seconds)
[2:55] * lxo (~aoliva@ has joined #ceph
[2:59] * joshd (~joshd@ip-66-33-206-8.dreamhost.com) Quit (Quit: Leaving.)
[3:03] * cmccabe (~cmccabe@ has left #ceph
[3:12] * lxo (~aoliva@ Quit (Read error: Connection reset by peer)
[3:28] * lxo (~aoliva@ has joined #ceph
[4:26] * lxo (~aoliva@ Quit (Read error: Connection reset by peer)
[4:40] * lxo (~aoliva@ has joined #ceph
[4:50] * lxo (~aoliva@ Quit (Quit: later)
[4:53] * MK_FG (~MK_FG@ Quit (Quit: o//)
[4:53] * MK_FG (~MK_FG@ has joined #ceph
[4:59] * lxo (~aoliva@ has joined #ceph
[5:47] * rsharpe (~Adium@70-35-37-146.static.wiline.com) Quit (Ping timeout: 480 seconds)
[6:49] * rsharpe (~Adium@70-35-37-146.static.wiline.com) has joined #ceph
[6:56] * iggy (~iggy@theiggy.com) has joined #ceph
[7:42] * neurodrone (~neurodron@cpe-76-180-162-12.buffalo.res.rr.com) Quit (Quit: zzZZZZzz)
[8:00] * atg (~atg@please.dont.hacktheinter.net) Quit (Remote host closed the connection)
[8:00] * atg (~atg@please.dont.hacktheinter.net) has joined #ceph
[8:02] * maswan (maswan@kennedy.acc.umu.se) Quit (Remote host closed the connection)
[8:02] * maswan (maswan@kennedy.acc.umu.se) has joined #ceph
[8:02] * iggy_ (~iggy@theiggy.com) has joined #ceph
[8:03] * Jiaju (~jjzhang@ Quit (Read error: Operation timed out)
[8:03] * iggy (~iggy@theiggy.com) Quit (Ping timeout: 480 seconds)
[8:04] * Jiaju (~jjzhang@ has joined #ceph
[8:04] * Nadir_Seen_Fire (~dantman@S0106001731dfdb56.vs.shawcable.net) has joined #ceph
[8:11] * DanielFriesen (~dantman@S0106001731dfdb56.vs.shawcable.net) Quit (Ping timeout: 480 seconds)
[8:33] * iggy_ is now known as iggy
[9:04] * macana (~ml.macana@ Quit (Ping timeout: 480 seconds)
[9:29] * macana (~ml.macana@ has joined #ceph
[10:20] * alexxy (~alexxy@ has joined #ceph
[10:53] <wonko_be> hey, any recommended way to connect XenServer to Ceph?
[10:53] <wonko_be> XenServer can import iSCSI
[10:53] <wonko_be> (and nfs)
[11:13] <murb> wonko_be: well doesn't it use the qemu for device emulation which has rbd support?
[11:14] <murb> or you could use the host kernel rbd block support and export that.
[11:17] <wonko_be> problem is that i can't divert from the licenced xenserver, and i assume i can't load just any script/module/...
[11:18] <murb> wonko_be: so you can't change the host kernel?
[11:19] <wonko_be> no, xenserver is the "closed appliance" thingy from citrix
[11:20] <wonko_be> afaik fiddling with the kernel isn't allowed (and would break my clients support contract i assume) :-)
[11:20] <wonko_be> I checked if they don't do some connection to S3, but alas...
[11:21] <wonko_be> so, is there a recommended way to get a ceph device out over iscsi?
[11:21] <wonko_be> or should i use nfs and create the vdi's as images
[11:22] * allsystemsarego (~allsystem@ has joined #ceph
[11:22] <wonko_be> an intermedia solution i see is to add a node, and import the ceph device there (rbd) and re-export that through iscsi (ietd)
[11:22] <wonko_be> but that might be a bit of a hassle
[11:23] <murb> wonko_be: you could do that.
[11:23] <wonko_be> however, that defies a bit the goal of a redundant storage cluster :)
[11:24] <wonko_be> i could remount the rbd on all ceph nodes, and export it from there... then maybe multipath
[11:24] <wonko_be> I'll build some test setups. maybe someone has some experience in this area
[11:29] <murb> wonko_be: it'll be very hacky, pity you can't do it properly :(
[11:29] <wonko_be> true
[11:29] <murb> and in my experience iSCSI multipath is a different world of pain.
[11:30] <wonko_be> that i have running a couple of times, quite stable
[11:30] <wonko_be> so no real worries there
[11:30] <murb> wonko_be: against ietd targets?
[11:30] <wonko_be> i'm just looking for the best way to tie ceph to ietd
[11:30] <wonko_be> yes, against ietd
[11:31] <wonko_be> lets say we've seen all possible bugs already
[11:31] <murb> okay :)
[12:04] <admix> i just ran into "libceph: tid 239 timed out on osd3, will reset osd" thing, is it fixed in the 0.28-1~bpo60+1 or should I get fresh git version?
[12:05] <admix> (under heavy traffic)
[14:10] * ghaskins_mobile (~ghaskins_@66-189-113-47.dhcp.oxfr.ma.charter.com) has joined #ceph
[15:03] * alexxy (~alexxy@ Quit (Ping timeout: 480 seconds)
[15:05] * aliguori (~anthony@cpe-70-123-132-139.austin.res.rr.com) has joined #ceph
[16:22] * neurodrone (~neurodron@cpe-76-180-162-12.buffalo.res.rr.com) has joined #ceph
[16:44] <sage> admix: mount -o osdtimeout=0 to turn off that timeout
[16:44] <admix> sage: thanks, going to try it right now :)
[16:48] * Nadir_Seen_Fire (~dantman@S0106001731dfdb56.vs.shawcable.net) Quit (Remote host closed the connection)
[17:01] <admix> some mds issue now; i'm off, thanks anyway
[17:09] * Tv (~Tv|work@ip-66-33-206-8.dreamhost.com) has joined #ceph
[17:18] * Dantman (~dantman@S0106001731dfdb56.vs.shawcable.net) has joined #ceph
[17:26] * mtk (~mtk@ool-182c8e6c.dyn.optonline.net) has joined #ceph
[17:31] * bchrisman (~Adium@c-98-207-207-62.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[17:47] * ghaskins_mobile (~ghaskins_@66-189-113-47.dhcp.oxfr.ma.charter.com) Quit (Quit: This computer has gone to sleep)
[18:17] * ghaskins_mobile (~ghaskins_@66-189-113-47.dhcp.oxfr.ma.charter.com) has joined #ceph
[18:21] <wonko_be> can anyone make out if this is bug, or just me being an idiot? http://pastie.org/1932408
[18:22] <wonko_be> i'm running the latest ceph code (just pulled the git in 2 hours ago)
[18:22] <wonko_be> if it's a bug, I'll document it and report it
[18:22] <wonko_be> (i have it on all the mds, not just this one)
[18:22] <sage> wonko_be: that is a bug. do you have a core file?
[18:23] <sage> yeah please stick it in the tracker!
[18:23] <wonko_be> tell me how to get it, and I'll provide you with a core file
[18:23] <wonko_be> i'll give you anything you need to fix it
[18:24] <sage> if enabled it would probably be /core
[18:24] <sage> but probably it's not there..
[18:24] <wonko_be> nope, can I enable it quickly?
[18:24] <wonko_be> i can easily trigger the bug again
[18:25] <sage> great. add
[18:25] <sage> debug mds = 20
[18:25] <sage> debug ms = 1
[18:25] <sage> to your [mds] section and then reproduce!
[18:25] <sage> and attach teh log to the bug
[18:25] <wonko_be> on it
[18:27] * bchrisman (~Adium@70-35-37-146.static.wiline.com) has joined #ceph
[18:27] <sage> sjust there?
[18:29] <Tv> sage: he's not in yet
[18:29] * ghaskins_mobile (~ghaskins_@66-189-113-47.dhcp.oxfr.ma.charter.com) Quit (Quit: This computer has gone to sleep)
[18:30] <wonko_be> sage: that log is 304MB big
[18:30] <wonko_be> anything particular you might need?
[18:30] <Tv> wonko_be: bzip2 -9 will probably make it be ~20MB
[18:30] <sage> yeah bzip
[18:30] <Tv> still largish..
[18:33] <wonko_be> anything special I need to do to signup in the redmine?
[18:33] <wonko_be> registered myself an account, can't log in now
[18:33] <sage> nope
[18:33] <sage> i think there's an email confirmation step
[18:34] <wonko_be> ah
[18:34] <wonko_be> let me wait for the greylisting then
[18:35] <Tv> wonko_be: btw my experience a few years back was that a 15 second greylist blocked about as much of the spam as a 1 hour one
[18:36] <Tv> and most non-governmental MTAs would resend in <1 minute
[18:36] <sage> wonko_be if the mail is lost i can prod it too
[18:37] <wonko_be> please do
[18:37] <wonko_be> login is "wonko"
[18:37] <wonko_be> Tv: my greylist is set to 5 seconds
[18:38] <sage> done
[18:38] <wonko_be> just "present it twice" is good enough
[18:38] <wonko_be> sage: got it
[18:47] <wonko_be> sage: max upload size is 5 MB, mine is 14 MB
[18:47] <wonko_be> (issue got id 1104)
[18:48] <sage> changed to 50mb
[18:51] <wonko_be> okay, added, let me know if you need more info
[18:51] <wonko_be> I'll keep the testcluster online this weekend, so i can try some stuff if needed
[18:58] <sjust> sage: I'm here
[18:59] <sage> sjust: can you review the latest stable commits and make sure everything makes sense?
[18:59] <sjust> sure
[19:15] * yehudasa (~yehudasa@ip-66-33-206-8.dreamhost.com) has joined #ceph
[19:18] * morse (~morse@supercomputing.univpm.it) Quit (Remote host closed the connection)
[19:20] * cmccabe (~cmccabe@ has joined #ceph
[19:21] * morse (~morse@supercomputing.univpm.it) has joined #ceph
[19:24] * morse (~morse@supercomputing.univpm.it) Quit (Remote host closed the connection)
[19:38] * verwilst (~verwilst@dD576934D.access.telenet.be) has joined #ceph
[19:40] <bchrisman> I have client logs from running testceph (src/client/testceph.cc) here: http://pastebin.com/2HmtA1D4. It's hanging on attempting to write to a file it has opened. I'm wondering if there's some sort of problem in how I've called ceph_write??? the write starts after the message 'ceph_open: success' in those logs
[19:41] <bchrisman> ahh my open arguments may be incorrect.
[19:45] <Tv> bchrisman: how would that translate to a hang? i think all hangs sound like bugs..
[19:45] <Tv> (unless you shot a daemon in the head, etc)
[19:45] <bchrisman> sure yeah...
[19:45] <bchrisman> I'll fix the code then sort out reproducing the hang.
[19:46] <bchrisman> (client hang)
[19:46] <Tv> any mention of caps makes me think about the bugs gregaf was looking at..
[19:49] <sjust> sage: those commits look good
[19:50] <sage> sjust k
[20:11] * verwilst (~verwilst@dD576934D.access.telenet.be) Quit (Quit: Ex-Chat)
[20:16] * morse (~morse@supercomputing.univpm.it) has joined #ceph
[20:29] * ghaskins_mobile (~ghaskins_@66-189-113-47.dhcp.oxfr.ma.charter.com) has joined #ceph
[21:28] * sjustlaptop (~sam@ip-66-33-206-8.dreamhost.com) has joined #ceph
[21:46] * imcsk8 (~ichavero@ has joined #ceph
[21:48] * sjustlaptop (~sam@ip-66-33-206-8.dreamhost.com) Quit (Quit: Leaving.)
[21:56] * ghaskins_mobile (~ghaskins_@66-189-113-47.dhcp.oxfr.ma.charter.com) Quit (Quit: This computer has gone to sleep)
[22:03] <imcsk8> hello, i'm testing ceph 0.28 and i get this error: ERROR! old-style section name(s) found: mon0, osd0. Please use the new style section names that include a period ??can somebody help me?
[22:04] <imcsk8> i'm using a previous working config from 0.26
[22:04] <sjust> imcsk8: we changed the format a bit
[22:04] <sjust> it'll still work, but to silence the warning, I think you need to rename sections like: mon0 -> mon.0
[22:06] <admix> also, mon and mds already support mds.hostname/mon.hostname syntax, but osd requires osd.0 (number), will this change too?
[22:07] <sjust> admix: i believe you will still need osd.0 (number)
[22:07] <admix> ok
[22:07] <sjust> admix: mon and mds work the other way because the roles (0,1..etc) are assigned dynamically
[22:07] <admix> ah, thanks
[22:09] <admix> I'm currently testing ceph for two-node (mds/osd/mon) setup with external third mon node, and still getting used to it
[22:12] <imcsk8> sjust, thanks
[22:12] <sjust> imcsk8: yep
[22:36] * ghaskins_mobile (~ghaskins_@66-189-113-47.dhcp.oxfr.ma.charter.com) has joined #ceph
[22:40] * ghaskins_mobile (~ghaskins_@66-189-113-47.dhcp.oxfr.ma.charter.com) Quit ()
[22:51] * Yulya__ (~Yu1ya_@ip-95-220-183-143.bb.netbynet.ru) has joined #ceph
[22:58] * Yulya_ (~Yu1ya_@ip-95-220-174-84.bb.netbynet.ru) Quit (Ping timeout: 480 seconds)
[23:19] * lxo (~aoliva@ Quit (Read error: Connection reset by peer)
[23:32] * lxo (~aoliva@ has joined #ceph
[23:34] * votz (~votz@dhcp0020.grt.resnet.group.UPENN.EDU) has joined #ceph
[23:59] * votz (~votz@dhcp0020.grt.resnet.group.UPENN.EDU) Quit (Quit: Leaving)

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