#ceph IRC Log

Index

IRC Log for 2011-05-17

Timestamps are in GMT/BST.

[0:04] * greglap (~Adium@12.248.40.138) has joined #ceph
[0:05] * greglap (~Adium@12.248.40.138) Quit ()
[0:08] <darkfaded> i need to make a netbootable linux with ceph
[0:08] <darkfaded> because i got 27(!!!) servers to play with
[0:08] <darkfaded> but almost no time
[0:18] * MarkN (~nathan@59.167.240.178) has left #ceph
[0:32] * aliguori (~anthony@cpe-70-123-132-139.austin.res.rr.com) Quit (Remote host closed the connection)
[0:39] * aliguori (~anthony@cpe-70-123-132-139.austin.res.rr.com) has joined #ceph
[0:58] * joshd (~joshd@ip-66-33-206-8.dreamhost.com) has joined #ceph
[1:21] * greglap (~Adium@cpe-76-170-84-245.socal.res.rr.com) has joined #ceph
[1:25] * greglap (~Adium@cpe-76-170-84-245.socal.res.rr.com) Quit ()
[1:35] <bchrisman> hmm.. not at all clear from the readdir_r manpage what it's supposed to return at the end of a directory??? going with the Client definition then..
[1:35] * Tv (~Tv|work@ip-66-33-206-8.dreamhost.com) Quit (Ping timeout: 480 seconds)
[1:36] <cmccabe> bchrisman: it's supposed to return NULL at the end of a directory right?
[1:36] <cmccabe> NULL in *result, 0 as the return code
[1:37] <joshd> rageguy: is your qemu happy?
[1:38] <bchrisman> cmccabe: readdir is??? readdir_r returns an int
[1:38] <cmccabe> bchrisman: that int will be a 0 on end-of-directory
[1:38] <cmccabe> bchrisman: and *result will be NULL
[1:39] <bchrisman> cmccabe: 0 + NULL pointer on indirect return mechanism?
[1:39] <cmccabe> yep
[1:40] <cmccabe> before you use readdir_r, though, read this: http://womble.decadent.org.uk/readdir_r-advisory.html
[1:40] * Dantman (~dantman@S0106001eec4a8147.vs.shawcable.net) Quit (Read error: Connection reset by peer)
[1:40] <bchrisman> yeah???
[1:41] * Dantman (~dantman@S0106001eec4a8147.vs.shawcable.net) has joined #ceph
[1:41] <cmccabe> on Linux, regular readdir is threadsafe unless you share the DIR* among multiple threads
[1:44] <cmccabe> allocating the correct buffer for readdir_r is actually pretty difficult
[1:44] <cmccabe> I guess that is still the most portable way to go, but man is it ugly
[1:45] <bchrisman> yeah
[1:46] <cmccabe> I think it's less efficient too, at least on linux
[2:10] <djlee> guys, when writing a series of files (seq. write), how does the actual object gets written in the disk? I don't think files/objects are written in a pure sequential manner, correct?
[2:47] * greglap (~Adium@198.228.210.243) has joined #ceph
[3:02] * sagelap (~sage@12.248.40.138) Quit (Ping timeout: 480 seconds)
[3:05] * joshd (~joshd@ip-66-33-206-8.dreamhost.com) Quit (Quit: Leaving.)
[3:08] * cmccabe (~cmccabe@208.80.64.174) has left #ceph
[3:18] * bchrisman (~Adium@70-35-37-146.static.wiline.com) Quit (Quit: Leaving.)
[3:18] <djlee> greg, are you available?
[3:23] * greglap (~Adium@198.228.210.243) Quit (Read error: Connection reset by peer)
[4:09] * alexxy (~alexxy@79.173.81.171) Quit (Ping timeout: 480 seconds)
[4:55] * djlee_ (~dlee064@des152.esc.auckland.ac.nz) has joined #ceph
[5:02] * djlee (~dlee064@des152.esc.auckland.ac.nz) Quit (Ping timeout: 480 seconds)
[5:12] * votz (~votz@dhcp0020.grt.resnet.group.upenn.edu) Quit (Quit: Leaving)
[6:04] * joshd (~jdurgin@adsl-75-28-69-238.dsl.irvnca.sbcglobal.net) has joined #ceph
[6:05] * neurodrone (~neurodron@cpe-76-180-162-12.buffalo.res.rr.com) has joined #ceph
[6:06] * greglap (~Adium@65-122-15-169.dia.static.qwest.net) has joined #ceph
[6:09] * lxo (~aoliva@201.82.32.113) Quit (Ping timeout: 480 seconds)
[6:14] * Dantman (~dantman@S0106001eec4a8147.vs.shawcable.net) Quit (Ping timeout: 480 seconds)
[6:24] * bchrisman (~Adium@c-98-207-207-62.hsd1.ca.comcast.net) has joined #ceph
[6:42] * greglap (~Adium@65-122-15-169.dia.static.qwest.net) Quit (Quit: Leaving.)
[6:51] * lxo (~aoliva@201.82.32.113) has joined #ceph
[7:57] * neurodrone_ (~neurodron@cpe-76-180-162-12.buffalo.res.rr.com) has joined #ceph
[7:57] * neurodrone (~neurodron@cpe-76-180-162-12.buffalo.res.rr.com) Quit (Read error: Connection reset by peer)
[7:57] * neurodrone_ is now known as neurodrone
[8:03] * lxo (~aoliva@201.82.32.113) Quit (Ping timeout: 480 seconds)
[8:03] * joshd (~jdurgin@adsl-75-28-69-238.dsl.irvnca.sbcglobal.net) Quit (Quit: Leaving.)
[8:31] * Jiaju (~jjzhang@222.126.194.154) has joined #ceph
[9:01] * yehuda_hm (~yehuda@bzq-79-182-97-171.red.bezeqint.net) Quit (Ping timeout: 480 seconds)
[9:18] * allsystemsarego (~allsystem@188.27.167.240) has joined #ceph
[9:35] * alexxy (~alexxy@79.173.81.171) has joined #ceph
[9:44] * neurodrone_ (~neurodron@cpe-76-180-162-12.buffalo.res.rr.com) has joined #ceph
[9:44] * neurodrone (~neurodron@cpe-76-180-162-12.buffalo.res.rr.com) Quit (Read error: Connection reset by peer)
[9:44] * neurodrone_ is now known as neurodrone
[9:45] * neurodrone_ (~neurodron@cpe-76-180-162-12.buffalo.res.rr.com) has joined #ceph
[9:45] * neurodrone (~neurodron@cpe-76-180-162-12.buffalo.res.rr.com) Quit (Read error: Connection reset by peer)
[9:45] * neurodrone_ is now known as neurodrone
[10:06] * yehuda_hm (~yehuda@192.116.104.89) has joined #ceph
[10:06] * neurodrone (~neurodron@cpe-76-180-162-12.buffalo.res.rr.com) Quit (Quit: zzZZZZzz)
[10:08] * votz (~votz@dhcp0020.grt.resnet.group.upenn.edu) has joined #ceph
[10:24] * lxo (~aoliva@201.82.32.113) has joined #ceph
[10:59] <admix> hey guys, any info on 0.28 release?
[12:42] * greglap (~Adium@mobile-198-228-226-123.mycingular.net) has joined #ceph
[13:17] <greglap> admix: I think sage will probably cut it today
[13:18] <greglap> djlee: sorry, I'm on vacation (actually sitting in an airport on my way there now...)
[13:18] <admix> greglap: great, thanks
[13:26] <rageguy> joshd: no, I put it on hold for a while
[13:26] * alexxy (~alexxy@79.173.81.171) Quit (Remote host closed the connection)
[13:27] * alexxy (~alexxy@79.173.81.171) has joined #ceph
[14:13] * rsharpe (~Adium@70-35-37-146.static.wiline.com) Quit (Ping timeout: 480 seconds)
[14:34] * wonko_be (bernard@november.openminds.be) has joined #ceph
[14:34] * greglap (~Adium@mobile-198-228-226-123.mycingular.net) Quit (Read error: Connection reset by peer)
[14:36] * neurodrone (~neurodron@cpe-76-180-162-12.buffalo.res.rr.com) has joined #ceph
[15:56] * Meths_ (rift@2.27.73.202) has joined #ceph
[16:03] * Meths (rift@2.25.214.205) Quit (Ping timeout: 480 seconds)
[16:45] * Dantman (~dantman@96.48.212.149) has joined #ceph
[17:20] * aliguori (~anthony@cpe-70-123-132-139.austin.res.rr.com) Quit (Quit: Ex-Chat)
[17:20] * yehuda_hm (~yehuda@192.116.104.89) Quit (Ping timeout: 480 seconds)
[17:43] * aliguori (~anthony@32.97.110.59) has joined #ceph
[17:50] * MK_FG (~MK_FG@188.226.51.71) Quit (Quit: o//)
[17:57] * MK_FG (~MK_FG@188.226.51.71) has joined #ceph
[17:57] * MK_FG (~MK_FG@188.226.51.71) Quit ()
[18:00] * MK_FG (~MK_FG@188.226.51.71) has joined #ceph
[18:02] * MK_FG (~MK_FG@188.226.51.71) Quit ()
[18:02] * tjikkun (~tjikkun@2001:7b8:356:0:204:bff:fe80:8080) Quit (Ping timeout: 480 seconds)
[18:03] * MK_FG (~MK_FG@188.226.51.71) has joined #ceph
[18:04] * MK_FG (~MK_FG@188.226.51.71) Quit ()
[18:05] * sagelap (~sage@ip-66-33-206-8.dreamhost.com) has joined #ceph
[18:06] * MK_FG (~MK_FG@188.226.51.71) has joined #ceph
[18:13] * MK_FG (~MK_FG@188.226.51.71) Quit (Quit: o//)
[18:21] * rsharpe (~Adium@70-35-37-146.static.wiline.com) has joined #ceph
[18:27] * sagelap (~sage@ip-66-33-206-8.dreamhost.com) Quit (Ping timeout: 480 seconds)
[18:28] * Tv (~Tv|work@ip-66-33-206-8.dreamhost.com) has joined #ceph
[18:32] * MK_FG (~MK_FG@188.226.51.71) has joined #ceph
[18:34] * Meths_ is now known as Meths
[18:47] * philipgian (~philipgia@athedsl-4504336.home.otenet.gr) has joined #ceph
[19:10] * joshd (~joshd@ip-66-33-206-8.dreamhost.com) has joined #ceph
[19:21] * cmccabe (~cmccabe@c-24-23-254-199.hsd1.ca.comcast.net) has joined #ceph
[19:29] <Tv> conference room is full of ppl
[19:41] * tjikkun (~tjikkun@2001:7b8:356:0:204:bff:fe80:8080) has joined #ceph
[19:45] <cmccabe> tv: that's good. if it were full of eels, there might be a problem
[19:45] <Tv> well the glass wall might make it a nice aquarium
[19:46] <cmccabe> just once, I'd like to see a fishbowl conference room used for that
[20:37] * yehuda_hm (~yehuda@bzq-79-182-97-171.red.bezeqint.net) has joined #ceph
[20:38] <yehuda_hm> cmccabe: if it was a hovercraft it'd be something completely different
[20:40] * aliguori (~anthony@32.97.110.59) Quit (Ping timeout: 480 seconds)
[20:50] * aliguori (~anthony@32.97.110.64) has joined #ceph
[21:06] * aliguori (~anthony@32.97.110.64) Quit (Ping timeout: 480 seconds)
[21:15] * aliguori (~anthony@32.97.110.59) has joined #ceph
[21:24] * Dantman (~dantman@96.48.212.149) Quit (Ping timeout: 480 seconds)
[21:30] * aliguori (~anthony@32.97.110.59) Quit (Quit: Ex-Chat)
[21:34] * Dantman (~dantman@S0106001731dfdb56.vs.shawcable.net) has joined #ceph
[21:40] * Yulya__ (~Yu1ya_@ip-95-220-188-224.bb.netbynet.ru) has joined #ceph
[21:46] * Yulya_ (~Yu1ya_@ip-95-220-157-30.bb.netbynet.ru) Quit (Ping timeout: 480 seconds)
[22:05] * aliguori (~anthony@32.97.110.59) has joined #ceph
[22:07] * sagelap (~sage@ip-66-33-206-8.dreamhost.com) has joined #ceph
[22:29] * jjchen (~jjchen@lo4.cfw-a-gci.greatamerica.corp.yahoo.com) has joined #ceph
[22:29] * sagelap (~sage@ip-66-33-206-8.dreamhost.com) Quit (Ping timeout: 480 seconds)
[22:38] * bchrisman (~Adium@c-98-207-207-62.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[22:44] * Juul_ (~Juul@131.243.47.57) has joined #ceph
[22:56] * Juul_ (~Juul@131.243.47.57) Quit (Quit: Leaving)
[23:02] <lxo> woah, the Gbit network wasn't enough to avoid a too-large logm message that prevents recovery after a failure that left only one of the three monitors active for several hours
[23:02] <lxo> nearly 100MB of requests for elections and more. anyone want to have a look?
[23:09] <cmccabe> I think the monitors need a quorum to operate
[23:09] <cmccabe> and I think that's 50% of them
[23:11] <lxo> yep. so 1 out of 3 is not enough, but it keeps logging messages it receives
[23:12] <cmccabe> I think so. Might be good to check with sage to confirm though
[23:12] <lxo> the problem is that when quorum is established, it tries to relay all this junk to the other monitors, but it never succeeds because it takes too long, and the others call for new elections before receiving and committing the huge file
[23:13] * sagelap (~sage@ip-66-33-206-8.dreamhost.com) has joined #ceph
[23:13] <cmccabe> sage might have an idea about the monitor stuff
[23:14] <lxo> we had an open bug about this, but it was thought that my upgrading the network from 100Mbps to 1Gbps would avoid the problem. so the software proved us wrong, generating a hundred (as opposed to ten) megabytes of logs that need recovering
[23:14] <cmccabe> a hundred megs should still only take a second or two to send over gigabit
[23:15] <cmccabe> I guess the problem may lie elsewhere, like in the parsing or handling of those logs
[23:15] <lxo> I moved the file aside on all monitors to let them recover. it had already been propagated to all monitors, but they kept resending it over and over after each election
[23:18] <lxo> bug 943
[23:20] <cmccabe> lxo: unfortunately greg is on vacation at the moment
[23:21] <lxo> no problem. I'll save the logs and add a note to the bug that it happened again
[23:49] <sagewk> tv: the x86_64 gitbuilder looks frozen
[23:49] <Tv> sagewk: checking
[23:50] <sagewk> er, stale i guess
[23:50] <Tv> yeah it's hung somewhere, has been for two days
[23:51] <Tv> in git fetch
[23:51] <Tv> which means the hang is probably on ceph.newdream.net, instead... checking
[23:52] <cmccabe> someone complained about the git server on the ML
[23:52] <cmccabe> DNS glitch?
[23:52] <Tv> whee git daemons that have been there since Feb 9th
[23:53] <Tv> and one from 2010
[23:53] <Tv> nothing that would match the May 15th date, though
[23:54] <sagewk> huh
[23:54] <Tv> well i kicked the git fetch, for now
[23:54] <Tv> i'll look at ceph.newdream.net more soon
[23:55] <Tv> trying to inject my topic into openstack meeting agenda ;)
[23:55] * sagelap (~sage@ip-66-33-206-8.dreamhost.com) Quit (Ping timeout: 480 seconds)
[23:57] <sagewk> tv: is there a hung sepia run i can look at?
[23:58] <Tv> sagewk: the last few runs had dbench complete ok, but then failed due to some oddity in autotest i haven't traced down (current plan: avoid that code, autotest barriers are not nice anyway, i'll replace them with my rpc stuff)
[23:59] <sagewk> ok. let me know next time you hit something i can look at

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