#ceph IRC Log


IRC Log for 2010-12-08

Timestamps are in GMT/BST.

[0:21] * alexxy (~alexxy@ Quit (Read error: Connection reset by peer)
[0:25] * alexxy (~alexxy@ has joined #ceph
[1:26] * DJLee (82d8d198@ircip3.mibbit.com) has joined #ceph
[1:27] <DJLee> hey Sage, Yehuda et al,
[1:27] <DJLee> regarding the bug track http://tracker.newdream.net/issues/479
[1:27] <DJLee> please ignore it for now, I think I kinda got it,
[1:29] <DJLee> I think you get the hang when you mount ceph on the same PC that's running the ceph system.
[1:30] <DJLee> something to do with network/local issues
[1:55] * alexxy[home] (~alexxy@ has joined #ceph
[1:58] * alexxy (~alexxy@ Quit (Ping timeout: 480 seconds)
[2:09] <bchrisman> my btrfs apparently doesn't support the user_xattr flag: btrfs: unrecognized mount option 'user_xattr'
[2:11] <DeHackEd> that's needed for ext3/4, maybe btrfs doesn't actually need it?
[2:11] <bchrisman> ahh
[2:11] <DeHackEd> man page is a bit vague
[2:11] <bchrisman> noticed it referenced on the ceph wiki
[2:12] <DeHackEd> and the man page says "Some file systems, such as ext2/3 and reiserfs, require...." implying maybe it's not a total necessity
[2:13] <cmccabe1> bchrisman: btrfs has user xattrs on by default. You don't need user_xattr
[2:13] * greglap (~Adium@ has joined #ceph
[2:14] <bchrisman> tx
[2:21] * greglap1 (~Adium@ has joined #ceph
[2:26] * greglap (~Adium@ Quit (Ping timeout: 480 seconds)
[2:30] * sjust (~sam@ip-66-33-206-8.dreamhost.com) Quit (Ping timeout: 480 seconds)
[2:31] <bchrisman> this is not quite clear to me… can you create a cephfs filesystem with mkcephfs *before* you startup the mon/mds/osd daemons? Looks like you can...?
[2:31] <DJLee> I think that's the step
[2:41] * DJLee (82d8d198@ircip3.mibbit.com) Quit (Quit: http://www.mibbit.com ajax IRC Client)
[2:45] * joshd (~joshd@ip-66-33-206-8.dreamhost.com) Quit (Quit: Leaving.)
[3:06] * bchrisman (~Adium@70-35-37-146.static.wiline.com) Quit (Quit: Leaving.)
[3:11] * greglap1 (~Adium@ Quit (Quit: Leaving.)
[3:27] * cmccabe1 (~cmccabe@adsl-75-37-25-40.dsl.pltn13.sbcglobal.net) has left #ceph
[5:14] * bchrisman (~Adium@c-24-130-226-22.hsd1.ca.comcast.net) has joined #ceph
[6:45] * ijuz (~ijuz@p4FFF558B.dip.t-dialin.net) Quit (Ping timeout: 480 seconds)
[6:54] * ijuz (~ijuz@p579994E2.dip.t-dialin.net) has joined #ceph
[7:00] * Guest248 (quasselcor@bas11-montreal02-1128535712.dsl.bell.ca) Quit (Remote host closed the connection)
[7:02] * bbigras (quasselcor@bas11-montreal02-1128535712.dsl.bell.ca) has joined #ceph
[7:03] * bbigras is now known as Guest267
[7:07] * Yoric (~David@dau94-10-88-189-211-192.fbx.proxad.net) has joined #ceph
[7:34] * f4m8_ is now known as f4m8
[9:05] * Yoric (~David@dau94-10-88-189-211-192.fbx.proxad.net) Quit (Quit: Yoric)
[10:00] * co_bth_ce (~co_malass@ has joined #ceph
[10:00] * co_bth_ce (~co_malass@ has left #ceph
[10:13] * verwilst (~verwilst@router.begen1.office.netnoc.eu) has joined #ceph
[10:25] * Yoric (~David@ has joined #ceph
[11:14] * allsystemsarego (~allsystem@ has joined #ceph
[11:15] * verwilst (~verwilst@router.begen1.office.netnoc.eu) Quit (Read error: Connection reset by peer)
[11:16] * verwilst (~verwilst@router.begen1.office.netnoc.eu) has joined #ceph
[13:31] * MarkN1 (~nathan@mail.zomojo.com) Quit (Ping timeout: 480 seconds)
[13:42] * MarkN (~nathan@mail.zomojo.com) has joined #ceph
[15:36] * verwilst (~verwilst@router.begen1.office.netnoc.eu) Quit (reticulum.oftc.net galapagos.oftc.net)
[15:36] * DLange (~DLange@dlange.user.oftc.net) Quit (reticulum.oftc.net galapagos.oftc.net)
[15:38] * verwilst (~verwilst@router.begen1.office.netnoc.eu) has joined #ceph
[15:38] * DLange (~DLange@dlange.user.oftc.net) has joined #ceph
[16:16] * __jt__ (~james@aca-104000.nat.emory.edu) has joined #ceph
[16:30] * Yoric_ (~David@ has joined #ceph
[16:30] * Yoric (~David@ Quit (Read error: Connection reset by peer)
[16:30] * Yoric_ is now known as Yoric
[16:51] * greglap (~Adium@ has joined #ceph
[17:35] * greglap (~Adium@ Quit (Quit: Leaving.)
[17:38] * bchrisman (~Adium@c-24-130-226-22.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[17:58] * sagephone (~yaaic@ has joined #ceph
[18:07] * verwilst (~verwilst@router.begen1.office.netnoc.eu) Quit (Quit: Ex-Chat)
[18:32] * bchrisman (~Adium@70-35-37-146.static.wiline.com) has joined #ceph
[18:42] * sagephone (~yaaic@ has left #ceph
[18:59] * joshd (~joshd@ip-66-33-206-8.dreamhost.com) has joined #ceph
[19:08] * Yoric (~David@ Quit (Quit: Yoric)
[19:11] * sjust (~sam@ip-66-33-206-8.dreamhost.com) has joined #ceph
[19:15] * alexxy (~alexxy@masq246.gtn.ru) has joined #ceph
[19:15] * alexxy[home] (~alexxy@ Quit (Read error: Connection reset by peer)
[19:18] * alexxy (~alexxy@masq246.gtn.ru) Quit ()
[19:18] * alexxy (~alexxy@ has joined #ceph
[19:37] * cmccabe (~cmccabe@dsl081-243-128.sfo1.dsl.speakeasy.net) has joined #ceph
[19:58] <sagewk> bchrisman: the ~rc in the version will go away in the actual release
[20:03] * pruby (~tim@leibniz.catalyst.net.nz) Quit (Remote host closed the connection)
[20:05] * pruby (~tim@leibniz.catalyst.net.nz) has joined #ceph
[20:15] <bchrisman> sure.. yeah
[20:31] <bchrisman> I put those packaging issues into the bug tracker… trivial or otherwise they were the only issues which stop it from building packages cleanly on rhel6b
[20:31] <bchrisman> as I mentioned, I got a cephfs running last night… looks pretty sweet…
[20:34] <sagewk> did you have a problem with cryptopp-devel? is that in rhel6?
[20:34] <bchrisman> it's not in rhel6, but it is in rpmforge (I didn't check epel for rhel6beta.. just noticed that later)
[20:34] <bchrisman> I had no problems with it.
[20:34] <sagewk> k
[20:35] <bchrisman> I presume that's used for building keys as well?
[20:35] <bchrisman> It looks like I used all that functionality and it works.
[20:35] <sagewk> yeah for aes and some hmac stuff?
[20:35] <sagewk> excellent :)
[20:37] <bchrisman> overall, impressive rig… I had a n00b-moment there with the iptables issue.. wondering if there's a good way to detect that and issue a message "check yer iptables"… I should've guessed immediately from the connection failure of the mons reported in /var/log/messages… but I imagine a lot of users could get trapped by that.
[20:38] <bchrisman> (particularly the ones who haven't been bitten in the butt by iptables and selinux to know to shut all that stuff off until you've got everything running)
[20:43] <cmccabe> bchrisman: it's hard to detect that an error is caused by selinux, because you just get EPERM
[20:45] <bchrisman> yeah… iptables is probably a bit easier to detect.
[20:46] <cmccabe> bchrisman: you can detect that iptables is on I guess. Although the problem could also be the iptables setup on another node.
[20:47] * DJLee (82d8d198@ircip1.mibbit.com) has joined #ceph
[20:47] <cmccabe> bchrisman: at the minimum, we could put a suggestion on the wiki to disable that stuff until everything else is working
[20:52] * pruby (~tim@leibniz.catalyst.net.nz) Quit (Remote host closed the connection)
[20:53] * pruby (~tim@leibniz.catalyst.net.nz) has joined #ceph
[20:54] <bchrisman> cmccabe: something in the mon log about failing to connect to other monitors? it shows up in /var/log/messages as test10 kernel: libceph: mon0 connection failed
[20:55] <bchrisman> cmccabe: which is a kernel message… so that's being reported by the ceph kernel module client
[20:56] <bchrisman> cmccabe: this is the message in mon log: -- >> pipe(0x1f6cb40 sd=6 pgs=0 cs=0 l=0).fault first fault
[20:57] <bchrisman> cmccabe: basically where it normally outputs, pipe.accept connect_seq.. the alternative is 'fault first fault' in the case of failing to connect (or hanging and waiting for connection to go through?)
[21:03] <cmccabe> bchrisman: fault first fault is something that I see a lot when there is a connectivity problem
[21:16] <bchrisman> cmccabe: that message would be a good place to warn about iptables.
[21:16] <bchrisman> ?
[21:16] <bchrisman> or does that happen too often during say.. outages?
[21:17] <cmccabe> bchrisman: gregaf knows the messenger better than I do-- he might have an opinion on this
[21:17] <cmccabe> bchrisman: the problem with changing the fault message is that it occurs for a lot of reasons, most of which don't have anything to do with iptables
[21:18] <bchrisman> yeah
[21:18] <cmccabe> on the other hand, I don't really have a better idea this time... :\
[21:19] <cmccabe> I guess potentially if we fail to pass the first traffic that we try to send, we could issue a different message than usual
[21:20] <cmccabe> actually, this might be better to handle at the distribution level. Doesn't red hat have some way that installing packages can modify the firewall?
[21:22] <bchrisman> will check into that
[21:52] * DJLee (82d8d198@ircip1.mibbit.com) Quit (Quit: http://www.mibbit.com ajax IRC Client)
[21:55] * allsystemsarego_ (~allsystem@ has joined #ceph
[22:03] * allsystemsarego (~allsystem@ Quit (Ping timeout: 480 seconds)
[22:44] * __jt__ (~james@aca-104000.nat.emory.edu) Quit (Quit: leaving)
[22:45] * __jt__ (~james@jamestaylor.org) has joined #ceph
[23:24] * allsystemsarego_ (~allsystem@ Quit (Quit: Leaving)
[23:29] * Yoric (~David@dau94-10-88-189-211-192.fbx.proxad.net) has joined #ceph

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