#ceph IRC Log


IRC Log for 2011-09-21

Timestamps are in GMT/BST.

[0:54] * bchrisman (~Adium@70-35-37-146.static.wiline.com) Quit (Quit: Leaving.)
[1:54] * bchrisman (~Adium@c-98-207-207-62.hsd1.ca.comcast.net) has joined #ceph
[1:58] <bchrisman> anybody seen this build issue? http://pastebin.com/DeiDHW8e
[2:01] * danielf (~df@ Quit (Quit: danielf)
[2:03] <Tv|work> bchrisman: funky.. what does "git tag -l" say?
[2:05] <Tv|work> bchrisman: cat src/.git_version
[2:07] <Tv|work> i don't see how that'll ever become "ewdream/master"
[2:07] <Tv|work> oh now i do
[2:08] <Tv|work> but how did you get that file to have that content is still a bit of a mystery
[2:08] <Tv|work> the bug is caused by insufficient paranoia from sage
[2:09] <Tv|work> bchrisman: if you tell me enough of your git repo, i'll reproduce & fix, but without it's hard to pinpoint why & where the fix belongs
[2:15] * greglap (~Adium@aon.hq.newdream.net) has joined #ceph
[2:15] * greglap1 (~Adium@aon.hq.newdream.net) Quit (Read error: Connection reset by peer)
[2:17] * Tv|work (~Tv|work@aon.hq.newdream.net) Quit (Ping timeout: 480 seconds)
[2:17] * adjohn (~adjohn@ Quit (Remote host closed the connection)
[2:17] * adjohn (~adjohn@ has joined #ceph
[2:23] * greglap (~Adium@aon.hq.newdream.net) Quit (Quit: Leaving.)
[2:47] * joshd (~joshd@aon.hq.newdream.net) Quit (Quit: Leaving.)
[3:05] * cp (~cp@ Quit (Quit: cp)
[3:07] * cp (~cp@ has joined #ceph
[3:07] * cp (~cp@ Quit ()
[3:09] * adjohn is now known as Guest11086
[3:09] * adjohn (~adjohn@ has joined #ceph
[3:17] * Guest11086 (~adjohn@ Quit (Ping timeout: 480 seconds)
[3:50] * adjohn (~adjohn@ Quit (Quit: adjohn)
[6:55] * wido (~wido@rockbox.widodh.nl) Quit (Remote host closed the connection)
[16:56] -solenoid.oftc.net- *** Looking up your hostname...
[16:56] -solenoid.oftc.net- *** Checking Ident
[16:56] -solenoid.oftc.net- *** No Ident response
[16:56] -solenoid.oftc.net- *** Found your hostname
[16:56] * CephLogBot (~PircBot@rockbox.widodh.nl) has joined #ceph
[16:57] * lx0 (~aoliva@9YYAABMSG.tor-irc.dnsbl.oftc.net) Quit (Read error: Connection reset by peer)
[16:57] * wido (~wido@rockbox.widodh.nl) has joined #ceph
[16:57] * lx0 (~aoliva@9YYAABNFN.tor-irc.dnsbl.oftc.net) has joined #ceph
[17:02] * aliguori (~anthony@cpe-70-123-132-139.austin.res.rr.com) Quit (Quit: Ex-Chat)
[17:03] * aliguori (~anthony@cpe-70-123-132-139.austin.res.rr.com) has joined #ceph
[17:21] * adjohn (~adjohn@50-0-92-177.dsl.dynamic.sonic.net) Quit (Quit: adjohn)
[17:22] * greglap (~Adium@ has joined #ceph
[18:15] * sagewk (~sage@aon.hq.newdream.net) has joined #ceph
[18:34] * jojy (~jojyvargh@70-35-37-146.static.wiline.com) has joined #ceph
[18:41] * joshd (~joshd@aon.hq.newdream.net) has joined #ceph
[18:49] * Tv|work (~Tv|work@aon.hq.newdream.net) has joined #ceph
[19:16] * The_Bishop (~bishop@port-92-206-29-232.dynamic.qsc.de) Quit (Remote host closed the connection)
[19:21] <bchrisman> Tv|work: that info: http://pastebin.com/ffj3q17S
[19:22] * The_Bishop (~bishop@port-92-206-29-232.dynamic.qsc.de) has joined #ceph
[19:31] * aliguori (~anthony@cpe-70-123-132-139.austin.res.rr.com) Quit (Quit: Ex-Chat)
[19:31] * aliguori (~anthony@cpe-70-123-132-139.austin.res.rr.com) has joined #ceph
[19:35] <greglap> all: okay, looks like maybe I just did a merge and forgot about it on that push, so if it looks like github's set up right it probably is :)
[19:37] <Tv|work> bchrisman: so src/.git_version is supposed to have two lines
[19:37] <Tv|work> bchrisman: it's not currently updated atomically, and the code reading it does this kind of mistakes if the format isn't perfectly
[19:37] <Tv|work> bchrisman: did you perhaps have an aborted compile before this? like, hit control-c?
[19:38] <bchrisman> last I did was a full git clean
[19:38] <bchrisman> as far as I remember...
[19:38] <bchrisman> if I remove the file and run a make, are there other files I'd need to remove?
[19:39] <bchrisman> (this is being performed by some automation scripting)
[19:39] <Tv|work> bchrisman: basically, in src/check_version, either one of these failed:
[19:39] <Tv|work> cur=`git rev-parse HEAD 2>/dev/null; git describe 2>/dev/null`
[19:39] <Tv|work> echo "$cur" > $1
[19:39] <Tv|work> bchrisman: i *can* make that be more paranoid about atomicity etc, i'm just wondering if it's worthwhile
[19:39] <Tv|work> it should not happen from a git clean & make
[19:40] <Tv|work> it really really should not
[19:40] <Tv|work> it should require some sort of failure or abort
[19:40] <bchrisman> hmm… will check that out.
[19:42] <bchrisman> probably something in our automation
[19:43] <Tv|work> bchrisman: if you had a repo with no tags, that might happen
[19:46] * pruby (~tim@leibniz.catalyst.net.nz) Quit (Ping timeout: 480 seconds)
[19:47] * pruby (~tim@leibniz.catalyst.net.nz) has joined #ceph
[20:03] <bchrisman> Tv|work: ahh my apologies… I was doing a little bit of hacking around to try to get those version things to ignore local merge commits but didn't revert that when we reverting to just building directly from the github source… my bad.
[20:03] * greglap (~Adium@ Quit (Read error: Connection reset by peer)
[20:09] * greglap (~Adium@ has joined #ceph
[20:11] <Tv|work> bchrisman: ahh
[20:17] * cp (~cp@ has joined #ceph
[20:18] <cp> It seems strange the mkcephfs will create the necessary directories for mons but not for osds
[20:21] <greglap> cp: with the OSDs it matters a lot more where the data ends up; we don't want to be automatically creating directories on the root disk if you forgot to mount another disk there or something
[20:21] <cp> Ah, that makes sense
[20:21] <cp> I actually forgot to mount a disk just now :)
[20:37] <Tv|work> greglap: the silly part is, using the root of the mount point as your osd location means you can still forget to actually mount it..
[20:38] * votz (~votz@pool-108-52-121-23.phlapa.fios.verizon.net) has joined #ceph
[20:38] <greglap> you still have to actually create it, though
[20:39] <greglap> and I think under a lot of code paths that won't work on startup because the OSD looks for its store and doesn't find anything
[20:39] <greglap> but I guess that's true regardless...
[20:45] <Tv|work> oh i mean more mkdir /mountpoint ; forget to mount it; cosd --mkfs
[20:47] <greglap> yeah, it's not proof against failure, but it seems like a lot better than nothing to me — and I think with some btrfs configs it actually won't let you do even that
[21:16] <Tv|work> what's another word for example.. when you have many use cases and you pick one of them to be representative example..
[21:18] <Tv|work> exemplar sounds too fancy, but that's what i'm looking for
[21:19] <Tv|work> gonna go with "Example:" in italics
[21:34] * bchrisman (~Adium@c-98-207-207-62.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[21:59] * cp (~cp@ Quit (Quit: cp)
[22:20] * bchrisman (~Adium@ has joined #ceph
[22:47] <Tv|work> new docs: http://ceph.newdream.net/docs/latest/start/ (only the object storage part, for now)
[22:47] <Tv|work> lots of restructuring around it
[22:48] <Tv|work> e.g. http://ceph.newdream.net/docs/latest/rec/
[22:50] <Tv|work> having more api docs would be good
[22:52] * cp (~cp@ has joined #ceph
[22:53] * _Tassadar (~tassadar@tassadar.xs4all.nl) has joined #ceph
[22:53] <_Tassadar> 'evening
[22:54] * sjustlaptop (~sam@aon.hq.newdream.net) has joined #ceph
[22:56] <Tv|work> g'day
[22:59] * cp (~cp@ Quit (Quit: cp)
[23:03] * DanielFriesen (~dantman@S0106001731dfdb56.vs.shawcable.net) has joined #ceph
[23:03] * Dantman (~dantman@S0106001731dfdb56.vs.shawcable.net) Quit (Read error: Connection reset by peer)
[23:09] * adjohn (~adjohn@ has joined #ceph
[23:14] * cp (~cp@ has joined #ceph
[23:15] * pruby (~tim@leibniz.catalyst.net.nz) Quit (Ping timeout: 480 seconds)
[23:17] * pruby (~tim@leibniz.catalyst.net.nz) has joined #ceph
[23:35] * danielf (~df@ has joined #ceph
[23:36] * adjohn (~adjohn@ Quit (Quit: adjohn)
[23:39] * Nightdog (~karl@190.84-48-62.nextgentel.com) has joined #ceph

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