=== Spads_ is now known as Spads
[01:45] <hallyn> arges: jdstrand:
[01:45] <hallyn>    - d/rules,control: Enable use of lttng for userspace tracing.
[01:45] <hallyn>  -- James Page <james.page@ubuntu.com>  Fri, 13 Mar 2015 07:42:45 +0000
[01:45] <hallyn> jamespage: ^  this is causing trouble (bug 1434999 )...
[01:45] <ubottu> bug 1432644 in libvirt (Ubuntu) "duplicate for #1434999 VM permanently tries to read /dev/shm/lttng-ust-wait-5" [High,Confirmed] https://launchpad.net/bugs/1432644
[01:47] <hallyn> meh, i think a deny rule is going to be the way to go
[01:49] <hallyn> i don't think removing the dependency is going to be acceptable
[01:51] <sarnold> at least the deny could be undone locally if someone wanted to use the tracing capabilities
[01:51] <sarnold> rebuilding to put it back would be a lot more effort
[02:33] <hallyn> yeah.  ok, i guess i'll go ahead and push with that fix
[02:34] <hallyn> hm, but what's the path
[02:34] <hallyn> /dev/shm/lttng* ?
[02:37] <hallyn> http://paste.ubuntu.com/10692865/
[02:53] <hallyn> weird thing is it seems like every error message documented is lttng-ust-wait-5.  something magical about 5?
[02:56] <hallyn> oh, there is - ./include/lttng/ust-abi.h:45:#define LTTNG_UST_ABI_MAJOR_VERSION                5
[02:57] <hallyn> smb: i pushed a (trivial) libvirt update, just fyi
[03:25] <sarnold> hallyn: ah, thanks, that -5 was driving me crazy :)
[03:28] <hallyn> i still left it as lttng-ust-wait-* in case major version changes :)
[03:29] <sarnold> having seent oo much of python-[123].[5678901234].. kinds of things, a nice * is sometimes perfect :)
[03:31] <hallyn> all right, to bed with me - \o
[03:32] <sarnold> nn
[04:15] <jdstrand> hallyn: interesting-- I was under the impression that the denial was causing vms to not start. I'm of course fine with an explicit deny if it is a harmless denial
=== iulian is now known as Guest32200
=== gusnan is now known as Guest15937
=== Guest15937 is now known as gusnan
=== vdagli is now known as hephaestus
[14:08] <hallyn> jdstrand: oh, maybe i misunderstood.
[14:08] <hallyn> if that's the case then i don' tknow how to handle it.
[14:08] <hallyn> i don't think disabling it in ceph will be acceptable
=== ara is now known as Guest40924
=== JanC_ is now known as JanC
=== ming is now known as Guest87143
=== Guest90957 is now known as mfisch