On 05/02/2012 05:11 AM, Daniel P. Berrange wrote:
On Tue, May 01, 2012 at 03:10:42PM -0400, Laine Stump wrote:
> This patch is one alternative to solve the problem detailed in:
>
>
https://bugzilla.redhat.com/show_bug.cgi?id=816465
>
> Some other unidentified library in use by libvirtd (in another thread)
> is apparently temporarily binding to a NETLINK_ROUTE raw socket with
> an address of "pid of libvirtd" during startup.
Can you identify this library.
I made a few attempts, but didn't have any luck and decided to post
these patches based on the other evidence I'd gathered. I agree that I
would much prefer understanding who is doing this, even if it doesn't
change the workaround method.
It should be possible to do so using
systemtap without all that much trouble.
My full experience with systemtap is using some of the examples from
your blog posting on the topic :-)
I would love to figure this out, though. The complicating factor I can
see (aside from me needing to learn how to write a systemtap script) is
that in this case stap needs to be run on a daemonizing process, from
the very beginning. If you can give me any better advice than "go read
the systemtap website", please do.
In the meantime, the people who developed the netlink event service code
have agreed to test out the patch in alternative 2 to verify that it
doesn't break communication between lldpad and libvirtd.