Re: testing libvirtd: don't hard-code $HOME/.libvirt/libvirt-sock

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



"Richard W.M. Jones" <rjones@xxxxxxxxxx> wrote:

> Jim Meyering wrote:
>> I've written a test to exercise libvirtd a little.
>> Unfortunately, if I'm already running libvirtd, that test fails
>> and I can't do anything about it:
>>
>>   $ ./libvirtd & ./libvirtd
>>   Failed to bind socket to '@/j/.libvirt/libvirt-sock': Address already in use
>>   [Exit 2]
>>
>> Any objection to providing an option to libvirtd that specifies
>> that directory (also used to form server->logDir)?
>>
>> It feels like it should come from the config file, but I haven't
>> looked at details yet.
>
> Yes it should be possible to set these, but isn't at the moment.
> Check out qemud/qemud.c function qemudInitPaths.
>
> Another possibility is to have libvirtd running as root, and libvirtd
> running as you, in which case they _shouldn't_ clash ...

The point is that if I leave one running as me, it shouldn't
interfere with "make check".  And if root runs "make check" (yes, I
know that doing that is totally crazy, but people do it all the time,
in spite of warnings), that shouldn't fail either.

--
Libvir-list mailing list
Libvir-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/libvir-list

[Index of Archives]     [Virt Tools]     [Libvirt Users]     [Lib OS Info]     [Fedora Users]     [Fedora Desktop]     [Fedora SELinux]     [Big List of Linux Books]     [Yosemite News]     [KDE Users]     [Fedora Tools]