Re: [PATCH 2/8] Basic framework for lock manager plugins

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

 



On Mon, Nov 29, 2010 at 02:49:26PM -0500, David Teigland wrote:
> On Wed, Nov 24, 2010 at 03:08:41PM -0500, David Teigland wrote:
> > LOCK_MODE_NONE - the lock manager is unused, and it's up to the
> > application to do its own locking or coordination when accessing the
> > resource, e.g. if there's a clustered db or fs on the device that does its
> > own coordination.  You seem to be calling this SHARED, but there's
> > actually no lock or exclusion that a lock manager would implement for this
> > kind of usage.
> 
> I failed to mention that with a more elaborate set of lock modes like a
> dlm uses, the CW "concurrent write" mode fits the use case of leaving the
> coordination to the application.  Simpler lock managers only support
> Shared or Exclusive, of course, in which case "None" would apply.

This kind of difference in capabilities between lock managers is a good
example of why I only provide the disk sharing mode to the plugin API,
and leave choice of locking mode to the plugin.

Daniel

--
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]