On Mon, Mar 24, 2014 at 14:41:52 -0600, Eric Blake wrote:
On 03/24/2014 07:50 AM, Jiri Denemark wrote:
> Some lock failure actions do not make any sense in combination with
> sanlock driver. Let's just report an error if someone tries to use them
> instead of causing unexpected and possibly quite bad thing to happen.
ACK series. However, are we missing some documentation patches to match?
I think the documentation is fine as it is:
The on_lockfailure element (since 1.0.0) may be used to configure what
action should be taken when a lock manager loses resource locks. The
following actions are recognized by libvirt, although not all of them
need to be supported by individual lock managers.
...
I mean it would be nice to explicitly state what actions are supported
by individual lock managers but doing it manually seems to fragile. Not
to mention we don't document such things elsewhere either :-)
I pushed the patches, thanks for the review.
Jirka