[PATCH] CTDB: Clean up interface between recovery daemon and IP takeover runs

[Amitay Isaacs]

On Thu, May 12, 2016 at 3:55 PM, Martin Schwenke <martin at meltin.net> wrote:

> By the end of this patch set there are only 2 places where IP takeover
> runs are done.  Banning is handled consistently in both calls but is
> also a little more conservative, with only the worst offender receiving
> banning credits.  This makes the code almost ready for an external
> IP takeover run helper (much like the recovery helper) to be
> implemented.
>
> There are a couple of other cleanups that remove unnecessary special
> cases that are already handled more generally:
>
> * ctdb-takeover: Do not set node unhealthy when "takeip" fails
>
>   This can only really be done with the banning consistency
>   improvements.
>
> * ctdb-recoverd: Drop explicit check to flag takeover run needed
>
>   This one is worth a closer look to make sure my assumptions are
>   correct.
>
> Please review and maybe push...
>
> peace & happiness,
> martin
>

Pushed to autobuild.

Amitay.