[ClusterLabs] Reduce failover time by concurent stop of 2 RG

[Ken Gaillot]

On Thu, 2022-01-27 at 13:26 +0100, John Gogu wrote:
> 
> 
> Hello group,
> I need a hint from you in regards of  the following situation:
> 
> GroupA
>   ResourceA1
>   ResourceA2
> 
> GroupB
>  ResourceB1
>  ResourceB2
> 
> 
> Ordering constraint start GroupA then GroupB symetrical=false
> 
> Colocation constraint GroupB with GroupA
> 
>  ResourceB2 and ResourceA2 have stop time 20min+ and failover time in
> case of ResourceA1 fail is always 40min+
> 
> There is any possibility / hint how to force stop in the same time of
> both Resource Groups to improve the failover time?
> 
> BR,
> J. Gogu

The only thing I can think of is on-fail=fence
-- 
Ken Gaillot <kgaillot at redhat.com>