[ClusterLabs] Antw: Re: Issue with DB2 HADR cluster

[Andrei Borzenkov]

On Wed, Apr 3, 2019 at 10:26 AM Valentin Vidic <Valentin.Vidic at carnet.hr> wrote:
>
> On Wed, Apr 03, 2019 at 09:13:58AM +0200, Ulrich Windl wrote:
> > I'm surprised: Once sbd writes the fence command, it usually takes
> > less than 3 seconds until the victim is dead. If you power off a
> > server, the PDU still may have one or two seconds "power reserve", so
> > the host may not be down immediately. Besides of that power-cycles are
> > additional stress for the hardware...
> >
> > So maybe you want to explain why and how much faster IPMI and PDU fencing are.
>
> SBD is slow for me too. Since it doesn't have a way to confirm the kill
> it needs to wait for various timeouts and these can be quite high. For
> example the IBM storage timeouts require this setup:
>
>   Timeout (watchdog) : 130

I assume this is path failover time? As I doubt storage latency can be
that high?

I wonder, does IBM have official guidelines for integrating SBD with
their storage? Otherwise where this requirement comes from?

>   Timeout (msgwait)  : 270
>
> On the same cluster IPMI fence executes in a second or two, but requires
> network connectivity.