[ClusterLabs] Antw: Re: [EXT] Re: Two node cluster without fencing and no split brain?
johnt at panix.com
Mon Jul 26 09:53:06 EDT 2021
> On Fri, Jul 23, 2021 at 8:55 AM Ulrich Windl <
> Ulrich.Windl at rz.uni-regensburg.de> wrote:
>> >>> "john tillman" <johnt at panix.com> schrieb am 22.07.2021 um 16:48 in
>> <1175ffcec0033015e13d11d7821d5acb.squirrel at mail.panix.com>:
>> > There was a lot of discussion on this topic which might have
>> > this question so I will ask it again in case someone missed it.
>> > It comes from a post (see below) that we were pointed to here by
>> > Is there something like the described "ping tiebreaker" in the current
>> > world of pacemaker/corosync?
>> Maybe explain how it should work:
>> If the two nodes cannot rech each other, but each can reach the ping
>> which node has the quorum then?
> Guess both - which is what is played down as 'disadvantage' in the
> below ;-)
It is not perfect, I agree, but it may be better than nothing at all.
As for how it worked in my head:
I would have used the switch's IP address as the "ping" tie breaker; a
common connection point between the two nodes. My assumption is that if
there was network loss by Node A then it would lose quorum. In the mean
time Node B would still reach the switch, achieve quorum, and start/move
Thank you for your responses. Ken G. posted a suggestion of using
"corosync-qdevice" on a third node. I'll look into that but I may not be
able to install software on a third node. If anyone can think of another
path to follow, please let me know.
>> > Best Regards,
>> > âJohn
>> >> Interesting read. Thank you for providing it!
>> >> In this follow up post
>> > l
>> >> the author mentions the following:
>> >> Ping tiebreaker
>> >> Some HA systems provide a ping tiebreaker. To make this work, you
>> pick a
>> >> address outside the cluster to ping, and any partition that can ping
>> >> address has quorum. The obvious advantage is that it's very simple
>> >> up â doesn't require any additional servers or shared disk. The
>> >> disadvantage (and it's a big one) is that it's very possible for
>> >> partitions to think they have quorum. In the case of splitâsite
>> >> recovery) type clusters, it's going to happen fairly often. If you
>> >> use this method for a single site in conjunction with fencing, then
>> >> will likely work out quite well. It's a lot better than no
>> >> one that always says "you have quorum". Having said that, it's
>> >> significantly inferior to a
More information about the Users