[ClusterLabs] Antw: Re: 2-Node Cluster Pointless?

[Digimer]

On 18/04/17 07:55 PM, Dimitri Maziuk wrote:
> On 04/18/2017 06:15 PM, Alex Bruneau wrote:
> 
>> Therein lies the true value: what happens when something really
>> messes up? How much do you really want to drive to the DC at 3am?
> 
> Yes, exactly: what happens when something really messes up? Will you
> lose a cool million bucks or will a few people get a 404 until you get
> back to work on Monday?
> 
> The whole SCARY SPLIT BRANE! RUN!! RUN AWAY!!! spiel is really quite
> pointless without the answer to that.

The software must assume that, if you've gone to the effort of setting
up an HA cluster, that the answer is closer to "really bad day" then "meh".

Certainly, the people creating the software have to assume that a
split-brain is devastating. Same for people who teach others and people
who write documentation.

That someone *might* build unimportant clusters doesn't change anything,
really. One could ask "if the service isn't important, why go to the
hassle of building a cluster at all? It's just avoidable complication".

So, in closing, I still argue that if you need HA, you always need fencing.

-- 
Digimer
Papers and Projects: https://alteeve.com/w/
"I am, somehow, less interested in the weight and convolutions of
Einstein’s brain than in the near certainty that people of equal talent
have lived and died in cotton fields and sweatshops." - Stephen Jay Gould