Hi Aleksandr,
Thanks for driving this. Someone should put a stop to the shutdown behavior
nonsense :)
But let me complicate things a bit for you.
I see multiple dimensions that are in play when node shutdown happens:
- Running operations (Compute, SQL, Cache): interrupt or wait for completion
> which is closes to non distributed fashion you mentioned.
If you use Ignite Services I'd say it's already distributed.
How people often do automation: they pick a single server and setup a system
cron job there. The server dies - their cron doesn't run, which isn't how it
should be.
Now, if
Hi Vladimir,
I think this is potentially an issue but I don't think this is about PDS at all.
The description is a bit vague, I have to say. AFAIU what you see is that when
the caches are persistent the streamer writes data faster than the nodes
(especially, backup nodes) process the writes.
Th
The general rule I always have in mind for breaking changes - there should be a
way to fallback to the old behavior.
Even if the old behavior is unreasonable, and people shouldn't rely on it, you
should let them acknowledge they are in the wrong - but continue to operate as
they used to.
And in
