Navin Agarwal
Navin Agarwal
Looks good. Can you please add tests for the scenario where in detached container, data stores are requested while its parent data store is initializing.
> @anthony-murphy Tony Murphy FTE Not quite sure. I tested with reverting my change around applying batch metadata regardless of the connected state of the ContainerRuntime, and that caused the...
> @agarwal-navin -- You say "No" does not introduce breaking change, but you're targeting `next`. Why is that? It's not breaking because I made a forward change some time back...
What happens to ops that were sent before forceReadonly() was called? I would think we want to send those ops since they were valid and then set to read-only. I...
Sure. But we should have an issue that tracks what happens to the pending ops. It will otherwise look like data loss from user's perspective.
Okay, that makes sense. Should we add a note to `forceReadonly` API that says pending ops may be lost and to avoid that check that container is not dirty before...
@alexvy86 Can you add tests to validate that the data store initialization does not happen unless handle.get() or request() is explicitly called on it. Basically, for summarizer clients, we don't...
Yes, sure. That sounds good.
@connorskees @markfields This is a risky change. We should ensure that are enough tests in place to not break anything. The scenario where a combination of new and old runtime...
@vladsud @pragya91 Can you guys please take a look at this PR? @jatgarg is not available right now and you both have the most context here.