Tests for scope and concurrency limit#418
Conversation
|
Part of restatedev/restate#4659 |
| } | ||
| // Briefly confirm the count stays at `limit`, defending against a slow rule activation | ||
| // where all 4 would have started before our first check. | ||
| delay(1.seconds) |
There was a problem hiding this comment.
TODO This is 100% not right.
@tillrohrmann how can i reliably assert all nodes have the same rule book propagated?
There was a problem hiding this comment.
I think this is something we need to expose in the PartitionProcessorStatus and then you can query it via the partition_state table.
There was a problem hiding this comment.
if this is soimething we can have for the schema registry in general, would be awesome btw, because i have some other tests that has the same issue: depending on the metadata propagation
|
Thanks a lot for adding these tests @slinkydeveloper 🦸. I am going to look into expose the current rule book version as part of the partition state. Then we can use this to wait until all partitions are using the latest rule book. |
slinkydeveloper
force-pushed
the
scope-tests
branch
from
061cdcc to
73410f5
Compare
|
@slinkydeveloper what's missing to move this PR forward? It would be really great to increase our test coverage of scoped invocations with concurrency limits :-) |
slinkydeveloper
force-pushed
the
scope-tests
branch
from
73410f5 to
1714320
Compare
slinkydeveloper
force-pushed
the
scope-tests
branch
from
1714320 to
fafe445
Compare
2 participants
No description provided.