disclaimer
the verbiage can change, doesn't need to be zoo snapshot.
Issue
We need better functionality to be able to debug what is going wrong across the stack with different start and stop points.
Different users and internal users will have different start and end points when debugging issues.
As someone debugging regions, I cannot start with a main.kcl and debug waveback.
I want a pipeline that can snapshot each stage for E2E testing as well as internal tesing to easily debug the stack.
Brainstorming
A new set of CLI commands that take a start and stop point and will produces a bunch of data for you to write to disk that can be used as inputs to any stage.
Scenarios
$ zoo snapshot kcl api ./main.kcl
This would output
./main.kcl.snapshot
./ezpz.snapshot
./api.snapshot
If you then ran
$ zoo snapshot ezpz api ./ezpz.snapshot
This would output
./ezpz.snapshot
./api.snapshot
and both of the ezpz and api files will be exactly the same!
Contents
We can talk to each team to figure out what we want to debug snapshot to be able to serialize/deseralize without the loss of information and it needs to be useful enough for others to start their debug process from.
disclaimer
the verbiage can change, doesn't need to be
zoo snapshot.Issue
We need better functionality to be able to debug what is going wrong across the stack with different start and stop points.
Different users and internal users will have different start and end points when debugging issues.
As someone debugging regions, I cannot start with a
main.kcland debugwaveback.I want a pipeline that can snapshot each stage for E2E testing as well as internal tesing to easily debug the stack.
Brainstorming
A new set of CLI commands that take a start and stop point and will produces a bunch of data for you to write to disk that can be used as inputs to any stage.
Scenarios
This would output
If you then ran
This would output
and both of the
ezpzandapifiles will be exactly the same!Contents
We can talk to each team to figure out what we want to debug snapshot to be able to serialize/deseralize without the loss of information and it needs to be useful enough for others to start their debug process from.