[Feature]: Running TCR on a slow environment

[mengdaming]

Contact Details

[email protected]

Feature Request

TCR should have the same behavior whichever the environment it's running in (slow machine, network with latency, etc.)

Feature Description

Current TCR implementation waits for 2s to re-arm the filesystem watching goroutine in order to cope with the changes done by TCR when using the commit-failure option. This works on machines running in a "normal" environment, but could fail in a "degraded" environment. We should look into a solution where we synchronize goroutines rather than relying on a timer.

Alternatives

Stay with current implementation

Additional Context

N/A

Code of Conduct

• I agree to follow this project's Code of Conduct