-
Type: Epic
-
Status: Resolved
-
Priority: Minor
-
Resolution: Fixed
-
Affects Version/s: None
-
Component/s: Monitoring
-
Tags:
-
Team(s):PLATFORM
-
Completion Level (0 to 5):5
Traditional monitoring using dashboard is hard to scale and limited to simple cases like detecting already known failure.
To answer the question of what happened on a complex system, where a single request is executed using external databases, services that handle retries policies and asynchronous processing, we need more than metrics we need distributed tracing.
This epic is about adding tracing and expose existing metrics to prometheus using opencensus.