[metrics-team] WIP: A high-level specification of processes behind Tor Metrics
karsten at torproject.org
Wed Nov 15 16:26:56 UTC 2017
In the spirit of release early, release often, I'm sharing a work in
It's a high-level specification of the processes behind Tor Metrics,
including CollecTor, Onionoo, ExoneraTor, metrics-web, and metrics-lib.
The purpose is to have a common understanding as a basis to:
- decide how to handle failures, errors, bugs, etc.
- decide what to log and on what severity,
- decide what to monitor and on what severity to report irregularities,
- write better unit tests to cover as many use cases as possible, and
- reflect design decisions and prepare evaluating frameworks for
The contribution so far is a list of all "processes" that make up Tor
Metrics, plus a detailed description of the processes behind ExoneraTor
with possible failure cases and how we handle them. Plus a few bugs that
I found while going through the code.
I guess my main questions are:
- Is this useful? It's sure a lot of work, but I feel that it can be
useful for us. Does it look useful for others, and if not, is there a
way how we can make it more useful?
- Are there more standardized, proven ways to formalize processes and
their failure cases? And I mean something that doesn't produce a lot of
overhead but instead saves us effort.
All the best,
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 528 bytes
Desc: OpenPGP digital signature
More information about the metrics-team