Closed Jusshersmith closed 4 years ago
Merging #208 into master will increase coverage by
0.55%
. The diff coverage is100%
.
@@ Coverage Diff @@
## master #208 +/- ##
=========================================
+ Coverage 62.14% 62.7% +0.55%
=========================================
Files 53 53
Lines 4169 4169
=========================================
+ Hits 2591 2614 +23
+ Misses 1391 1367 -24
- Partials 187 188 +1
Impacted Files | Coverage Δ | |
---|---|---|
internal/auth/metrics.go | 94.73% <100%> (ø) |
:arrow_up: |
internal/proxy/metrics.go | 87.87% <100%> (+69.69%) |
:arrow_up: |
Problem
We send 'request' metrics in various places however they aren't all the same type of metric, which makes things confusing while actually utilising the metrics unless the actual code is inspected.
For example:
We send these 'request' metrics as time values: https://github.com/buzzfeed/sso/blob/f302bca23e5b8b841dfb411d3d43df344772e6a6/internal/auth/metrics.go#L69
We send these 'provider.request' metrics as incremented values: https://github.com/buzzfeed/sso/blob/521ff9c1529f7015d9cdc4d2a75832192535db18/internal/auth/providers/okta.go#L179
Solution
Slightly rename the metrics that are of a
time
type so we can easily distinguish between the value types.Note
This could be classed as a 'breaking' change for anyone using this metric for reporting/monitoring purposes, however not for the actual functionality of SSO itself.