Skip to content
This repository was archived by the owner on Mar 5, 2024. It is now read-only.

Idiamond metrics#131

Merged
pingles merged 9 commits intomasterfrom
idiamond-metrics
Aug 3, 2018
Merged

Idiamond metrics#131
pingles merged 9 commits intomasterfrom
idiamond-metrics

Conversation

@pingles
Copy link
Copy Markdown
Contributor

@pingles pingles commented Aug 2, 2018

This builds upon the work by @idiamond-stripe to improve handling of Prometheus and StatsD metrics.

It renames some of the metrics @idiamond-stripe introduced to bring them closer in line with the naming recommendations (_total only for counters etc.).

Currently gRPC Client metrics (those exported within the agent) don't report correctly- once that's fixed this can be merged.

idiamond-stripe and others added 7 commits July 21, 2018 13:40
- Replace go-metrics Prometheus metrics with the Prometheus go client
library
- Replace go-metrics statsd with a client that does not perform in
memory aggregation
- Add additional logging as to the duration of requests
* reorder interceptors to capture client metrics
* document client metrics that'll be exported for the agent
@codecov
Copy link
Copy Markdown

codecov bot commented Aug 2, 2018

Codecov Report

Merging #131 into master will increase coverage by 0.7%.
The diff coverage is 59.67%.

Impacted file tree graph

@@            Coverage Diff            @@
##           master     #131     +/-   ##
=========================================
+ Coverage   38.15%   38.86%   +0.7%     
=========================================
  Files          21       24      +3     
  Lines         815      831     +16     
=========================================
+ Hits          311      323     +12     
- Misses        475      480      +5     
+ Partials       29       28      -1
Impacted Files Coverage Δ
pkg/aws/metadata/server.go 5.55% <ø> (+0.1%) ⬆️
pkg/server/gateway.go 0% <0%> (ø) ⬆️
pkg/aws/sts/gateway.go 0% <0%> (ø) ⬆️
pkg/k8s/pod_cache.go 54.8% <0%> (ø) ⬆️
pkg/aws/metadata/log.go 20.83% <0%> (-1.9%) ⬇️
pkg/aws/metadata/handler_health.go 0% <0%> (ø) ⬆️
pkg/aws/metadata/handler_role_name.go 83.33% <100%> (ø) ⬆️
pkg/k8s/metrics.go 100% <100%> (ø)
pkg/aws/sts/metrics.go 100% <100%> (ø)
pkg/aws/metadata/metrics.go 100% <100%> (ø)
... and 7 more

@pingles pingles requested review from a team and tombooth August 2, 2018 15:52
@pingles
Copy link
Copy Markdown
Contributor Author

pingles commented Aug 2, 2018

If someone from @uswitch/cloud could take a look and see what you think- I think this should be the last thing needed before we push out an RC and start testing in some of our clusters.

docs/metrics.md Outdated
system, check the timing of each RPC call, and monitor the size of the
credentials cache. By default, Prometheus metrics are exported on
`localhost:9620` and StatsD metrics are sent to `127.0.0.1:8125`. StatsD
metrics are not aggregated and flushed every 100ms.
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

are not aggregated and flushed every 100ms. I think that should be "are not aggregated and are flushed every 100ms"

@pingles pingles merged commit ecf1dd2 into master Aug 3, 2018
@pingles pingles deleted the idiamond-metrics branch August 3, 2018 08:39
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants