[BUG] ApiGateway metrics being skipped "metric":"IntegrationLatency","msg":"skipping metric unmatched by associator"

[overlordchin]

Is there an existing issue for this?

• I have searched the existing issues

YACE version

0.61.2

Config file

apiVersion: v1alpha1 discovery: jobs: - type: AWS/ApiGateway regions: - us-east-1 period: 300 length: 300 metrics: - name: 4XXError statistics: [Sum] - name: 5XXError statistics: [Sum] - name: IntegrationLatency statistics: [Maximum, Average, p99] - name: Latency statistics: [Maximum, Average, p99]

Current Behavior

ApiGateway v1 metrics appear to work correctly and show up in the metrics page and prometheus.
V2 have log messages as follows:
{"account":"redacted","arn":"","caller":"discovery.go:248","dimensions":"ApiName=my-api,Resource=/myenResource,Stage=mystage,Method=POST","job_type":"AWS/ApiGateway","level":"debug","metric":"IntegrationLatency","msg":"skipping metric unmatched by associator","region":"us-east-1","ts":"2024-07-16T21:08:36.186928023Z","version":"0.57.1"}

Expected Behavior

metrics are not skipped and show up in prometheus and are searchable by dimensions that show up on v1 resources

Steps To Reproduce

1. ensure metrics are visible with dimensions in cloudwatch for given resource
2. add above configuration to yace
3. restart yace to ensure config is loaded
4. read log messages from yace and observe no errors.
5. add debug flag to log and restart yace to observe skipping messages

Anything else?

We have tried both with and without the enable max dimension association and aws-sdk-v2 features, neither seems to have any impact. It is unclear why these are being skipped and adding required dimension names causes it to find no metrics at all and even the v1 metrics disappear

Additionally we have tried adding dimension requirements which causes ALL apigateway metrics to be skipped