We created a parameterized dashboard in Datadog and noticed several tag discrepancies,
such as env
vs. environment
or prod
vs. production
.
AWS does not provide any sane way to query for tags across different services,
as this Twitter thread laments,
so I looked for ways to perform a sort of “reverse lookup” from Datadog.
The following technique turned out pretty useful:
First, go to Datadog > Metrics > Summary and add a tag you’re looking for (e.g. environment:prod
).
The list of metrics is filtered down to incoming metrics from resources with that tag.
Make note of at least one metric each service; for demo purposes I’ll use aws.s3.bucket_size_bytes
.
Leave Metrics Summary and navigate to Metrics Explorer. Fill in the following:
Graph:
aws.s3.bucket_size_bytes # the metric from Metrics Summary
Over:
environment:prod # the tag you're searching
One graph per:
bucketname # any human-friendly attribute
Tada! Datadog renders graphs to the right with each bucket’s name in the title.