With the above example, we can see the 'bytes_processed' metric fed into our Datadog dashboard with a 'count' type. The problem we are seeing is that we cannot get the metric type override specified in the example config to work. We tried adding the type override in the annotations like so:
metrics = [
{ xxx_processed_bytes_total = {
name = "bytes_processed",
type = "gauge" } },
...
]
We've tried a number of different ways of structuring the data in the annotations but none have seemed to provide any results.
I can confirm that we tried running this with the 7.33.0 datadog-agent tag and attempted to add a config file manually on our datadog-agent pods to try and get this type override to work, but unfortunately that hasn't achieved any results.
Additional environment details (Operating System, Cloud provider, etc):
We're running on AWS EKS and have deployed a datadog daemonset and a datadog-clusterchecks deployment.
Describe the results you received:
Received a metric in Datadog, where the metric name was a concatenated string of all the words in the metric specification combined with a "_". From the example config posted above our metric came into the dashboard like so: "name_bytes_processed_type_gauge" with a type of 'count'.
Describe the results you expected:
We expected to see the bytes_processed metric with a overriden 'gauge' type.
Any clarity on this would be quite appreciated, thank you.
Hi,
Using the openmetrics integration via autodiscovery running in Kubernetes. The config is embedded into the annotations in this specified format.
The annotations for our deployment look like this:
With the above example, we can see the 'bytes_processed' metric fed into our Datadog dashboard with a 'count' type. The problem we are seeing is that we cannot get the metric type override specified in the example config to work. We tried adding the type override in the annotations like so:
We've tried a number of different ways of structuring the data in the annotations but none have seemed to provide any results.
I can confirm that we tried running this with the 7.33.0 datadog-agent tag and attempted to add a config file manually on our datadog-agent pods to try and get this type override to work, but unfortunately that hasn't achieved any results.
Additional environment details (Operating System, Cloud provider, etc): We're running on AWS EKS and have deployed a datadog daemonset and a datadog-clusterchecks deployment.
Describe the results you received: Received a metric in Datadog, where the metric name was a concatenated string of all the words in the metric specification combined with a "_". From the example config posted above our metric came into the dashboard like so: "name_bytes_processed_type_gauge" with a type of 'count'.
Describe the results you expected: We expected to see the bytes_processed metric with a overriden 'gauge' type.
Any clarity on this would be quite appreciated, thank you.