Open AaronClark-CRWD opened 1 year ago
I tested this by spinning up a pod with 1 container, then 4, then 5 containers. Could not reproduce it.
To be clear, this helm chart is simply a wrapper around https://github.com/fluent/fluent-bit so if there is an issue with the metadata it would most likely be upstream. But if there is a valid test that shows it breaking I could look into it further.
What we observed is if pod has more than 4 container then logscale is not populated with namespace_name and pod_name
e.g. namespace : wba-qa and pod_name : dep-wba-atm-preferences-single-7b5d4b9f79-t4g5j
I have verified this with 1,2,3,4 container in pod and logscale show log in console.
This behavior is in all environment.