You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Despite seeing the logs with level=info, they are being classified as ERROR in GCP Log Explorer. This results in incorrect visibility and metrics in our cluster.
Attached screenshots demonstrating the issue:
Expected Behavior
Logs with a level of INFO from Cronitor should be displayed as INFO severity in GCP Logging, ensuring accurate metric tracking.
My assumption is that all logs are being sent to standard error, which causes them to be displayed as such in the logs.
We would appreciate your assistance in resolving this issue.
The text was updated successfully, but these errors were encountered:
matimercado
changed the title
Logs in GCP Logs Explorer Displayed as Errors Despite Ievel=info in cronitor
Logs in GCP Logs Explorer Displayed as Errors Despite level=info in cronitor
Jan 13, 2025
Thanks for reaching out about this. It looks like this is an issue with the logging library that we chose for this project. I'm surprised this is the first time it's come up!
It looks like this library is also no longer releasing updates, so fixing it looks like it's a matter of swapping out logging libraries. If you'd like to open a PR that does that we'd be happy for assistance testing out a replacement! I'll also bring up with the team to figure out when we could prioritize a release ourselves.
Hi folks!
We are currently running the latest version of the Cronitor Helm chart (v0.5.0) in our GKE cluster. The deployment was performed as follows:
Problem
Despite seeing the logs with
level=info
, they are being classified as ERROR in GCP Log Explorer. This results in incorrect visibility and metrics in our cluster.Attached screenshots demonstrating the issue:
Expected Behavior
Logs with a level of INFO from Cronitor should be displayed as INFO severity in GCP Logging, ensuring accurate metric tracking.
My assumption is that all logs are being sent to standard error, which causes them to be displayed as such in the logs.
We would appreciate your assistance in resolving this issue.
The text was updated successfully, but these errors were encountered: