I have a Flask server running on GCE's container-optimized image. This server prints a log for every request. Normally, I can view the logs my application prints in GCP Stackdriver Logging console with no problem at all.


As of a few days ago, I could no longer see any logs from the VM in the Logging console. I have not touched the VM for the last 2 weeks: no code deployment, no restart, nothing of the sort. The server is clearly still running, since it continues to respond to requests as you would expect.

To be clear, I can see logs like "Compute Engine reset", but I can see none of the logs generated by my application's code.

Attempted solutions

First, I reset the VM instance using the GCE console. Did not work.

Second, I stopped the VM instance, then started it again. Did not work.

Third, I SSHd into my VM instance, and ran logger "Test logs!". I could still see nothing in the Logging Console.

Fourth, I made sure I had no filters in the logging console.

Fifth, I checked logs for other VM instances. No problems there. Only this one specific VM is facing this issue.

My hunch is that something happened to the logging agent on this VM. But I have no idea how to resolve that :(

Any help would be much appreciated!


The behavior you reported probably was linked to an internal Stackdriver event which happened few days ago.

Diagnosis of the incident: Customers may not see logs from later than Apr 19, 09:27 US/Pacific. Logs recently ingested into Stackdriver logging may not be available in the service.

From April 19th until April 20th some users were affected by an internal Stackdriver Logging not showing new logs issue. It was solved on 2019-04-20 02:49 PDT.

  • But we see an issue from Apr 16. – Ismail Khan Apr 24 at 5:14

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.