Closed ThEcTecan closed 3 weeks ago
When the upgrade fails, it will log a stack trace, and it will not automatically shut down.
Are you running on an orchestrator such as Kubernetes, that could be killing the pod if it doesn't become healthy in a certain amount of time?
How much time passes between Computing severities for vulnerabilities where severity is currently NULL
being logged and the container being shut down?
Are you running on an orchestrator such as Kubernetes, that could be killing the pod if it doesn't become healthy in a certain amount of time?
Thanks, that is exactly what happened.
I am indeed using docker compose to start the container in an Azure App Service. According to Azure App Service on Linux FAQ there is a default timeout of 230s for the container to start up.
After increasing the timeout, I could see that Computing severities for vulnerabilities where severity is currently NULL
took ~12 minutes and the upgrade ran through successful.
Hmmm, 12min is a lot for this operation. It already utilizes SQL statement batching so it doesn't perform a lot of requests to the database.
Do you have high latency between API server and database?
Current Behavior
Upgrading Dependency Track apiserver from version 4.10.1 to 4.11.x fails with no logged exception. You can see that the v4110Updater has started it's execution, but the logs stop with
Afterwards the docker container is shutdown.
Here are the full logs of the upgrade process:
``
Steps to Reproduce
Expected Behavior
The upgrade succeeds or there is an error message logged, indicating why the upgrade fails.
Dependency-Track Version
4.11.3
Dependency-Track Distribution
Container Image
Database Server
Microsoft SQL Server
Database Server Version
Azure SQL Database (latest)
Browser
Microsoft Edge
Checklist