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
I found a way to workaround, but I believe it requires more info.
When grafana container starts, I observe this
Error checking server process execution privilege. error: could not get current OS user to detect process privileges
Looking around, I found a reference grafana/grafana#63599 which suggests some stuff, and then in the manual I found a note on bind mounts which suggest to use user: '0'
Adding this seems to fix the problem. Unclear, though, why this is needed only now.
My setup is a MacOS Tahoe with Docker Desktop 4.59.1 (217750)
The need to run the Grafana container as root (user: '0') on macOS with bind mounts started after changes in the DevLake Grafana Dockerfile for beta9. These changes, made for OpenShift compatibility, switched some operations to root and altered group ownership and permissions on key directories before switching back to a non-root user. This setup works well for OpenShift, but on macOS with Docker Desktop, bind mounts can cause permission issues if the container user (now UID 101) doesn't exist on the host, leading to errors like "Error checking server process execution privilege" and "attempt to write a readonly database" source.
Setting user: '0' in your Docker configuration bypasses these permission problems by running Grafana as root, which matches the workaround recommended in upstream Grafana documentation for macOS bind mounts. There aren't any migration notes or troubleshooting docs in DevLake about this, but the timing and nature of the Dockerfile changes explain why this is needed now.
If you or others encounter this, using user: '0' is the recommended fix for macOS with bind mounts. For more details, see the upstream Grafana docs on bind mounts and the related issue. If this becomes a common problem, it may be worth sharing your findings with the community.
This issue has been automatically marked as stale because it has been inactive for 60 days. It will be closed in next 7 days if no further activity occurs.
Activity
LunaticMuch commentedon Feb 3, 2026
I found a way to workaround, but I believe it requires more info.
When grafana container starts, I observe this
Looking around, I found a reference grafana/grafana#63599 which suggests some stuff, and then in the manual I found a note on bind mounts which suggest to use
user: '0'Adding this seems to fix the problem. Unclear, though, why this is needed only now.
My setup is a MacOS Tahoe with Docker Desktop
4.59.1 (217750)dosubot commentedon Feb 3, 2026
The need to run the Grafana container as root (
user: '0') on macOS with bind mounts started after changes in the DevLake Grafana Dockerfile for beta9. These changes, made for OpenShift compatibility, switched some operations to root and altered group ownership and permissions on key directories before switching back to a non-root user. This setup works well for OpenShift, but on macOS with Docker Desktop, bind mounts can cause permission issues if the container user (now UID 101) doesn't exist on the host, leading to errors like "Error checking server process execution privilege" and "attempt to write a readonly database" source.Setting
user: '0'in your Docker configuration bypasses these permission problems by running Grafana as root, which matches the workaround recommended in upstream Grafana documentation for macOS bind mounts. There aren't any migration notes or troubleshooting docs in DevLake about this, but the timing and nature of the Dockerfile changes explain why this is needed now.If you or others encounter this, using
user: '0'is the recommended fix for macOS with bind mounts. For more details, see the upstream Grafana docs on bind mounts and the related issue. If this becomes a common problem, it may be worth sharing your findings with the community.To reply, just mention @dosu.
Docs are dead. Just use Dosu.
klesh commentedon Feb 9, 2026
Thanks for reporting the issue!
DoSubot summarized it perfectly—let’s jot it down somewhere so the next person doesn’t stumble into the same hole.
@LunaticMuch, how did you do the upgrade? Any thoughts on the best spot to park the warning?
github-actions commentedon Apr 11, 2026
This issue has been automatically marked as stale because it has been inactive for 60 days. It will be closed in next 7 days if no further activity occurs.
github-actions commentedon Apr 18, 2026
This issue has been closed because it has been inactive for a long time. You can reopen it if you encounter the similar problem in the future.