lambi

side note: according to their website, it should be able to add packages even as non-root, but I have found not way of doing so... https://www.influxdata.com/blog/docker-run-telegraf-as-non-root/

If someone is running into issues with telegraf, docker etc., some notes on how I finally got it working (also as a future reference for me). Root cause: after telegraf version 1.20.2, telegraf does no longer run as root. It is therefore no longer able to access the docker status, and it is no longer possible to install packages via "post arguments" or by using the telegraf console as-is. Workaround would be staying on telegraf 1.20.2, which has some issues itself: a) possible security issue due to its root privileges b) will not receive any updates (new functionality and patches) c) docker monitoring seems not to be possible anymore due to it using an old version of the docker API (error message in log "[inputs.docker] Error in plugin: Error response from daemon: {"message":"client version 1.21 is too old. Minimum supported API version is 1.24, please upgrade your client to a newer version"}" Solution (working for me as of august 2025): install latest telegraf version telegraf:alpine give the telegraf user the rights to access the docker.sock by adding an extra parameter --user telegraf:$(stat -c '%g' /var/run/docker.sock) make sure your telegraf-conf does not use any additional packages (since those are not installed yet, the container will most likely not start). open the Unraid-CLI, and enter the telegraf container as root via docker exec -it --user root <telegraf-docker-name> sh now you can install any addtional packages via the CLI, e.g. apk add smartmontools apk add lm-sensors apk add lm-sensors-detect apk add nvme-cli apk add ipmitool now re-enable all inputs using additional packages (if needed). you should now have the latest version of telegraf running with additional packages and a telegraf user able to access the docker status.

Thanks, that worked!

yes I did. result: 2025-04-09 15:17:40: ERROR: SQLITE: Long running query Stmt: [PRAGMA journal_mode = DELETE] 2025-04-09 15:17:40: ERROR: Active query(0): PRAGMA journal_mode = DELETE which, according to https://forums.urbackup.org/t/error-when-use-urbackupsrv-remove-unknown/2026/2, happens because the server service is still running while executing the statement

I know, but that command requires the server to be stopped, otherwise the SQLite DB throws errors (see link in my above post). But I cannot figure out how to stop the server on the command line.

not sure if this has come up before... I am trying to run the "remove-unknown" command for urbackup from the CLI (since there are a couple of backup entries that are orphaned), but in order to to that, I need to stop the server from CLI (https://forums.urbackup.org/t/error-when-use-urbackupsrv-remove-unknown/2026/2) while keeping the docker running of course. I cannot however seem to be able to stop the server from the CLI, any suggestions (neither "systemctl stop" nor "service stop" work)? (maybe I am just completely missing something atm).