Opendronemap on Fedora 34

Project Discussion
1

Hi,

I try to run Opendronemap project on my Fedora 34. It’s based on Docker container, and i know Fedora uses Podman instead. I have to replace all docker’s occurences in the Docker file. Every images are correctly downloaded (i see it in Cockpit’s podman app), pod and containers are running but the last one called “webapp” doesn’t want to run. I have these logs when i try to run :

Error: unable to start container
8a05556cb80099473d81bc0bef5954979348e77ab78004ccdad07f98c81519d6: 
executable file /bin/bash -c 'chmod +x /webodm/*.sh && /bin/bash -c "/webodm/wait-for-postgres.sh db /webodm/wait-for-it.sh -t 0 broker:6379 -- /webodm/start.sh"'
not found in $PATH: No such file or directory: OCI runtime attempted to invoke a command that was not found

Can anyone explain me why i can’t run ?

2

Maybe try to install podman-docker and podman-compose. And then simply run the Docker commands. Or directly use podman-compose.

3

That error is coming from the container image itself. It’s saying one of those commands is broken inside the container. It’s not a Podman / Docker issue.

4

Is it possible to enter in the offending container even if it isn’t started to edit files into ?

@ Kees de Jong : all necesaries packages are already installated in my desktop (podman, podman-compose etc).

5

Remove the entry point lines in the dockerfile and rebuild it. Then you can interactively edit things inside the container and run the command.

6

You probably want to see this.

7

OK i’ve found solution :

I have to edit Docker file and add the line “command” and edit “entrypoint” like this is said here :

Deleting affected containers and rebuild them solve problems.

Thanks for your help.

8

On the broken container do:

podman run -t [container name] < /bin/bash

This will put you in a shell in that container. From there you can troubleshoot why the application isn’t starting and then get it started.

However, this won’t fix it. Anytime you restart the container you’d have to do this. So once you’ve figured out what is wrong, you want to update and rebuild the container image, then redeploy that container.