Tutor for 2u/OCM: FAQs and Debugging

Issue: tutor dev quickstart fails with

• This sometimes just goes away if you wait a minute and try again

Issue: tutor dev stop doesn’t actually stop anything

• Check your container names with docker ps. They should be of the form tutor_nightly_dev_myservice.

  • If they do not have the nightly suffix, you may be accidentally running the latest tutor release instead of tutor-nightly. You can check this by running tutor config printroot. If you are running the nightly version, the directory name will be /some/thing/tutor-nightly. Otherwise it will just be /some/thing/tutor

  • If they say local instead of dev, run tutor local stop

tutor images build --no-cache all seems to still be using cached layers

• try using tutor dev dc build --no-cache instead

Possible code mounting issue?

• Error response from daemon: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: exec: "./manage.py": stat ./manage.py: no such file or directory: unknown

• Came up any time a command that used docker-compose ‘run’ under the hood. (eg tutor dev run lms bash)

• Resolved by just doing a tutor stop tutor start with no mount

⚠️ The current environment stored at /Users/username/Library/Application Support/tutor/env is not up-to-date: ...

• If you had been running tutor-nightly and installed a non-nightly release brach of a plugin it may have updated your tutor-nightly install to the latest named release.

• you can check your tutor and plugin versions with pip list

• re-installing the nightly branch of tutor should resolve this

ERROR: for cms Cannot start service cms: Ports are not available: listen tcp 0.0.0.0:8001: bind: address already in use

• This often pops up if you have been running minikube or otherwise using kubectl

  • Sometimes minikube delete will fix it

• lsof -i:8001 will show you what’s running there. you can use ps -ef | grep <process> to find the process id and kill it