The problem is this:
$ tutor local logs lms -f
docker-compose -f /home/tutor/.local/share/tutor/env/local/docker-compose.yml -f /home/tutor/.local/share/tutor/env/local/docker-compose.prod.yml --project-name tutor_local logs --follow lms
/usr/lib/python3/dist-packages/requests/__init__.py:91: RequestsDependencyWarning: urllib3 (1.25.10) or chardet (3.0.4) doesn't match a supported version!
RequestsDependencyWarning)
Attaching to tutor_local_lms_1
lms_1 | [2020-12-04 15:33:47 +0000] [7] [INFO] Starting gunicorn 20.0.4
lms_1 | [2020-12-04 15:33:47 +0000] [7] [INFO] Listening at: http://0.0.0.0:8000 (7)
lms_1 | [2020-12-04 15:33:47 +0000] [7] [INFO] Using worker: sync
lms_1 | [2020-12-04 15:33:47 +0000] [10] [INFO] Booting worker with pid: 10
lms_1 | [2020-12-04 15:33:47 +0000] [12] [INFO] Booting worker with pid: 12
lms_1 | [2020-12-04 15:34:17 +0000] [7] [CRITICAL] WORKER TIMEOUT (pid:10)
lms_1 | [2020-12-04 15:34:17 +0000] [7] [CRITICAL] WORKER TIMEOUT (pid:12)
lms_1 | [2020-12-04 15:34:17 +0000] [12] [INFO] Worker exiting (pid: 12)
lms_1 | [2020-12-04 15:34:17 +0000] [10] [INFO] Worker exiting (pid: 10)
lms_1 | [2020-12-04 15:34:18 +0000] [24] [INFO] Booting worker with pid: 24
lms_1 | [2020-12-04 15:34:18 +0000] [25] [INFO] Booting worker with pid: 25
lms_1 | [2020-12-04 15:34:48 +0000] [7] [CRITICAL] WORKER TIMEOUT (pid:24)
lms_1 | [2020-12-04 15:34:48 +0000] [7] [CRITICAL] WORKER TIMEOUT (pid:25)
lms_1 | [2020-12-04 15:34:48 +0000] [25] [INFO] Worker exiting (pid: 25)
lms_1 | [2020-12-04 15:34:48 +0000] [24] [INFO] Worker exiting (pid: 24)
lms_1 | [2020-12-04 15:34:49 +0000] [50] [INFO] Booting worker with pid: 50
lms_1 | [2020-12-04 15:34:50 +0000] [51] [INFO] Booting worker with pid: 51
lms_1 | [2020-12-04 15:35:20 +0000] [7] [CRITICAL] WORKER TIMEOUT (pid:50)
lms_1 | [2020-12-04 15:35:20 +0000] [7] [CRITICAL] WORKER TIMEOUT (pid:51)
lms_1 | [2020-12-04 15:35:20 +0000] [50] [INFO] Worker exiting (pid: 50)
lms_1 | [2020-12-04 15:35:20 +0000] [51] [INFO] Worker exiting (pid: 51)
lms_1 | [2020-12-04 15:35:21 +0000] [76] [INFO] Booting worker with pid: 76
lms_1 | [2020-12-04 15:35:21 +0000] [78] [INFO] Booting worker with pid: 78
The gunicorn workers take a bit too long to start and then they hit a timeout. They get restarted, take a bit too long to start again, etc. etc. etc.
The solution (in my case) is to increase the timeout a bit. I have set it to 120 by adding -t 120 to the gunicorn command from the openedx Dockerfile. Then the gunicorn workers start without problems.