Problem: Runner does not handle error if it cannot write its config_file

[MannerMan]

Issue

Hi!

Using remote runner feature, it seems as if the runner cannot write its config_file, it will not catch this error and silently continue. I encountered this problem when mounting a volume though docker with incorrect permissions. This has the effect of each restart of the runner becomes a new registration.

Possible fix:
develop...MannerMan:semaphore:fix-runner-config-write-error-handling

Impact

Service (scheduled tasks, alerts)

Installation method

Docker

Database

Postgres

Browser

No response

Semaphore Version

2.10.7

Ansible Version

No response

Logs & errors

No logs, the error is silently ignored

Manual installation - system information

No response

Configuration

{
  "runner": {
    "registration_token": "<redacted>",
    "config_file": "/var/lib/semaphore/semaphore-runner.json",
    "api_url": "https://semaphore.<redacted>/api",
    "max_parallel_tasks": 4
  },
  "port": "3000",
  "interface": "",
  "tmp_path": "/var/lib/semaphore/tmp",
  "concurrency_mode": ""

Additional information

No response

[m-soltani]

@MannerMan
It seems that I am facing the same issue. Have you find any workarounds yet?