How to force machine restart if checks are failing?

1

All checks are failing on this machine (despite the processes seemingly being up), but the machine is not restarted.

Screenshot 2025-03-08 at 3.03.38 PM
Screenshot 2025-03-08 at 3.03.38 PM2474×730 95.7 KB

I want the machine to restart if a check is failing for more than X attempts.

2

Can we see your TOML config?

3

Here you go:

primary_region='iad'
kill_signal='SIGINT'
kill_timeout=60
swap_size_mb=512

[deploy]
  strategy = "bluegreen"

[checks]
  [checks.react-router]
    # We need a relatively high grace period due to the time it takes to import react-router build.
    grace_period="60s"
    interval="2s"
    method="GET"
    path="/ping"
    port=3000
    timeout="1s"
    type="http"
  [checks.ready]
    grace_period="60s"
    interval="2s"
    method="GET"
    path="/checks/ready"
    port=3000
    timeout="1s"
    type="http"
  [checks.redis]
    grace_period="60s"
    interval="5s"
    method="get"
    path="/checks/redis"
    port=3000
    timeout="1s"
    type="http"
  [checks.postgres]
    grace_period="60s"
    interval="5s"
    method="get"
    path="/checks/postgres"
    port=3000
    timeout="1s"
    type="http"

[http_service]
  internal_port=3000
  force_https=true
  auto_stop_machines='off'
  auto_start_machines=true
  min_machines_running=2
  processes=['app']
  [http_service.concurrency]
    type="requests"
    soft_limit=100
  [http_service.http_options.response]
    pristine=true

[[vm]]
  cpus=2
  cpu_kind = "performance"
  memory="4gb"

[[restart]]
  policy='always'
  processes=['app']
1 Like
4

Hmm, the staff advice on this old post is that it is not possible as a config option:

I suggested here a little while ago how it can be done if you don’t mind doing it yourself:

5

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.