GRPC request got a "DeadlineExceeded desc = context deadline exceeded" for a first request after idle time

Questions / Help
1

Good day everyone!
I’ve run a GRPC service on fly.io with the next config:

# fly.toml app configuration file generated for ****** on 2024-08-04T13:56:17+03:00
#
# See https://fly.io/docs/reference/configuration/ for information about how to use this file.
#

app = 'my_perfect_app_name'
primary_region = 'waw'

[[services]]
  protocol = 'tcp'
  internal_port = 50051
  auto_stop_machines = "off"
  auto_start_machines = true
  min_machines_running = 1
  processes = ['app']

  [[services.ports]]
    port = 443
    handlers = ['tls']

    [services.ports.tls_options]
      alpn = ["h2"]

    [services.concurrency]
      type = "connections"
      hard_limit = 25
      soft_limit = 20

  [[services.tcp_checks]]
      grace_period = "60s"
      interval = "15s"
      restart_limit = 0
      timeout = "2s"


[checks]
  [checks.transcription_service_tcp_check]
    grace_period = "60s"
    interval = "15s"
    port = 50051
    timeout = "10s"
    type = "tcp"

[deploy]
  wait_timeout = "2m"

[[vm]]
  memory = '1gb'
  cpu_kind = 'shared'
  cpus = 2

As I got it, with auto_stop_machines = "off" and min_machines_running = 1 it must be running all time (I see that in the dashboard / service / machines).

After idle time for a first request I get the next (via ghz):

Status code distribution:
  [OK]                 199 responses   
  [DeadlineExceeded]   1 responses     

Error distribution:
  [1]   rpc error: code = DeadlineExceeded desc = context deadline exceeded

And for a second request:

Status code distribution:
  [OK]   200 responses

What is wrong with my config? Why after an idle time have I the DeadlineExceeded desc = context deadline exceeded?

2

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