hello,
I am trying to setup a NOMAD test environment using the (very useful!) -dev mode.
My test env has:
- a traefik instance which acts as a load balancer
- a number of HTTP containers
Traefik will map URLs to the correct container using tags. This works OK for my prod environment, but somehow I cannot replicate it in my test environment, because it seems that the Traefik container is not able to talk to the NOMAD endpoint (http://127.0.0.1:4646). Actually, I can see it from traefik logs:
time="2022-12-16T14:33:03Z" level=error msg="Provider connection error failed to load initial nomad services: Get \"http://127.0.0.1:4646/v1/services\": dial tcp 127.0.0.1:4646: connect: connection refused, retrying in 6.97842529s" providerName=nomad
time="2022-12-16T14:33:10Z" level=error msg="Provider connection error failed to load initial nomad services: Get \"http://127.0.0.1:4646/v1/services\": dial tcp 127.0.0.1:4646: connect: connection refused, retrying in 9.787400986s" providerName=nomad
time="2022-12-16T14:33:20Z" level=error msg="Provider connection error failed to load initial nomad services: Get \"http://127.0.0.1:4646/v1/services\": dial tcp 127.0.0.1:4646: connect: connection refused, retrying in 19.033258275s" providerName=nomad
Here is my Traefik job spec:
job "traefik" {
datacenters = ["dc1"]
type = "service"
group "traefik" {
count = 1
network {
mode = "host"
port "http"{
static = 80
}
port "admin"{
static = 8080
}
}
service {
name = "traefik-http"
provider = "nomad"
port = "http"
}
task "server" {
driver = "docker"
config {
image = "traefik:2.9.5"
ports = ["admin", "http"]
image_pull_timeout = "10m"
args = [
"--api.dashboard=true",
"--api.insecure=true",
"--entrypoints.http.address=:${NOMAD_PORT_http}",
"--entrypoints.traefik.address=:${NOMAD_PORT_admin}",
"--providers.nomad.exposedByDefault=false",
"--providers.nomad=true",
"--providers.nomad.endpoint.address=http://127.0.0.1:4646",
"--providers.nomad.endpoint.tls.insecureSkipVerify=true",
"--serversTransport.forwardingTimeouts.dialTimeout=600s",
"--serversTransport.forwardingTimeouts.idleConnTimeout=600s"
]
}
}
}
}
and this is the jobspec of the container which should be reachable from the load balancer:
job "shinyapp" {
datacenters = ["dc1"]
type = "service"
group "app" {
count = 1
network {
mode = "host"
port "shinyserver" {
to = 3838
}
}
service {
name = "app"
port = "shinyserver"
provider = "nomad"
tags = [
"traefik.enable=true",
"traefik.http.routers.http.rule=PathPrefix(`/shiny`)"
]
}
task "server" {
driver = "docker"
resources {
cpu = 2000
memory = 5000
}
config {
image = "rocker/shiny:4"
image_pull_timeout = "10m"
ports = ["shinyserver"]
auth_soft_fail = true
}
}
}
}
Any idea on how I can fix the problem? Thanks.
Matteo