Simple Nomad Job without Docker or other dependencies

Nomad
1

I am in a restricted environment where I cannot install Docker or Node.JS . Anyone have a simple nomad JOB thatI can test with ? I have created my first cluster and would love to try a job .

I did create a job using this link Running bash scripts as a task in nomad using raw_exec driver but I am getting the following error

Task Group “cache” (failed to place 1 allocation):

  • Constraint “missing drivers” : 4 nodes excluded by filter .

I see the job in the UI but it shows pending and on my command line it shows “deployment in progress” perhaps a simple job is not my problem but would love to try something different .

Any help is appreciated

2

Hi @sammy676776 !

For the most simple job imaginable, I typically use this - which just calls sleep using the raw_exec driver

job "example" {

  group "group1" {
    task "sleep" {
      driver = "raw_exec"
      config {
        command = "sleep"
        args    = ["infinity"]
      }
      resources {
        cpu    = 10
        memory = 10
      }
    }
  }
}

For a simple HTTP server, I like to use this job which just uses python3’s built-in http.server module

job "example2" {
  group "group" {
    network {
      mode = "host"
      port "http" { static = 8888 }
    }

    task "pyhttp" {
      driver = "raw_exec"
      config {
        command = "python3"
        args    = ["-m", "http.server", "${NOMAD_PORT_http}", "--directory", "/tmp"]
      }

      service {
        name     = "py"
        port     = "http"
        provider = "nomad"
        check {
          type     = "http"
          path     = "/"
          interval = "3s"
          timeout  = "1s"
        }
      }
    }
  }
}

Also note that raw_exec driver is not available by default on normal clusters -you have to enable it manually as a plugin option

If you are on Linux, it’s probably better to just use the exec driver instead, which leverages chroot, and cgroups for resource isolation.

3

Thank you . Will try and post the result .

4

Perhaps there is some other issue going on in my new cluster that I built as even this simple job failed .

in my Server.hcl I have

> plugin "raw_exec" {
>     config {
>       enabled    = true
>       no_cgroups = true
>     }
>   }

client.hcl

options {
      "driver.allowlist" = "raw_exec"
    }
}
plugin "raw_exec" {
    config {
      enabled    = true
      no_cgroups = true
    }
  }

When I submit the job it appears in the UI but shows failed . It also shows the alloc and clients . How can I be sure that the server has communicated to clients properly ? Is there anything I can check ? J

I ran “nomad server members -detailed” and I can see all server details including the clients . I can go to UI and see both servers and clients . When I run the job does it deploy to the clients and if it was successful will it say something in client or server log ? If so where ?

thanks in advance

5

Meant to say I also ran “nomad node status -stats” and it shows all clients . I am running your “sleep” raw_exec job and here is the status

nomad job status test
ID            = test
Name          = test
Submit Date   = 2023-02-24T03:37:59-05:00
Type          = service
Priority      = 50
Datacenters   = <removed>
Namespace     = default
Status        = pending
Periodic      = false
Parameterized = false

Summary
Task Group  Queued  Starting  Running  Failed  Complete  Lost  Unknown
group1      0       0         0        6       0         0     0

Future Rescheduling Attempts
Task Group  Eval ID   Eval Time
group1      dd05e2f2  14m5s from now

Latest Deployment
ID          = 648402a6
Status      = running
Description = Deployment is running

Deployed
Task Group  Desired  Placed  Healthy  Unhealthy  Progress Deadline
group1      1        2       0        2          2023-02-24T03:47:59-05:00

Allocations
ID        Node ID   Task Group  Version  Desired  Status  Created     Modified
66e60599  24d34bf1  group1      2        run      failed  1m55s ago   1m54s ago
8a0272fe  24d34bf1  group1      2        stop     failed  9m56s ago   1m55s ago
e861a442  15ceea08  group1      0        stop     failed  15m7s ago   9m56s ago
22defb0d  24d34bf1  group1      0        stop     failed  17m7s ago   15m7s ago
a3d18006  098cf534  group1      0        stop     failed  18m7s ago   17m7s ago
1d3fdcf0  dd964482  group1      0        stop     failed  18m37s ago  18m7s ago

Based on the above output I also ran

nomad alloc status --short 66e60599
ID       = 66e60599-f557-8d35-f021-266e7e7013c5
Name     = test.group1[0]
Created  = 4m10s ago
Modified = 4m9s ago

Tasks
Name   State  Last Event      Time                       Lifecycle
sleep  dead   Not Restarting  2023-02-24T03:45:59-05:00  main