Registering a Server

If you have access to a virtual machine (VM) or bare metal server, you can register it as a Tapis system. To do so, you must first describe the system in a JSON object.

The following example can be modified for your VM.

system_def = {
  "id": "<userid>.<vm_hostname>",
  "description": "Tapis v3 execution system",
  "systemType": "LINUX",
  "host": "<network_address>",
  "effectiveUserId": "<system_account>",
  
  "defaultAuthnMethod": "PKI_KEYS",
  "rootDir": "/home/<system_account>",
  "canExec": True,
  "jobRuntimes": [ { "runtimeType": "DOCKER" }, { "runtimeType": "SINGULARITY" } ],
  "jobWorkingDir": "workdir",
}

Be sure to review the following and make any required updates:

• <userid>.<vm_hostname> - A unique id for this system. As indicated, we recommend using a combination of your username and the hostname of the computer.
• <network_address> - The network address of the machine (e.g., an IP address or domain)
• <system_account> - The POSIX account on the system that Tapis should use when accessing the system. Note that when registering individual servers, there is often only one POSIX account that is used, however, it is possible to support a multiple users when the usernames on the system match those authenticating with Tapis. In that case, settingeffectiveUserId to the value "${apiUserId} will instruct Tapis to use the API user’s identity (i.e., their username) when accessing the system.
• /home/<system_account> - The root directory; Tapis will prepend this path to all paths issued against this system.
• If you want to use a password (instead of keys) to delegate authentication to Tapis, replace "PKI_KEYS" with "PASSWORD" for the defaultAuthnMethod.

Note that although it is possible, we have not provided any login credentials in the system definition. For security reasons, it is recommended that login credentials be updated using a separate API call as discussed below.

If your server uses SLURM to schedule jobs, you can add the following attributes to the system definition:

{
  "jobIsBatch": True,
  "batchScheduler": "SLURM",
  "batchSchedulerProfile": "tacc",
  "batchLogicalQueues": [
    {
      "name": "tapisNormal",
      "hpcQueueName": "debug",
      "maxJobs": 50,
      "maxJobsPerUser": 10,
      "minNodeCount": 1,
      "maxNodeCount": 16,
      "minCoresPerNode": 1,
      "maxCoresPerNode": 68,
      "minMemoryMB": 1,
      "maxMemoryMB": 16384,
      "minMinutes": 1,
      "maxMinutes": 60
    }
  ],
  "batchDefaultLogicalQueue": "tapisNormal",
}

Additional Resources

• Additional details about the Systems endpoints - API specification