Developer documentation

Under Development

This tool is under active development. The documentation is not complete yet. If you have any questions, please contact us via GitHub Issues

Architectural drivers

• Slurm WLM has a reduced set of commands, but the number of optional arguments required by each one is significantly large
• The OpenAPI specification of slurmrest changes significantly on each new version, as well as the supported arguments by each wlm version.
• A wlm can provide support to multiple OpenAPI specification versions
• Each OpenApi specifaction may significantly differ from each other

Client:

https://github.com/SLURM-CLI-API-Proxy/SLURM-CLI-API-Proxy-client/blob/main/slurm_api_client/docs/SlurmApi.md

References

• Schmed's Slurm REST API documentation
• Schmed's Slurm command line documentation
• Slurm version vs supported API versions

graph TD;


    Users -->|SSH| sinfo[sinfo];
    Users -->|SSH| sbatch[sbatch];
    Users -->|SSH| squeue[squeue];
    Users -->|SSH| scontrol[scontrol];

    subgraph "Head Node"
        slurmctld --> slurmdbd[slurmdbd];
        sinfo --> slurmctld;
        sbatch --> slurmctld;
        squeue --> slurmctld;
        scontrol --> slurmctld;
    end

    subgraph "Compute Node"
        slurmd1[slurmd] 
    end

    subgraph "Compute Node"
        slurmd2[slurmd]
    end

    slurmctld --> slurmd1;
    slurmctld --> slurmd2;

    slurmdbd --> Database["Slurm Accounting Database"];

graph TD;


    LocalScript --> sinfop
    LocalScript --> sbatchp
    LocalScript --> squeuep
    LocalScript --> scontrolp


    Users -->|SSH| sinfo[sinfo];
    Users -->|SSH| sbatch[sbatch];
    Users -->|SSH| squeue[squeue];
    Users -->|SSH| scontrol[scontrol];
    Users --> sinfop
    Users --> sbatchp
    Users --> squeuep
    Users --> scontrolp

    sinfop[sinfo*] --> |HTTP Request| slurmrestd;
    sbatchp[sbatch*] --> |HTTP Request| slurmrestd;
    squeuep[squeue*] --> |HTTP Request| slurmrestd;
    scontrolp[scontrolp*] --> |HTTP Request| slurmrestd;


    subgraph "Slurm controller"
        slurmrestd --> slurmctld[slurmctld];
        slurmctld --> slurmdbd[slurmdbd];
        sinfo --> slurmctld;
        sbatch --> slurmctld;
        squeue --> slurmctld;
        scontrol --> slurmctld;
    end

    subgraph "Compute Node"
        slurmd1[slurmd] 
    end

    subgraph "Compute Node"
        slurmd2[slurmd]
    end

    slurmctld --> slurmd1;
    slurmctld --> slurmd2;

    slurmdbd --> Database["Slurm Accounting Database"];

open_api_client example

import openapi_client
from openapi_client.models.v0039_job_submission import V0039JobSubmission
from openapi_client.models.v0039_job_submission_response import V0039JobSubmissionResponse
from openapi_client.rest import ApiException
import os
import json
from pprint import pprint

# Defining the host is optional and defaults to http://localhost
# See configuration.py for a list of all supported configuration parameters.
configuration = openapi_client.Configuration(
    host = "http://slurm-controller:6820"
)

# The client must configure the authentication and authorization parameters
# in accordance with the API server security policy.
# Examples for each auth method are provided below, use the example that
# satisfies your auth use case.

# Configure API key authorization: user
#configuration.api_key['user'] = os.environ["API_KEY"]

# Uncomment below to setup prefix (e.g. Bearer) for API key, if needed
# configuration.api_key_prefix['user'] = 'Bearer'

# Configure API key authorization: token
configuration.api_key['token'] = os.environ["SLURM_JWT"]

# Uncomment below to setup prefix (e.g. Bearer) for API key, if needed
# configuration.api_key_prefix['token'] = 'Bearer'

# Enter a context with an instance of the API client
with openapi_client.ApiClient(configuration) as api_client:
    # Create an instance of the API class

    api_instance = openapi_client.SlurmApi(api_client)

    #https://slurm.schedmd.com/rest_api.html#v0.0.42_job_submit_req


    #job_request = {'script': '#!/bin/bash', 'job': {'name': 'dajob'}}



    job_request = {
        "script": "#!/bin/bash",  
        "job": {
            "environment": ["ALL"],
            #"current_working_directory": "/home/testuser",
            #"name": "jname"
            #"output": "slurm-%j.out"
        }
    }


    # job_request = {
    #     "script": """#!/bin/bash
    #         ls > /home/hcadavid/temp_file.txt
    #         sleep 10
    #         """,  # Modified to include sleep command
    #     "job": {
    #         "environment": ["PATH=/bin/:/usr/bin/:/sbin/",
    #                         "MY_VAR=value1",
    #                         "ANOTHER_VAR=value2",                            
    #                         ],
    #         "current_working_directory": "/home/hcadavid",
    #         "name": "test slurmrestd job",
    #         "output": "slurm-%j.out"
    #     }
    # }

    # Convert the dictionary to a JSON string
    json_req_string = json.dumps(job_request, indent=2)


    #Empty job submission (from the documenation)
    #v0039_job_submission = openapi_client.V0039JobSubmission() # V0039JobSubmission | submit new job

    v0039_job_submission_instance = V0039JobSubmission.from_json(json_req_string)


    try:
        # submit new job
        api_response = api_instance.slurm_v0039_submit_job(v0039_job_submission_instance)
        print("The response of SlurmApi->slurm_v0039_submit_job:\n")
        pprint(api_response)
    except Exception as e:
        print("Exception when calling SlurmApi->slurm_v0039_submit_job: %s\n" % e)