[BUG] Cluster fails to deploy because create_slurm_accounts.py fails

[cartalla]

Describe the bug

When deploying a new cluster I sometimes get the following errors from create_slurm_accounts.py.
If I rerun the script then it passes.

2024-05-13 19:43:53,708 p=4087 u=root n=ansible | TASK [ParallelClusterHeadNode : Run /opt/slurm/config/bin/create_slurm_accounts.py to make sure it works] ***
2024-05-13 19:44:05,640 p=4087 u=root n=ansible | fatal: [local]: FAILED! => changed=true

  cmd: |-
    set -ex

    export SLURM_ROOT=/opt/slurm
    /opt/slurm/config/bin/create_slurm_accounts.py --accounts /opt/slurm/config/accounts.yml --users /opt/slurm/config/users_groups.json --default-account unassigned -d

DEBUG:2024-05-13 14:43:54,470: Checking account infrastructure existence and fairshare
    INFO:2024-05-13 14:43:54,470:     Creating account infrastructure with fairshare=10, parent=None

INFO:2024-05-13 14:43:59,127: Updating infrastructure account parent from None to root
    ERROR:2024-05-13 14:43:59,334: Couldn't set ParentName for account infrastructure to root.
    command: ['/opt/slurm/bin/sacctmgr', 'modify', '-i', 'account', 'infrastructure', 'set', 'Parent=root']
    output:
     Nothing modified
    Traceback (most recent call last):
      File "/opt/slurm/config/bin/create_slurm_accounts.py", line 152, in update_slurm
        subprocess.check_output([self.sacctmgr, 'modify', '-i', 'account', account, 'set', f'Parent={exp_parent}'], encoding='UTF-8', stderr=self.devnull) # nosec
      File "/usr/lib64/python3.9/subprocess.py", line 424, in check_output
        return run(*popenargs, stdout=PIPE, timeout=timeout, check=True,
      File "/usr/lib64/python3.9/subprocess.py", line 528, in run
        raise CalledProcessError(retcode, process.args,
    subprocess.CalledProcessError: Command '['/opt/slurm/bin/sacctmgr', 'modify', '-i', 'account', 'infrastructure', 'set', 'Parent=root']' returned non-zero exit status 1.

ERROR:root:Unhandled exception in /opt/slurm/config/bin/create_slurm_accounts.py
    Traceback (most recent call last):
      File "/opt/slurm/config/bin/create_slurm_accounts.py", line 354, in <module>
        app = SlurmAccountManager(args.accounts, args.users, args.default_account)
      File "/opt/slurm/config/bin/create_slurm_accounts.py", line 90, in __init__
        number_of_changes = self.update_slurm()
      File "/opt/slurm/config/bin/create_slurm_accounts.py", line 272, in update_slurm
        raise RuntimeError("Some slurm updates failed")
    RuntimeError: Some slurm updates failed
    Traceback (most recent call last):
      File "/opt/slurm/config/bin/create_slurm_accounts.py", line 354, in <module>
        app = SlurmAccountManager(args.accounts, args.users, args.default_account)
      File "/opt/slurm/config/bin/create_slurm_accounts.py", line 90, in __init__
        number_of_changes = self.update_slurm()
      File "/opt/slurm/config/bin/create_slurm_accounts.py", line 272, in update_slurm
        raise RuntimeError("Some slurm updates failed")
    RuntimeError: Some slurm updates failed

[cartalla]

I was able to reproduce the error with the following:

sacctmgr -i delete account infrastructure
export SLURM_ROOT=/opt/slurm
/opt/slurm/config/bin/create_slurm_accounts.py --accounts /opt/slurm/config/accounts.yml --users /opt/slurm/config/users_groups.json --default-account unassigned -d