Skip to content

Your POSIX identity on Linux and HPC offerings

When you get access to a Linux or HPC offering — for example a SLURM cluster or a storage service that provisions a Unix account for you — the service provider assigns you two numeric identifiers:

  • a UID (user id), which identifies your personal account, and
  • a primary GID (group id), which identifies your default group.

These numbers are how the underlying systems track file ownership, disk quotas and job scheduling. You normally never type them yourself: you log in with your username and SSH key, and the cluster maps you to your UID and GID behind the scenes.

What to expect

  • They are stable. Your UID and primary GID stay the same for the lifetime of your account, so files you create keep their ownership and your jobs are accounted to you consistently.
  • They are unique within a provider. Each service provider keeps its own pool of identifiers, so no two people on that provider's systems share a UID, and a user's id never collides with a project or role group id.
  • They are managed for you. You do not request or change these numbers — the provider allocates them automatically from a reserved pool when your account is created.

Your service provider can look up the UID and primary GID that were assigned to your account, which is useful when you open a support request about file permissions or scheduling.

Assigned UID and GID of offering users

Note

If you have accounts with more than one service provider, your UID and GID may differ between them — each provider numbers its own users independently. Within a single provider they stay consistent across all of that provider's offerings.