Post by josephreynolds1 on Jun 15, 2021 16:48:34 GMT
What is the right way to model BMC FieldMode in Redfish?
> Keeps track of whether the BMC is in field mode. Enabling field mode is intended to be a one-way operation. Therefore the application implementing this interface could return a NotAllowed error when the user attempts to set the property value to false.
Can FieldMode be added as a property under the Manager type? Or should this be a Manager.Oem property? Or would this be better modeled as a BIOS setting?
Per github.com/openbmc/phosphor-dbus-interfaces/blob/master/yaml/xyz/openbmc_project/Control/FieldMode.interface.yaml, field mode is:
> An interface to enable field mode, which, when enabled, indicates that the system is in production.> Keeps track of whether the BMC is in field mode. Enabling field mode is intended to be a one-way operation. Therefore the application implementing this interface could return a NotAllowed error when the user attempts to set the property value to false.
Here is how I understand FieldMode false/disabled:
- The primary use case is for manufacturing. FieldMode is false/disabled during early parts of the manufacturing process and is locked into the true/enabled state prior production use.
- Examples of BMC behavior changes when FieldMode is disabled:
- Additional functions are enabled, for example setting MAC addresses.
- Tolerances are reduced, for example a smaller number of retries before signalling an error.
- Error handling is changed. For example, the system may stop on error and not try to recover.
I understand the Redfish spec focus is on users. However, users may wish to ensure FieldMode is enabled, for example, as part of a "security settings" page.
Can FieldMode be added as a property under the Manager type? Or should this be a Manager.Oem property? Or would this be better modeled as a BIOS setting?