libstoragemgmt: Add support for retrieving hwraid mode from storage t…

[joehandzik]

…argets

Signed-Off-By: Joe Handzik joseph.t.handzik@hpe.com

[joehandzik]

We'll see if we hit the same failures. Again, please let me know if there's something structurally wrong with my PRs. My cli code looks...slightly different from yours. But when I try to conform to the model in the rest of the code, I get runtime errors. So let me know if my cli code as it stands is inadequate in particular.

[cathay4t]

Please try to use enum lsm_system_hwraid_mode_type instead of uint32_t *hwraid_mode.

[cathay4t]

Please be informed, once #51 merged,
this pull request needs the rebase.

[joehandzik]

Yup, definitely (re: rebase). Thanks for documenting!

[joehandzik]

Hmm, started looking into the enum usage, and I can't say that I agree. If this was a greenfield class I'd implement an enum, but since I put the system mode inside the System python class, I need to make sure the values I have there don't conflict with the corresponding C values for system statuses (see: STATUS_UNKNOWN, STATUS_OK, etc). I then defined values for LSM_SYSTEM_HWRAID states that wouldn't conflict with the other LSM_SYSTEM values. If I create an enum, I wouldn't necessarily be protected from accidentally setting a future HWRAID value equal to an existing LSM_SYSTEM value.

tldr; I just emulated what already exists for LSM_SYSTEM #defines, and I think it would be more confusing to implement some LSM_SYSTEM values differently from the way existing LSM_SYSTEM values are defined. Is that assertion reasonable, or do you still want me to aggregate LSM_SYSTEM_HWRAID values into an enum?

[cathay4t]

@joehandzik To avoid overlap, I am working on this pull request:

1. Use similar design in pull request New API for retrieving system firmware version. #56 .
2. Change it to enum lsm_system_hwraid_mode_type to enum lsm_system_mode_type
   which could expand to include 'SAN, NAS, SAN&NAS, OBJ and etc` values in the future.
3. Add MegaRAID support.

[joehandzik]

Sounds good, sorry for the delay @cathay4t.