How does High Availability (HA) work on Fabric Interconnects with only Blade Servers attached?

[RedNectar]

Hi Experts,

I'm familiar with the way HA works with UCS B-Series Chassis - the Chassis has a SEEPROM that stores the cluster state which both Fabric Interconnects can see (and write to)

But how does HA work if you have only C-Series Blade servers attached, such as a Hyperflex CLuster?

I will assume that the Cluster State is written to one of servers (which one? the first one discovered???) but where? I imagine it has to be some FLASH somewhere, and I'm guessing the logical answer is the CIMC configuration storage.

But I'm only guessing.

Can anyone confirm this, or point me to a reference that explains it?

[neilmart · Accepted solution]

Hello,
If you run 'show cluster extended-state' from local-mgmt (or from the first UCSM command line prompt) it tells you exactly which servers the backup db's are stored. For Chassis' it is stored on the SEEPROM, as you mentioned. For Rack Servers I believe it is stored on the CIMC as you suggested.
Best Regards,

Neil Martin .ılı.ılı.
CX Technical Consultant Engineer
Cisco Unified Computing Systems and Server Virtualization

SUSE Certified Administrator
In Enterprise Linux 12

Vmware Certified Professional
Data Center Virtualization 2019