UCS Central High Availability deployment

[qbgentili]

Hi All,

Looking to use UCS Central in a High Availability deployment.  I was reading through the guide and looking for clarification and advice regarding the shared storage that is required for the two VMs. Is it recommended to use iSCSI for the connection to the storage, or would a Raw Device Mapping (RDM) work just as well?

These VMs will be deployed in a VMware ESXi 5.1 environment with access to NetApp storage.

Thank you.

[Jeffrey Silberman · Accepted solution]

Hi Bryan,,

According to the Installation and Upgrade Guide, on page 21, it says "You must add the disk as an RDM".

Please keep in mind that if you're already running your hypervisor in H/A cluster mode, then you don't need UCS Central H/A Cluster mode as well --- standalone mode should be fine.   (Please check out Page 43 of the latest UCS Central Best Practice Guide).  

Hope this helps,

  -Jeff

[qbgentili]

Jeff,

Thanks for your reply.  That's what I was looking for. Also, thank you for the links to the documents!

Bryan

[Raghavendra P]

Hi Bryan

I have clarification and information on the UCS central to provide high availability in our environment. Below is my scenario:-

We are hosting UCS in two different to provide production and non-production environment and we planned to install UCS central to manage it in centralized location.

Considering the UCS central provided HA capability by using RDM or NFS volume, common method of UCS Central appliance installed in both datacenter (one in primary and one in secondary) and add all the UCS managers to single UCS central instance and if any of the DC fails we still have the central instance running in either of the DC. But in our environment RDM and NFS both are not standard solution with this assumption please suggest if we can configure UCS central like below explained.

Host UCS central VM appliance in primary datacenter with external database (MSSQL) connection for collecting the statistics during the central installation if in case of the appliance failure or datacenter failure occurs considering MSSQL database is replicated across to other datacenter using storage replication or SQL based replication can we recover in secondary datacenter with newer appliance deployment and pointing to the same SQL database either from primary datacenter or from replication recovered database to same configuration.

Also please suggest if any other method is feasible for our requirement. Thanks.

Regards

Raghu