Steinar Saugsta1 Super Contributor.
Super Contributor.

NNM1 10 and HA between two Datacenters



I'm planning to install NNMI 10.2x in HA (clustering). The requirements is that it should be resiliance between two different datacenters/locations. It will be running on Oracle Linux servers and there will be a remote Oracle DB. From my point of view the NNMi documentation is just talking about a solution for HA which I understand have to be in the same datacenter/location. Is it possible to configre a HA solution (iusing clustering) between two datacenter? Maybe I'm missing some knowledge which tells that he cluster could be "shared" between two datacenter.

Thanks in advacne for all feedback/advice.


Steinar S

2 Replies
Acclaimed Contributor.. AndyKemp Acclaimed Contributor..
Acclaimed Contributor..

Re: NNM1 10 and HA between two Datacenters

Geographical redundancy is not currently a part of the NNMI product right now, just local  physical.   We were faced with this situation as well when we migrated from NNM 7.53 to NNMi 8.1.  How we overcame this was to run two seperate installations in two different geographical locations, both with local redundancy (HA using Veritas VCS at that time, now we use application failover) .    Out implementation is Hot/Hot, but we use the northbound interface to send incidents  to our Manager of Managers and just swap that stream  between the two instances as needed for downtime, patching, and application issues.   Its nice enough that we havent had any downtime for years, one is always up.

Have a nice day 🙂

Andy Kemp
I've lasted longer in the technology industry than most certifications.
Absent Member.. dhendry Absent Member..
Absent Member..

Re: NNM1 10 and HA between two Datacenters


It is possible to configure your HA cluster between 2 datacenters. The component that you have to focus on is the clustering software that you are using. The premise is that NNMi will interact with the clustering software in order to control the HA functions. If you meet the requirements for the cluster configuration, then it does not matter where your servers are located.

1 - Latency. 5ms was the threshold we were using.  This may be different for your cluster software.
(Note: The RHCS documentation actually states the threshold as 2ms)

2 - Network. This may be obvious, but you need to configure the servers to be on the same VLAN.

For our setup, we are using RHEL 6.8 and RHCS (Red Hat Cluster Suite) with a Postgres DB. The 2 datacenters are in the same city. Currently we are running NNMi 10.10. The components are different, but the premise remains the same.

I hope that helps.

The opinions expressed above are the personal opinions of the authors, not of Micro Focus. By using this site, you accept the Terms of Use and Rules of Participation. Certain versions of content ("Material") accessible here may contain branding from Hewlett-Packard Company (now HP Inc.) and Hewlett Packard Enterprise Company. As of September 1, 2017, the Material is now offered by Micro Focus, a separately owned and operated company. Any reference to the HP and Hewlett Packard Enterprise/HPE marks is historical in nature, and the HP and Hewlett Packard Enterprise/HPE marks are the property of their respective owners.