Solarwinds HA Pool configuration in Multi subnet Scenario | How to configure this without

We are working on HA pool configuration and having concerns regarding using a DNS Admin user for SolarWinds HA pool configuration.

Is there any workaround to configure SolarWinds HA effectively without using DNS Admin user, like using a NetScaler or LB or any other methods ?

Looking forward your prompt and expert opinion.

Find more posts tagged with

solarwinds ha

Accepted answers

All comments

chad.every

Yes, this article references using a Load Balancer with the SolarWinds Platform.
https://documentation.solarwinds.com/en/success_center/orionplatform/content/core-load-balancer-requirements.htm

Using High Availability DNS/VIP options are only for the convenience when a load balancer is not available. Starting in version 2025.1, we provided the option to disable SolarWinds managed DNS/VIP for customers using a load balancer.

akhilraj891

Hi Chad , thanks for the update.

I believe , LB can be used only for WebServers and Polling engines right . In our Scenario we have to setup DR for Primary Orion . If Primary DC goes down Secondary Orion should be active.

From the guide , I understand it can be configured using HA/DR pool (Muttisubnet) + DNS .

Can you please help with steps if above scenario can be achieved without DNS dependency ?

If incase the link between Primary and Secondary goes down , both Primary and Secondary became active -active correct ? How to handle this situation ?

chad.every

You can use a network load balancer for the MPE engines too. You can use your load balancers to handle the DNS aspect (depending on your LB setup, e.g. Global NLB vs Local NLB)

The biggest items to know when using HA for DR are:

The Database is the most critical part. Make sure your database can failover between sites and that the SQL Listener does not change. The SolarWinds Platform will attempt to reconnect to the listener if it goes down. Depending on how you have your database configured, this might require manually updating the SQL Listener DNS accordingly.
In the event of a complete outage and connectivity between sites is down, there is a high chance of having 'split brain' issues with HA, where both engines will think they are primary. We have plans in the future to minimize those occurrences by allowing a 3rd member to sit as a 'witness only', but that is a ways out. In the scenarios when split brain occurs, it would be best to contact support with a 'system down' case so they can fix it. We do have plans to introduce a repair option for HA in the near future so contacting support would not be required. As long as everything goes as planned, that repair option would come in the 2025.4 release.