Active-Active Setup

can you please elaborate. What is going into Active/Active, what SWI products are you using? What are other concerns?

This is the topic I am referring to:

https://solarwindscore.my.site.com/SuccessCenter/s/article/Orion-Active-Active-solution?language=en_US

The PDF as well:https://documentation.solarwinds.com/archive/pdf/orionplatform/Orion_Active-Active_Deployment_Guide.pdf

I am inclined to say that this might not be working anymore since v2022.x the whole database Schema and platform has changed since this version. I am asking in my peer groups if someone has seen this setup in the wild and post 2020 platform version.

if you like to discuss this over a call let me know and I can give you my contact details over a direct message.

i am located in Central Europe, so probably your mornings, my afternoons might be a good fit.

Alright, so we had a longer discussion about this. Credit goes to tobyw_loop1 dgsmith80 marcrobinson Deltona shuth 

The common theme is: Rather use HA than Acive-Active. Or use a separate instance where you only poll the info "duplicate" that you really need. Also the downside is, you need duplicate licenses.

The document you have is all publicly available info that we are aware of. Some of us are SWI Consultants and might have some "own experience documentation".

Either way, let us know if you want some more discussion with us around it.

Cheers.

 HerrDoktor  has mentioned most of the points. I have had my own share of problems in the past with Active-Active setup

Firstly why do we need an Active-Active setup ? If its mainly for Disaster Recovery then based on my experience I would go for a parallel setup altogether. 

1. Build a like to like setup, downside is definitely SolarWinds license cost plus operational cost (but then this can be controlled based on what sort of infrastructure you host your SolarWinds on), but then as well talk to SolarWinds about the licensing cost if your plan is to use the other setup only during Disaster Recovery.

2. There are several ways in which you can control your secondary environment, I am sure we have some drawbacks as it wont be completely automated but we can come online in less than 30 to 60 mins with a parallel setup OR even in 60 seconds

3. I had the exact same setup Production and DR which is like to like, when we onboard any configuration on Production I migrate the same configuration onto DR.

4. Do we need to poll the devices in parallel thats totally upto you based on your recovery time, in my case I dont poll the devices from both my environments rather I keep the devices in unmanaged state on the DR, but I make sure they are in both places Prod and DR and bring them online only when Prod is affected.

5. At times I as well poll it from both Production and DR, once in a month to make sure everything is intact, but then i make sure my alert actions are stopped on DR (which can anyways be  switched ON in a second when required).

6. If its on cloud to save operational cost we can as well shutdown our DR VM's and we can as well have a smaller DB as its not always polling the devices from DR. 

7. As I mentioned in the points above, how do you want it to be ? Main factors that you need to consider is Cost, Recovery Time and Restoration Time. But its definitely easier to maintain and manage a parallel setup.

Hope this helps.   

__PRESENT