Select to view content in your preferred language

Configure DataStore in HA deployment

419
4
03-19-2025 09:59 AM
CarstenHogertz
Emerging Contributor

Hi community,

i've got a High Available ArcGIS Enterprise deployment. There are two VMs, both with an ArcGIS Server and DataStore installation. The AGS are active/active together in site. Now I registered the DS on the first VM with <ipVM1>:6443/arcgis. Everything worked fine.

My question is for the DS on the second VM. Should I use <ipVM1>:6443/arcgis or <ipVM2>:6443/arcgis?

Thanks for your help.

Tags (1)
0 Kudos
4 Replies
ArchitSrivastava
Frequent Contributor

Hello @CarstenHogertz ,

As per my understanding, we should use "<ipVM1>:6443/arcgis" as utilized for registering the first ArcGIS Data Store machine as it is recommended in the documentation.

For more details:

https://enterprise.arcgis.com/en/portal/latest/administer/windows/add-standby-machine.htm#ESRI_SECTI...

Hope it helps!

0 Kudos
Marc_Graham
Frequent Contributor

Hi @CarstenHogertz 

I'm not sure that your deployment pattern is optimal. I recommend you read some of the documentation like https://enterprise.arcgis.com/en/portal/latest/administer/windows/configure-highly-available-system.... 

You have not provided much detail on your deployment.  Do you have arcgis portal on each machine as well? What data stores are you referring to? Just relational?  Each different type of data store has different deployment patterns and requirements.  

 

0 Kudos
CarstenHogertz
Emerging Contributor

Hi @Marc_Graham,

The environment comprises four virtual machines.
VM1 & VM2 contain AGS & relational DS, VM3 & VM4 the portal. Everything is on high availability. The AGS are registered as hosted servers in the portal. The web adapters also run on dedicated VMs.


The DS backup runs on a shared network folder. Both DS backup directories point to the same folder on the shared network.


What would be your suggestion for configuring the DS?

0 Kudos
JakeSkinner
Esri Esteemed Contributor

Hi @CarstenHogertz, as @ArchitSrivastava mentioned, you can configure the second data store with VM1.  This will configure the second data store as the standby data store.  In the event the primary data store on VM1 is broken, it will fail over to the standby data store on VM2.

0 Kudos