Deployment Architecture

Server Migration

AMAN0113
Explorer

We are planning to migrate a server that plays multiple roles as a DS, HEC, Proxy, SC4S, Syslog etc., to multiple servers by possibly trying to split the roles. Eg; server A to play DS role, server B to take care of HEC Services and so on.
What would be the easiest approach to achieve this?  Seems like a lot of work. Would it be recommended to do so in the first place? What are the criteria we should have in mind while doing this migration.

0 Karma

gcusello
SplunkTrust
SplunkTrust

Hi @AMAN0113 ,

it's possible to share the DS only if it has to manage less than 50 clients.

Even if, the most work in on the devices:

if you are managing the addressing od the Deployment Server using a dedicated app, it's very easy to move the DS to another server, if not it's better to move the other servers.

Anyway, To move the DS, you should create a custom Add-On, called e.g. TA_deploymentClient, containing only two files:

  • apps.conf
  • deplymentclient.conf

So these are the steps to move the DS:

  • on the old DS, add a new serverclass  for all clients pointing to the new add-on TA_deploymentClient
  • on all clients remove deploymentclient.conf from the %SPLUNK_HOME/etc/system/local folder
  • restart Splunk on all clients
  • verify (on the DS) that in all clients there's a new app installed
  • install Splunk on a new system,
  • copy apps from $SPLUNK_HOME/etc/deployment-apps to the same folder of the new system
  • copy %SPLUNK_HOME/etc/system/local/serveclasses.conf in thesame folder of the new system,
  • push new new configurations

To move the HEC, you have to re-create the inputs, eventually using the same tokens, but, in the devices, you have to point to the new destination.

It could be easier if you have a Load Balancer in front of the the HEC servers, becausein this case, you have only to modify the configuration in the LB.

Same thing for the syslog receivers: you have to  copy all the inputs from th old to the new servers and then move the pointing adresses in each device.

See what' the highest number of connection you have (syslog, clients or HEC) and maintain it in the original server, moving the others.

Ciao.

Giuseppe

Got questions? Get answers!

Join the Splunk Community Slack to learn, troubleshoot, and make connections with fellow Splunk practitioners in real time!

Meet up IRL or virtually!

Join Splunk User Groups to connect and learn in-person by region or remotely by topic or industry.

Get Updates on the Splunk Community!

Optimize AI at Scale: Must-Attend Observability Sessions at .conf26

conf26   With nearly 70 breakout sessions on the docket, the .conf26 Observability track is designed to help ...

How much can you really learn in 3 minutes?

Observability can certainly be hard to understand – there's a lot of jargon and buzzwords and it seems to ...

Event Series: The Agentic SOC: Trust Before Autonomy

AI is fundamentally changing security operations, but true progress requires more than just automation—it ...