Deployment Architecture

Deployment app on multiple serverclass

nieminej
Loves-to-Learn Lots

I wanted to add same base configuration for workstations and have serverclasses divided by organizations but base app would be same on everyone.

Now I have problem: 

When you make changes (add host through webgui to one serverclass) and click save, it changes bundle epoch time under global_bundles and then other serverclasses say that file does not exist on server when clients try to download app. And then if I run reload deploy-server it's fine again. But everytime if I need to add client on any workstation serverclass it breaks all other serverclasses. It's pretty rough to run reload deploy-server command everytime because there will be pretty high load on the DS. 

Is there any other way to handle this than making class-specific base apps?

Running 9.4.1 12vCPU/12GB RAM. 

0 Karma

nieminej
Loves-to-Learn Lots

 

@kiran_panchavat 

But everytime I make changes through webGUI it makes serverclass reload and all other serverclasses lose bundle before I make full server reload 

0 Karma

isoutamo
SplunkTrust
SplunkTrust

When you have this kind (big) environment with several serverclasses etc. you should use only conf files with git to manage that environment!

With that way your DS wouldn't do anything else than what you tell it to do.

PLA1310C_Splunk .conf24 Interactive Workshop this is something you should read.

0 Karma

kiran_panchavat
SplunkTrust
SplunkTrust

@nieminej 

I'm uncertain about this, please open a Splunk support ticket to investigate the issue further.

Did this help? If yes, please consider giving kudos, marking it as the solution, or commenting for clarification — your feedback keeps the community going!
0 Karma

kiran_panchavat
SplunkTrust
SplunkTrust

@nieminej 

Instead of reloading the entire deployment server, you can reload specific server classes. This can be done using the command ./splunk reload deploy-server -class <serverclass name>. This way, only the changes made to the corresponding deployment apps will get reloaded, reducing the load on the deployment server. 

Instead of immediate reload deploy-server, use a scheduled reload (e.g., via cron) to batch updates:

In this situation, we can’t designate a specific server class.  With multiple server classes involved, setting up individual cron jobs for each one becomes impractical.

0 * * * * /opt/splunk/bin/splunk reload deploy-server -auth admin:<password>

  • The shared app stays static, and GUI changes only affect serverclass membership, not the app bundle itself. Batching reloads reduces DS load frequency

Reload deployment server via API/Splunk SDK

https://community.splunk.com/t5/Splunk-Dev/With-the-Splunk-Python-SDK-how-do-I-reload-deploy-client-... 

App:

Adds a simple |reloadds search command to reload deployment server configs from disk, as well as an alert action which does the same thing.

Another app in splunkbase to reload DS

https://splunkbase.splunk.com/app/7339 

Check this for more details: https://community.splunk.com/t5/Getting-Data-In/Deployment-Server-reload-configs-without-restarting-... 

REST call:

curl -ku admin:your_password https://your_splunk_server:8089/servicesNS/-/system/deployment/server/config/_reload 

Use the URL below, substituting your host name for myHostName and the serverClass you want to reload for serverClassName

https://myHostName:8089/services/deployment/server/serverclasses/serverClassName/reload 

Did this help? If yes, please consider giving kudos, marking it as the solution, or commenting for clarification — your feedback keeps the community going!
0 Karma

isoutamo
SplunkTrust
SplunkTrust
You shoudn't ever write admin password in clear text in any file (like cron) in any system. You have lost your splunk installation(s) security by that way!!!!
0 Karma
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!

Persistent Queue at TcpOut — One of Splunk's Most Practical Features

Splunk introduced persistent queueing at the tcpout layer as one of the most practical resilience features in ...

Skip the Awkward Silence: Have a .conf-ersation at .conf26

Picture this. You arrive at .conf26 already having your socializing and networking plans mapped out. No ...

Rethinking Zero Trust: From Product Purchases to Logical Control Evidence

Implementing Zero Trust (ZT) across complex environments often falters at the very beginning due to a ...