Getting Data In

High Availability option available for Heavy forwarder .

Nraj87
Explorer

In Current Splunk deployment  we have 2 HFs, One used for DB connect another one used for the HEC connector and other.

And the requirement is if One HF goes done other HF can handle all the functions.  

so  is there High Availability option available for Heavy forwarder OR for DB connect APP ?

Labels (1)
Tags (1)
0 Karma

PickleRick
SplunkTrust
SplunkTrust

HF on its own does not have any "HA" or "clustering" feature. There are possibilities to manually craft some active-passive solutions but they require quite a lot of "external architecting" - doing a lot of stuff semi-manually on the OS level. I think there was a .conf presentation about it a few years ago but I can't find it.

0 Karma

gcusello
SplunkTrust
SplunkTrust

Hi @Nraj87 ,

HA is a continuous open issue for DB-Connect.

The easiest solution, as @isoutamo hinted, is to install DB-Connect in a Search Head Cluster, so the Cluster gives you the requested HA features, but not all the customers want to thave an input processor (like DB-Connect) in front end systems accessed by all the users.

I hope that Splunk will design a solution for this as soon as possible, for the moment, there's a request in ideas.splunk.com : https://ideas.splunk.com/ideas/EID-I-85 that you could vote.

There are two problems to solve for the HA:

  • the checkpoint,
  • the input enablement.

A not automatic workaround is to install DB-Connect on at least two HF and create a scheduled script that makes a KV-Store backup from the main HF and a restore in the secondary one, in this way you align the checkpoint between the two HFs, obviously in case or Disaster you'll have duplicated data for the period from the last alignment.

Then you have to manually start the secondary DB-Connect in case of Disaster and stop it when the Disaster period is closed.

It's a porkaround not a solution! waiting for the solution from Splunk that's late from many years (vote for it!).

Ciao.

Giuseppe

isoutamo
SplunkTrust
SplunkTrust
You could create own SHC cluster for HF without any normal SH activity from end user. There is no need to use regular end user SHC, just the functionality is what is needed!

gcusello
SplunkTrust
SplunkTrust

Hi @isoutamo ,

I proposed this solution to Splunk for a project and I'm waiting for an answer that probably will not arrive!

In my opinion it should work.

Ciao.

Giuseppe

0 Karma

isoutamo
SplunkTrust
SplunkTrust

There is coming DBX 4.0 version which already contains HA feature. I'm not sure when it's coming into GA, but you could check current discussions from slack https://splunk-usergroups.slack.com/archives/C22R341NG/p1736956361104829

 

0 Karma

isoutamo
SplunkTrust
SplunkTrust

Hi

some TAs will support some kind of HA e.g. DB Connect, but I think that most didn’t.

With DB Connect you could use SHC configuration for managing HA. I’m not sure how well this is currently working in general TAs? This needs some kind of mechanism to use distributed checkpoint status e.g. kvstore.

r. Ismo

0 Karma
Get Updates on the Splunk Community!

Technical Workshop Series: Splunk Data Management and SPL2 | Register here!

Hey, Splunk Community! Ready to take your data management skills to the next level? Join us for a 3-part ...

Spotting Financial Fraud in the Haystack: A Guide to Behavioral Analytics with Splunk

In today's digital financial ecosystem, security teams face an unprecedented challenge. The sheer volume of ...

Solve Problems Faster with New, Smarter AI and Integrations in Splunk Observability

Solve Problems Faster with New, Smarter AI and Integrations in Splunk Observability As businesses scale ...