DB Connect 2 and Search Head Clustering: Should resource pool members be part of the SHC, or separate nodes outside of the SHC?

[a212830]

Hi,

I am working on upgrading to Splunk 6.2 and DB Connect 2. In my lab, I have SHC setup, and DB Connect2. Looking at the doc, it looks like db connect on the captain sends request to resource pool members to do the work. Should the resource pool members be part of the SHC, or should they be separate nodes, outside of the SHC.

[jcoates_splunk]

separate nodes, they should not be SHC members.

[a212830]

Thanks. So, which layer then? Are they independent, and not managed by any of the mgmt tools? (Deployer, CM, Deployment Server)...

[jcoates_splunk]

that's correct -- I assume that if you need pools in addition to SHC, it's because you're pumping data at a significant rate. Forwarders are the ideal home for that.

[bandit]

Wouldn't it be better to have this hosted on the SHC for redundancy of the database feeds? If I lose the forwarder forwarder, don't I lose all db feeds?

[jcoates_splunk]

That would be better in theory, but in practice it will not work. On an SHC, only the captain will run DBX jobs.