Splunk App for VMWare: Why the DCN doesn't seem to collect anything?

[tsekali]

Hello everyone. A few days ago I setup splunk app for VMWare on a single-instance infrastructure with one indexer (for searching and indexing) hosted on Linux, a few universal forwarders for windows machines and an ESXi host. VMWare app was setup and a Data Collection Node was deployed from OVA file. I can see the DCN on the web interface with greek "ticks" and the vCenter as well, but the graphs are stuck to the "No Data" notice. Also, tried " sourcetype=vmware:perf* OR sourcetype=vmware:inv:hierarchy " but nothing. DCN collects nothing. Scheduler started, Gateway working, roles for user and admin are assigned, resources OK and vSphere 5.1 with WSDL (at least as I am told). Your assistance would be a lot appreciated!

[belka]

Check to see if your DCN has enough disk space in the $SPLUNK_HOME/ directory. The DCN vm ships with 5GB set, which is also the default minimum for the dispatcher to run. You can run a search or two, but then it runs out of space and you cannot search, that is, gather VMWare data at the DCN to forward to the VMWare App. You can increase the size of the $SPLUNK_HOME/ directory to fix it or configure the minimum disk space for the dispatcher to under 5 GB. The five GB limit makes it easier to ship the VM. A vm of say 10 GB storage would take up 10 GB and be harder to move around (5GB was hard enough).

Try that and see if that helps solve the problem.

[thomrs]

There is a hidden dashboard that may help. I found it very useful in troubleshooting.

/en-US/app/splunk_for_vmware/hydra_framework_status?earliest=-6h%40h&latest=now&form.namespace=vmware

For best results make sure you forward the DCN logs.

http://docs.splunk.com/Documentation/VMW/3.1.3/Configuration/Adjustdatacollectionnodecapabilities