What can be done when an indexer doesn't join the cluster?

[ddrillic]

We have a cluster of four nodes and one of them just crashed. We brought it up, but it hasn't joined the cluster. Rolling restart is an operation of over an hour in this case. So, is there any other way to make this node join the cluster?

 ./splunk list cluster-peers

Shows the node as down, but the Splunk server is actually up on this node.

[akocak · Accepted solution]

it is good that it rejoined, if it didn't , I generally run below on splunk cli (splunk_home\bin) on working and not working indexer.

(./)splunk cmd btool server list --debug
(./)splunk cmd btool outputs list --debug

if you have it ssl enabled also:

(./)splunk cmd btool inputs list --debug

then compare values on simple textdiff tool. Since you configure your indexer from deployment server, any misconfiguration one would appears.

also search internal logs for errors in:
index=_internal source=splunkd.log -> to get an overview of what splunkd is doing
source=splunkd_access.log -> to see all incoming REST calls & response codes
source=metrics.log -> to see metrics about how splunk is performing

https://conf.splunk.com/files/2016/slides/indexer-clustering-basics-internals-and-debugging.pdf
http://docs.splunk.com/Documentation/Splunk/6.6.3/Troubleshooting/Usebtooltotroubleshootconfiguratio...

[ddrillic]

It rejoined by itself after some time ...