How to resolve "ssl23_get_client_hello unknown protocol" error on indexer and "TcpOutputFd Read error" on forwarder?

[splunkreal]

Hello guys,

I'm using this on deployment-apps (universal forwarder deployment) :

[tcpout]
defaultGroup = default-autolb-group

[tcpout:default-autolb-group]
server = indexer:9997

[tcpout-server://indexer:9997]
sslCertPath = $SPLUNK_HOME/etc/apps/APP_OUTPUTS_BASE_PPR/local/server.pem
sslPassword = password
sslRootCAPath = $SPLUNK_HOME/etc/apps/APP_OUTPUTS_BASE_PPR/local/cacert.pem
sslVerifyServerCert = false

I'm 99% sure sslPassword for my cacert.pem is not 'password', so isn't it working for server.pem instead (default splunk cert) ?

If I use the correct cacert pass, I get ssl23_get_client_hello unknown protocol on indexer and TcpOutputFd Read error on forwarder.

Thanks for your clarification.

[gcusello]

Hi realsplunk,
sorry for this detail:
the steps you have to do are:

• insert correct password in inputs.conf (verify where is the inputs.conf file where there is sslPassword: $SPLUNK_HOME/etc/system/local) on your indexers;
• restart indexers, password will be encrypted
• insert correct password in inputs.conf in your TA (in local directory);
• restart forwarder, password will be encrypted

If you insert outputs.conf in default directory, password will be encrypted in local directory and not encripted in default directory.
Bye.
Giuseppe

[splunkreal]

So why at http://docs.splunk.com/Documentation/Splunk/6.2.3/Admin/Outputsconf we have :

sslPassword =
The password associated with the CAcert.
The default Splunk CAcert uses the password "password".
There is no default value.*

Thanks a lot!