We are experiencing that OBIEE 12c (newest patch) is down on random occasions. When it happens, OBIEE is down for a few minutes, then it gets back up.

we are seeing the following message in obis1-diagnostic:

[2019-05-23T12:20:15.657+01:00] [OBIS] [ERROR:1] [] [] [ecid: ] [sik: ssi] [tid: 170c]  [nQSError: 12002] Socket communication error at call=send: (Number=10054) An existing connection was forcibly closed by the remote host. [[

file: server\Utility\Comm\Src\SUCMSocket.cpp; line: 913

no other log files show error messages in the same time period

The only metalink issue I have found that describes this symptom is doc Id. 2496436.1

However, that issue describes a problem with the scheduler. Not with OBIEE going down. The issue describes a scenario where you have 3 nodes in a cluster. How do I figure out how many nodes our BI cluster has? We did a default installation of OBIEE 12c