EAC Doesn't Handle Correctly The Next Dgraph Startup After Dgraph Failed To Start And "Failed" State Was Saved Into EAC

(Doc ID 2329019.1)

Last updated on NOVEMBER 16, 2017

Applies to:

Oracle Commerce Guided Search / Oracle Commerce Experience Manager - Version 11.3 to 11.3 [Release 11.x]
Information in this document applies to any platform.

Symptoms

In certain conditions (ie. performance issues when running multiple Baselines at once on ITL Host or under heavy load), Dgraphs might not start within the configured Dgraph startup timeout value (120 seconds by default).

When reaching the configured Dgraph startup timeout value, the Endeca Application Controller (EAC) gives an error (ie. Component 'AuthoringDgraph' did not start in startup timeout of 120 seconds), but the Dgraph actually finishes the startup procedure. If this happens during a Baseline, Baseline will stop completely.

Also, the Dgraph state is changed to "Failed" in EAC.

Afterwards, EAC doesn't handle correctly the Dgraph "Failed" state when the next Dgraph startup is requested (ie. Baseline, runcommand DistributeIndexAndApply, runcommand AuthoringDgraph cycle etc.), although the Dgraph starts and the "Failed" state is changed into "Running" state.

EAC fails instantly and gives an error message (ie. Server component 'AuthoringDgraph' failed to start), indicating to further look into the Dgraph logs.
But the Dgraph logs show only the normal startup procedure.

Afterwards, when the next Dgraph startup is requested (as in, when the Dgraph state is other than "Failed"), there are no more errors.

Changes

 

Cause

Sign In with your My Oracle Support account

Don't have a My Oracle Support account? Click to get started

My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms