After a few minutes (5/10min) of POST requests, I get this error : {“ok”:false,”message”:”backend closed connection”} in my console but I do not see error in App Search API logs... How is it possible ?
What can be the cause of this error ?
It's hard to tell without more information, but my guess would be, the message ("backend closed connection") didn't come from the App Search API. The error might have happened before reaching Enterprise Search server, somewhere between the proxy server and the Enterprise Search server.
Are you running on Cloud? In the admin console, you should have "Enterprise Search logs" and "Proxy logs". The records there might tell you more about what's going on.
@StevenMartins The "service unavailable" message suggests that the Enterprise Search instance is not able to handle the incoming requests, because something is putting too much pressure on the server, or the configured capacity is not sufficient.
I would advise to restart the deployment, and if that doesn't help, contact support, so they can open a ticket to investigate the problem with your deployment more closely.
I view 503 Service Unavailable, but there is a 502 "[...] read: connection reset by peer" before and I there are nothing in metrics that confirms insufficient capacity
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.