SocketTimeoutException

Issue

What is the underlying cause of java.lang.RuntimeException: java.net.SocketTimeoutException

2021-06-18 07:55:41.652 ERROR [liferay/segments_entry_reindex-2][ElasticsearchIndexSearcher:165] java.lang.RuntimeException: java.net.SocketTimeoutException: 120,000 milliseconds timeout on connection http-outgoing [ACTIVE]
at com.liferay.portal.search.engine.adapter.search.SearchSearchRequest.accept(SearchSearchRequest.java:34)

Environment

Liferay DXP 7.3

Resolution

It is something that is more of an environmental-specific issue than a Liferay product issue.
It clearly means a SocketTimeoutException in the case of the accept(). Therefore, tweak the timeout barrier such that it is more flexible on network delays to avoid this issue might be tried. Especially if the server application will be running on a remote machine. Aside from that, the issue must be investigated to see what is causing network delays, such as a broken router.
There might be chances that consumers have a shorter TCP keep-alive timeout configured in their environment than the 120 seconds socket timeout we set on the HTTP client or their firewall may drop connections earlier than the keepalive probe could be sent. Therefore, it would be helpful in this matter to check whether they have any network device (Loadbalancer, Firewall, Proxy, etc.) between their client and server which has a session timeout or similar and either increase that timeout or lower the tcp_keep_alive kernel parameter.

Additional Information

The above just a workaround that can be done at the moment. However, regarding the reported concern here about what might be blocking the requests or what time is needed to set for the load balancer, this totally depends on how the environment is configured. So to move forward on resolving this, one needs to work with the infrastructure team/network team/operations team to identify any potential causes for early termination of requests.
There are some unofficial related articles below for more information.