Skip to content

Handle large exception

Alok Joshi requested to merge handle_large_exception into master

Problem symptom: Search service pods run into OOM issues, affecting service availability.

Problem scenario: Search API http request is made with too many clauses in the query.

Problem behavior: In such cases, Search service is responding correctly with 400 response code for input with too many clauses. When it tries to log this error/exception, Search service pod runs out of memory, due to the large size of error message coming from Elastic.

Solution: Log only part of the error message, when the error message is too large.

Edited by Alok Joshi

Merge request reports

Loading