You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When using a lot of different dissect filters, in particular, where all of these are matching against a single event, the warning messages from failed matches spams the log with messages of the following:
[2018-09-20T14:00:14,305][WARN ][org.logstash.dissect.Dissector] Dissector mapping, field not found in event {"field"=>"somefield", "event"=>{"host"=>"ludicra", "@timestamp"=>2018-09-20T04:00:14.108Z, "@version"=>"1", "message"=>"sdgasdg"}}
There is one of these messages for each failed match. On a busy server with lots of patterns, this adds up. Can we reduce the level of these messages to DEBUG to avoid spamming the logs?
The text was updated successfully, but these errors were encountered:
Sure. I can do that. I think it is a bug as that warning is supposed to be rare. I fixed it while implementing a POC of sequencing.
Be aware, though, the addition of sequencing should ameliorate the need for multiple dissects.
This has a performance impact as well. Some users are seeing a perf hit from all the additional logging. When the Logstash logs were diverted to /dev/null, dissect was much faster than grok, but when the failures were sent to the log file they were about the same.
When using a lot of different dissect filters, in particular, where all of these are matching against a single event, the warning messages from failed matches spams the log with messages of the following:
There is one of these messages for each failed match. On a busy server with lots of patterns, this adds up. Can we reduce the level of these messages to DEBUG to avoid spamming the logs?
The text was updated successfully, but these errors were encountered: