Logstash freezes post update to 6.2.4

Elastic Stack Logstash
1

I am running three logstash instances which were on 6.2.3 version. Today I upgraded all of them to 6.2.4. The update process was smooth on 2 of the instances and no problem was noticed.

On the third instance, logstash was in a "locked" up state post update. The logstash logs had no other entry except Pipeline started and the process state itself was S(Sleeping).Eventually the system load fell to 0. Then I looked at /var/log/messages which had the following error traceback (posted on pastebin as the traceback is too big)

https://pastebin.com/Uns7PKZX

This error was repeated a bunch of times. To recover, I had to empty the queue and restart logstash. After this the instance worked normally.

My pipeline has beats input plugin and elasticsearch and TCP output plugin.

2

Same here.

All ugly after this.
Apr 24 22:24:00 ent-mocsmlh04 systemd: Starting logstash...
Apr 24 22:24:22 ent-mocsmlh04 logstash: Sending Logstash's logs to /var/log/logstash which is now configured via log4j2.properties
Apr 24 22:25:22 ent-mocsmlh04 logstash: Exception in thread "Ruby-0-Thread-9@[main]>worker0: /usr/share/logstash/logstash-core/lib/logstash/pipeline.rb:385" org.logstash.ackedqueue.QueueRuntimeException: deserialize invocation error
Apr 24 22:25:22 ent-mocsmlh04 logstash: at org.logstash.ackedqueue.Queue.deserialize(Queue.java:733)
Apr 24 22:25:22 ent-mocsmlh04 logstash: at org.logstash.ackedqueue.Batch.deserializeElements(Batch.java:62)

3

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.