Intermittent connectivity issues between Filebeat and Logstash

Elastic Stack Logstash
1

I have few Filebeat containers and a single logstash container running on a paticular namespaces of a kubernetes cluster.

However, I see intermittent connectivity issues between Filebeat and Logstash. (Logs below)
Due to this I see that Filebeat repeatedly sends the same events to Logstash.

<>|2020-01-23T15:43:14.925Z|ERROR|logstash/sync.go:155|Failed to publish events caused by: read tcp 100.96.10.3:59614->100.68.143.9:5044: i/o timeout|
|---|---|---|---|
|2020-01-23T15:43:16.813Z|ERROR|pipeline/output.go:121|Failed to publish events: read tcp 100.96.10.3:59614->100.68.143.9:5044: i/o timeout|
|2020-01-23T15:43:16.814Z|INFO|pipeline/output.go:95|Connecting to backoff(tcp://local-logstash:5044)|
|2020-01-23T15:43:16.818Z|INFO|pipeline/output.go:105|Connection to backoff(tcp://local-logstash:5044) established|</>

2

I have the same issue. Im running both Logstash and Filebeat 7.5.2 on Docker. Both containers are in the same docker-network on the same host. Moving the Filebeat out of Docker onto the host has the same issue. Recreating the containers+networks or restarting the Docker-daemon does not solve the problem either.
Filebeat manages to write events to Logstash, but is reconnecting every 60s due to this i/o timeout error.
As far as I can tell, there is no networking issue between these containers. Ping, telnet and curl do work flawlessly.

Edit: Using Filebeat and Logstash Image 7.5.1 does not have that issue in the very same setup.

3

Btw, I'm also able to telnet into logstash port from filebeat container