Random "Timed out ProduceRequest in flight" messages #4116
Unanswered
CatalinCojocaru
asked this question in
Q&A
Replies: 1 comment
-
Bom dia, conseguiu resolver? |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
I am getting some random timeout errors while publishing messages using Confluent.Kafka. The application runs in a Kubernetes cluster and is built using the .NET 6 framework.
When the default timeout (60000ms) is reached, the message is successfully published.
These are the logs:
And by changing
SocketTimeoutMs
from 60000ms to 5000ms, the message is published after 5 seconds:I can't find a pattern for this behavior, but it happens pretty often I would say, both when the application is deployed and on my local environment.
Based on the logs from App Insights for one of our environments:
And from another environment where we don't have to many messages published:
I tried changing the default values for some of the configuration settings, like the default value for
LingerMs
from 5ms to 100ms, but didn't help.Does anyone have any idea what could be the reason behind this? I don't know if there is something wrong with my code/configuration, or something wrong with the Kafka broker.
The broker is not owned or maintained by our team, but I know it is built on top of Apache Kafka, using Strimzi, running in Kubernetes.
Client configuration:
CompressionType: CompressionType.Snappy
Acks: Acks.Leader
SecurityProtocol: SecurityProtocol.SaslSsl
EnableSslCertificateVerification: true
MessageMaxBytes: 100000000
SocketKeepaliveEnable: true
MessageSendMaxRetries: 10
RetryBackoffMs: 100
LingerMs: 100
SocketTimeoutMs: 5000
Confluent.Kafka nuget version:
1.9.3
Operating system:
aspnet:3.1.28-bionic
More logs with
Debug = "broker,topic,msg"
Beta Was this translation helpful? Give feedback.
All reactions