azure-functions-kafka-extension
azure-functions-kafka-extension copied to clipboard
Azure
Kafka extension for Azure Functions
Hi @TsuyoshiUshio - we have a function app vnet connected in premium plan with the kafka trigger. When the message intake is high , I am seeing the error that...
I am not getting the messages from kafka. The partitions are getting assigned as visible from application log as well as kafka queries so there doesnt seem to be any...
We have a k8s cluster that has 6 KafkaTrigger azure functions listening to 6 partitions. We have a time limit to consume all the messages produced on the topic were...
Hello, Ssl Config substitutions do not work on the HttpTrigger with Kafka. You need to hardcode. This is a huge config issue. They do work in the KafkaTrigger. SslKeyLocation =...
https://github.com/confluentinc/confluent-kafka-dotnet/issues/1544 We have Azure Load Balancer issue for getting timeout. For avoiding it, we can set the `connection max idle ms` as less than 4 min. Once the new version...
https://github.com/Azure/azure-functions-kafka-extension#linux-premium-plan-configuration The above doc suggests: /home/site/wwwroot/bin/runtimes/linux-x64/native when running on a Linux Premium Isolated ASE the correct path is: /home/site/wwwroot/runtimes/linux-x64/native
We want to use this trigger for consumers on a multi-region kafka topic, which requires use of the client.rack setting to specify which region a particular consumer should consume from....
Given this function `public static void InvoiceConsumer( [KafkaTrigger(Configs.BrokerList, "invoice", ConsumerGroup = "azfunc", Protocol = Configs.Protocol, AuthenticationMode = Configs.AuthenticationMode, Username = Configs.UserName, Password = Configs.Password)] KafkaEventData[] incomingEvents,...)` Where **both** MyTopicKey and...
When having multiple partition the trigger will scale up on high load but will not scale down on low load. The trigger will ask the instance/host to scale down but...
Scale in when lag is consitently below threshold and do not scale out when we increasing under the threshold