ApsaraMQ for Kafka:FAQ

Billing

• Why am I charged for pay-as-you-go or Serverless instances even if I do not use them?

• How do I convert an instance from topic-based billing to partition-based billing?

• Why am I charged based on peak traffic?

• How does peak traffic billing work?

• Why am I charged for adding extra partitions?

• How does billing work for connector ecosystem integration?

Networking

• Cannot ping the endpoint of an ApsaraMQ for Kafka instance

• How can an ECS instance access an ApsaraMQ for Kafka instance in a different zone?

• Can I access an ApsaraMQ for Kafka instance over the public network?

• How do I access an ApsaraMQ for Kafka instance through a proxy?

• Do I have to select the 172.16.0.0/12 CIDR block for a classic network?

• Is interconnection supported between different VPCs?

Client errors

• Client errors and solutions when using ApsaraMQ for Kafka

• ApsaraMQ for Kafka: "Not authorized to access group" error message

Sending messages

• Where can I find best practices for sending messages?

• How do I test message sending?

• After sending a message from a client, how do I confirm that it was sent successfully?

• How many connections does a producer establish to brokers?

• Does setting a callback in a Java client affect the message sending speed?

• Why is the sending latency long for PHP?

• How do I configure client parameters?

• Client errors and solutions when using ApsaraMQ for Kafka

• Cannot ping the endpoint of an ApsaraMQ for Kafka instance

• Why do I receive the "Cluster authorization failed" error when I send messages after I enable the idempotence feature?

Consuming messages

• Best practices for subscribers

• How do I troubleshoot exceptions when a client first connects to ApsaraMQ for Kafka?

• Why does frequent rebalancing occur on the consumer?

• Why does the consumer fail to pull messages or pull messages slowly?

• Why is it not recommended to use the Sarama Go client to send and receive messages?

• What causes message accumulation when I use ApsaraMQ for Kafka?

• The 'Not authorized to access group' error message is returned by ApsaraMQ for Kafka

Instances

• What open source versions do instances support?

• How do I select an instance type?

• How do I obtain the VPC information required for instance deployment?

• What are the possible impacts of a broker upgrade?

• You cannot change an instance's region

• When does ApsaraMQ for Kafka delete old messages?

• How do I quickly test whether the ApsaraMQ for Kafka server-side is working correctly?

• Are delayed messages supported?

• Can I access an ApsaraMQ for Kafka instance over the public network?

• Is interconnection supported between different VPCs?

• Do I have to select the 172.16.0.0/12 CIDR block for a classic network?

• Is message compression supported?

• Why are messages not deleted after they expire?

• Why is there a large drift between the creation time of a queried message and the specified query time point?

• Instance Health Check Guide

• How do I recover an instance?

• How do I release an instance?

• Why does it take a long time to upgrade an instance version?

Topics

• Why is the total number of topics (and partitions) limited?

• Why can't the number of partitions in a topic be reduced?

• Why are messages sent to a topic unevenly distributed among partitions?

• What are the storage types for topics?

• If an instance is billed by topic, how are the numbers of topics, groups, and partitions calculated?

• Why is the partition time for a topic displayed as 1970 or "--"?

• Why can't the number of partitions be reduced after a topic is deleted?

• Is the compact cleanup policy supported?

• What are the common issues related to automatic topic creation?

• Topic quota for adding partitions

• For an instance that is billed by topic, what is the number of topics and partitions included in the traffic specification?

• What do I do if topic partitions are skewed?

• How do I handle inactive topics?

Groups

• What specifications are available?

• Why is the partition time for a topic displayed as 1970?

• Why can messages be consumed even if the group does not exist?

• Why is there a large difference in partition consumption times, significant lag, or out-of-order consumption?

• What does the latest consumption time shown in the console mean?

• How do I view which IP addresses are consuming messages?

• Where can I find best practices for consumption?

• Does mounting an NFS on the consumer affect the consumption speed?

• How do I manage consumer offsets in the console?

• Why can't I see consumer client information in the console?

• Why can't I see the subscription relationship of a group in the console?

• Why are consumer offsets not deleted after they expire?

• Why is the same partition consumed by multiple consumer threads?

• Why has the number of groups on the Group Management page increased?

• Why is the status of a group always "Deleting"?

• Why is the message accumulation for a group "0" or not displayed?

API issues

• The GetConsumerProgress API call returns a LastTimestamp value of -1

Monitoring and alerting

• Why do I need to upgrade my instance to support the monitoring and alerting feature?

• What do I do if the status of an alert item shows "Insufficient Data"?

• How do I monitor open source Apache Kafka?

• Why can an Alibaba Cloud account view monitoring information but a RAM user cannot?

• Why can't I log on to the machine where ApsaraMQ for Kafka is deployed?

• Why do I still receive alerts for message accumulation after I delete a group?

• Why is the message accumulation in alerts inconsistent with the message accumulation in the console?

• Issues with Monitoring Data

Ecosystem integration

• Filebeat Sending Failures