-
Notifications
You must be signed in to change notification settings - Fork 14
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
update(MirrorMaker): improve configuration concepts (#586)
Co-authored-by: Harshini Rangaswamy <harshini.rangaswamy@aiven.io>
- Loading branch information
1 parent
2f4654f
commit 6fd6d73
Showing
3 changed files
with
120 additions
and
47 deletions.
There are no files selected for viewing
139 changes: 92 additions & 47 deletions
139
docs/products/kafka/kafka-mirrormaker/concepts/mirrormaker2-tuning.md
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -1,50 +1,95 @@ | ||
--- | ||
title: MirrorMaker 2 common parameters | ||
title: Configuration parameters for Aiven for Apache Kafka® MirrorMaker 2 | ||
--- | ||
|
||
MirrorMaker 2 (MM2) offers a suite of parameters to help with data replication and monitoring within Apache Kafka® ecosystems. | ||
This topic outlines common parameters you can adjust, along with tips for | ||
validating MM2's performance. | ||
|
||
1. Increase the value of `kafka_mirrormaker.tasks_max_per_cpu` in the | ||
advanced options. Setting this to match the number of partitions can | ||
enhance performance. | ||
1. Ensure the interval seconds for the following settings match. You | ||
can reduce these intervals for more frequent data synchronization: | ||
- Advanced options: | ||
- `kafka_mirrormaker.emit_checkpoints_interval_seconds` | ||
- `kafka_mirrormaker.sync_group_offsets_interval_seconds` | ||
- Replication flow: | ||
- `Sync interval in seconds`. | ||
1. To exclude internal topics, add these patterns to your topic | ||
blacklist: | ||
- `.*[\-\.]internal` `.*\.replica` `__.*` `connect.*` | ||
1. Depending on your use case, consider adjusting these parameters: | ||
- `kafka_mirrormaker.consumer_fetch_min_bytes` | ||
- `kafka_mirrormaker.producer_batch_size` | ||
- `kafka_mirrormaker.producer_buffer_memory` | ||
- `kafka_mirrormaker.producer_linger_ms` | ||
- `kafka_mirrormaker.producer_max_request_size` | ||
|
||
## MirrorMaker 2 validation tips | ||
|
||
To ensure MirrorMaker 2 is up-to-date with message processing, monitor | ||
these: | ||
|
||
1. **Consumer lag metric**: Monitor the `kafka.consumer_lag` metric. | ||
|
||
1. **Dashboard metrics**: If MirrorMaker 2 stops adding records to a | ||
topic, the `jmx.kafka.connect.mirror.record_count` metric stops | ||
increasing, showing a flat line on the dashboard. | ||
|
||
1. **Retrieve latest messages with \`kt\`**: Use | ||
[kt](https://github.com/fgeller/kt) to retrieve the latest messages | ||
from all partitions with the following command: | ||
|
||
``` | ||
kt consume -auth ./mykafka.conf \ | ||
-brokers SERVICE-PROJECT.aivencloud.com:PORT \ | ||
-topic topicname -offsets all=newest:newest | \ | ||
jq -c -s 'sort_by(.partition) | .[] | \ | ||
{partition: .partition, value: .value, timestamp: .timestamp}' | ||
``` | ||
Learn about the configuration layers in Aiven for Apache Kafka® MirrorMaker 2, including service, replication flow, and integration settings. | ||
Optimize data replication and performance in your Kafka ecosystem. | ||
|
||
## Configuration layers | ||
|
||
Aiven for Apache Kafka® MirrorMaker 2 configurations are organized into three layers: | ||
**service**, **replication flow**, and **integration**. Each layer controls a specific | ||
aspect of the replication process. | ||
|
||
### Service configurations | ||
|
||
Service configurations control the behavior of nodes and workers in the | ||
Aiven for Apache Kafka® MirrorMaker 2 cluster. | ||
|
||
**Example of a service configuration**: | ||
|
||
- Parameter: [`kafka_mirrormaker.emit_checkpoints_enabled`](https://aiven.io/docs/products/kafka/kafka-mirrormaker/reference/advanced-params#kafka_mirrormaker_emit_checkpoints_enabled) | ||
- Description: Enables or disables periodically emitting consumer group offset | ||
checkpoints to the target cluster. | ||
- Impact: | ||
- Automatically restarts the workers. | ||
- Restarts all connectors and tasks. | ||
|
||
### Replication-flow configurations | ||
|
||
Replication-flow configurations manage the behavior of connectors, such as Source, Sink, | ||
Checkpoint, and Heartbeat. | ||
|
||
**Example of a replication-flow configuration**: | ||
|
||
- Parameter: [`topics`](https://registry.terraform.io/providers/aiven/aiven/latest/docs/resources/mirrormaker_replication_flow) | ||
- Description: Specifies a list of topics or regular expressions to replicate. | ||
For more information, see the [topics included in a replication flow](/docs/products/kafka/kafka-mirrormaker/concepts/replication-flow-topics-regex). | ||
- Impact: | ||
- Automatically restarts the affected connectors. | ||
- Restarts their associated tasks. | ||
|
||
### Integration configurations | ||
|
||
Integration configurations fine-tune the interaction between producers and consumers | ||
within connectors. | ||
|
||
**Example of an integration configuration**: | ||
|
||
- Parameter: [`consumer_fetch_min_bytes`](https://registry.terraform.io/providers/aiven/aiven/latest/docs/resources/service_integration#nested-schema-for-kafka_mirrormaker_user_configkafka_mirrormaker) | ||
- Description: Sets the minimum amount of data the server should return for a fetch | ||
request. | ||
- Impact: | ||
- Automatically restarts the workers. | ||
- Restarts all connectors and tasks. | ||
|
||
:::note | ||
Most configuration parameters are derived from | ||
[KIP-382: MirrorMaker 2.0 - Configuration Properties](https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=95650722#KIP382:MirrorMaker2.0-ConnectorConfigurationProperties). Refer to this resource for additional details. | ||
::: | ||
|
||
## Common parameters | ||
|
||
This section describes common parameters that can be adjusted to optimize the performance | ||
and behavior of Aiven for Apache Kafka MirrorMaker 2 replication. | ||
|
||
1. **Optimize task allocation**: | ||
Increase the value of | ||
[`kafka_mirrormaker.tasks_max_per_cpu`](/docs/products/kafka/kafka-mirrormaker/reference/advanced-params#kafka_mirrormaker_tasks_max_per_cpu) | ||
in the advanced configuration. | ||
Setting this to match the number of partitions can improve performance. | ||
|
||
1. **Align interval settings**: | ||
Ensure the following interval settings match to achieve more frequent and synchronized | ||
data replication: | ||
- **Advanced configurations**: | ||
- [`kafka_mirrormaker.emit_checkpoints_interval_seconds`](/docs/products/kafka/kafka-mirrormaker/reference/advanced-params#kafka_mirrormaker_emit_checkpoints_interval_seconds) | ||
- [`kafka_mirrormaker.sync_group_offsets_interval_seconds`](/docs/products/kafka/kafka-mirrormaker/reference/advanced-params#kafka_mirrormaker_sync_group_offsets_interval_seconds) | ||
- **Replication flow**: | ||
- `Sync interval in seconds` | ||
|
||
1. **Exclude internal topics**: | ||
Add these patterns to your topic blacklist to exclude internal topics: | ||
- `.*[\-\.]internal` | ||
- `.*\.replica` | ||
- `__.*` | ||
- `connect.*` | ||
|
||
1. **Adjust integration parameters**: | ||
Modify these integration parameters based on your use case to improve producer and | ||
consumer performance: | ||
- `consumer_fetch_min_bytes` | ||
- `producer_batch_size` | ||
- `producer_buffer_memory` | ||
- `producer_linger_ms` | ||
- `producer_max_request_size` |
27 changes: 27 additions & 0 deletions
27
docs/products/kafka/kafka-mirrormaker/howto/monitor-replication-execution.md
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,27 @@ | ||
--- | ||
title: Monitor replication execution | ||
--- | ||
Apache Kafka® MirrorMaker 2 uses Kafka® Connect for monitoring and state management, helping you track replication flows and address issues. | ||
|
||
## Monitoring tips | ||
|
||
Follow these tips to ensure that replication is up-to-date with message processing: | ||
|
||
1. **Monitor consumer lag:** Use the `kafka.consumer_lag` metric to track replication | ||
progress and identify delays. | ||
|
||
1. **Track dashboard metrics:** Check the `jmx.kafka.connect.mirror.record_count` metric. | ||
If MirrorMaker 2 stops adding records to a topic, this metric will show a flat line, | ||
indicating no new records are being replicated. | ||
|
||
1. **Retrieve the latest messages using `kt`:** Use the | ||
[**kt**](https://github.com/fgeller/kt) tool to fetch the latest messages from all | ||
partitions. Run the following command: | ||
|
||
```bash | ||
kt consume -auth ./mykafka.conf \ | ||
-brokers SERVICE-PROJECT.aivencloud.com:PORT \ | ||
-topic topicname -offsets all=newest:newest | \ | ||
jq -c -s 'sort_by(.partition) | .[] | \ | ||
{partition: .partition, value: .value, timestamp: .timestamp}' | ||
``` |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters