Integrating Flink with Datadog
I am having several issues regarding Flink and Datadog integration. First, the issue is that Datadog uses dogstatsD instead of statsD which is not included in Flink documentation
Another issue is that if you go to Datadog's Integrations page, Flink integration is missing. I have tried installing graphite but I am having several issues with that as well due to python 3.6, I tried virtualenv as well, but thought of going with datadog, which is giving me hard time as well.
Two approaches that may work:
It looks like flink has an HTTP connector to send metrics to Datadog, which at first glance looks to send over the Datadog metrics API instead of dogstatsd.
Dogstatsd is not very different from statsd otherwise, so it's often easy to modify statsd libraries to work for dogstatsd. This project on GitHub seems to be such a project, and may come in handy.
- Unable to run flink jar : Multiple factories for identifier 'default' that implement
- Data Loss in Flink Job with Iceberg Sink After Restart: How to Ensure Consistent Writes?
- Apache Flink Branching
- Setting YARN application name for Flink Jobs
- Flink Kubernetes S3 state support
- Apache Flink and State Store from Kafka Data Source
- java.lang.ClassNotFoundException: org.apache.flink.connector.base.source.reader.RecordEmitter
- Confused about FLINK task slot
- Unit Tests Failing when upgrading Flink to 1.18
- Flink sql pipeline print?
- Flink: java.lang.NoSuchMethodError: AvroSchemaConverter
- Does unaligned checkpointing in Flink have disadvantages beyond extra I/O, and does it affect end-to-end exactly-once semantics?
- Flink-BigTable - Any connector?
- Flink state using RocksDB
- How to configure process function in Apache Flink
- Apache Flink type information error for POJO containing List of Tuples
- Give Flink type information to serialize accumulator object without Kryo
- Flink classLoader cannot be null Kryo
- Cannot stop flink job with savepoint when it is restarting
- JcaPEMKeyConverter is provided by BouncyCastle, an optional dependency. To use support for EC Keys you must explicitly add dependency to classpath
- Apache Flink with multiple Kafka sources. Ensure one topic is fully read before consuming data on the other topic
- Timestamps of result elements of window function in Apache Flink
- Flink State Processor API with Windowed CoGroup and Unaligned Checkpoints
- Flink SQL Hints in Confluent Flink Window Query
- Unable to consume data using the latest Pyflink Kafka connector
- How to handle late events using Flink's table API?
- How to deploy Kinesis Data Analytics without data loss
- Whether Processing Time Temporal Join supports FOR SYSTEM_TIME AS OF syntax or not
- Flink temporal join inconsistencies
- NoClassDefFoundError when running flink with kafka connector