Releases: snowplow/stream-collector
2.6.0
Changelog
- Replace all metrics implementations with statsd (#223)
- Enable AWS MSK IAM Authentication (#232)
- Remove warning about missing config option (#222)
- Inspect X-Forwarded-Proto for http -> https redirects (#221)
- Change docker base image to eclipse-temurin:11-jre-focal (#228)
- Publish distroless docker image (#229)
2.5.0
Changelog
- Use shorter app name (#217)
- Update copyright notices to 2022 (#216)
- Disable warnings for illegal headers (#178)
- Change default telemetry url (#209)
- Add configuration option for allow list for collector redirects (#131)
- Increase default value of pubsub backoffPolicy totalBackoff (#187)
- Move logging of thread pool creation out of KinesisSink (#129)
- CollectorServiceSpec should test number of events written to sink (#188)
- Improve graceful shutdown (#12)
2.4.5
2.4.4
2.4.3
2.4.2
2.4.1
Stream Collector 2.4.1 released
Quickfix release
Changes
OSS terraform modules unique id not propagated into telemetry event (#182)
Discourse
2.4.0
Stream Collector 2.4.0 released
This release moves the SSL configuration from the config file to the JVM system properties.
All components were given a minimal configuration file in examples.
Telemetry data collection is enabled. Only application name and version are send every hour by default.
Changes
Make it possible to configure the collector without a file (#173)
Add telemetry (#167)
Handle LimitExceededException when testing if stream exists (#174)
Include aws-java-sdk-sts to enable web token authentication (#169)
Use sbt-dynver to set application version (#166)
Publish arm64 and amd64 docker images (#165)
Change docker base image to adoptopenjdk:11-jre-hotspot-focal (#164)
Use JRE defaults for https configuration (#163)
Bump akka-http to 10.2.6 (#162)
Bump akka to 2.6.16 (#161)
Discourse
2.3.1
This is a maintenance release that upgrades some dependencies to newer versions. And one small fix to help aws users in the cn-northwest-1 region.
2.3.0
This release adds a new health endpoint that can be monitored via GET requests to the /sink-health path. These requests will receive a response of 200 if no issues are detected with the sink to which the collector is trying to write, and 503 otherwise.
The new sink health endpoint is meant to serve as an alarm trigger for cloud outages; but it will always return 200 for non-cloud sinks, such as nsq, kafka and stdout.
Changes
Add outage protection features to Kinesis, SQS and Pubsub sinks (#132)
Pass Kinesis partitioning key as SQS message attribute (#146)
Fix typo in PubSub sink useragent string (#147)
Use base image from DockerHub (#107)
Attach jar files to Github releases (#108)
Remove Bintray from resolutionRepos (#144)
Bump scopt to 4.0.1 (#143)
Bump pureconfig to 0.15.0 (#142)
Bump jackson-databind to 2.9.10.8 (#141)
Bump json4s-jackson to 3.6.11 (#140)
Bump specs2-core to 4.11.0 (#139)
Bump sbt-scalafmt to 2.4.2 (#138)
Bump sbt-tpolecat to 0.1.17 (#137)
Bump sbt-buildinfo to 0.10.0 (#136)
Bump sbt-assembly to 0.15.0 (#135)
Bump sbt to 1.5.1 (#134)
Add link to Snowplow's public roadmap in README (#145)