release-notes/3.8.9.md
RabbitMQ 3.8.9 is a maintenance release.
It focuses on bug fixes and usability improvements.
This release no longer supports Erlang 21.3.
This release is compatible with Erlang 23.
This release requires Erlang/OTP 22.0 or later.
22.3 or 23.0 releases are recommended.
Provisioning Latest Erlang Releases explains
what package repositories and tools can be used to provision latest patch versions of Erlang 22.3.x.
See 3.8.0 release notes upgrade and
compatibility notes first if upgrading from an earlier release series (e.g. 3.7.x).
See the Upgrading guide for general documentation on upgrades and RabbitMQ change log for release notes of other releases.
When upgrading to this release from 3.7.15 or an older version, extra care has to be taken.
Because older RabbitMQ CLI tools can be incompatible with Erlang 22+ releases, RabbitMQ must be upgraded at the same time as Erlang.
Alternatively the node can be upgraded to 3.7.18 on Erlang 21.3 first,
then Erlang 22.x or 23.x, then RabbitMQ to most recent 3.8.x release.
Any questions about this release, upgrades or RabbitMQ in general are welcome on the RabbitMQ mailing list.
Starting with 3.8.0, an unintentionally introduced assertion could prevent classic mirrored queue
mirrors from starting successfully in case the primary replica changed concurrently with their startup.
GitHub issue: rabbitmq/rabbitmq-server#2449
Raft implementation's failure detector default polling interval has been increased from 1s to 5s. The previously used default results in too frequent leader elections in networks with high packet loss (say, double digit percent).
This has no correctness or safety ramifications, however, automated tests that assert on Raft properties or expect a very fast leader re-election may have to set it back so that peer failures are detected quicker.
GitHub issue: rabbitmq/rabbitmq-server#2450
The now deprecated intrusive health check endpoint, GET /api/healthchecks/node, reported failures in the response
but always set the status to 200 OK. It now uses the status of 500 in case of a failure.
RabbitMQ Monitoring guide explains why this health check is no longer recommended and what the alternatives are.
GitHub issue: rabbitmq/rabbitmq-management#840
When a client doubly encodes a request into JSON, the plugin tries to detect that and will produce a more helpful 400 response instead of a 500.
GitHub issue: rabbitmq/rabbitmq-management#839
When a classic mirrored queue with a replica on every cluster node (this is highly discouraged) was federated, multiple links could be started by the plugin, potentially making message transfers concurrent and affecting ordering.
GitHub issue: rabbitmq/rabbitmq-federation#114
It is now possible to federate exchanges in different virtual hosts in the same cluster for more than one hop. For example, federate messages from an exchange in virtual host B to C, while B itself is federated from virtual host A. Previously such messages would have been filtered by the message cycle detection algorithm which assumed that exchanges are only federated across clusters.
GitHub issue: rabbitmq/rabbitmq-federation#116
Kubernetes API events registered by this plugin mixed up reason and message fields.
Contributed by Gabriele Santomaggio.
GitHub issue: rabbitmq/rabbitmq-peer-discovery-k8s#70
Improved compatibility with Azure AD. This plugin can now be configured to fetch scopes from a non-standard
field in the decoded token value (other than the standard scopes field).
Contributed by @Robert-Gustafsson.
GitHub issue: rabbitmq/rabbitmq-auth-backend-oauth2#53
Warning: The source code archive provided by GitHub only contains the source of the broker, not the plugins or the client libraries.
Please download the archive named rabbitmq-server-3.8.9.tar.xz.