Centrifugo PRO v6.8.3

Centrifugo PRO is an extended version of Centrifugo with a set of additional features. These features can provide your business with unique benefits – drastically save development time, reduce resource usage on a server, protect your backend from misusing, and put the system observability to the next level.

More details and feature description can be found in Centrifugo PRO docs.

What's changed

Improvements

• Everything from Centrifugo OSS v6.8.3
• Even Redis Cluster sharded PUB/SUB distribution — a new use_precomputed_partition_tags Redis broker option spreads sharded PUB/SUB partitions evenly across Redis Cluster nodes using a precomputed table of hash tags. The default bare-index tags ({0}, {1}, ...) hash via CRC16 into clustered slots, so on larger clusters load ends up badly skewed (at 16 nodes / 32 partitions roughly half the nodes receive zero PUB/SUB traffic). When enabled, sharded_pub_sub_partitions must equal one of the supported sizes (16, 32, 64, 128, 256, 512, 1024, 2048, 4096). Note that enabling it changes the Redis key and channel names, so toggling it on a running deployment is a key-namespace migration. See Even partition distribution with precomputed tags.
• OpenTelemetry cloud resource detectors — the opentelemetry.resource_detectors option lists cloud platforms (gcp, aws) whose resource attributes Centrifugo detects from the platform metadata service at startup and attaches to exported metrics and traces (region/zone, instance/host, container/task info, etc.). Values from OTEL_RESOURCE_ATTRIBUTES take precedence over detected ones. This pairs well with the OSS change that reports the node ID as service.instance.id to give autoscaled deployments distinct, correctly-labeled time series — see Cloud resource detectors.