Runtime Provisioner 0.8.0 Changelog
0.8.0 - 2026-09-29
-
AXPD-11927 - Breaking:
ksml-provisioneris renamed toruntime-provisioner, along with its chart, its Kubernetes resources, its image and its Go module. Read the “How to Upgrade the KSML Provisioner to Runtime Provisioner 0.8.0” guide in the Axual documentation first. Endpoint paths and values keys are unchanged -
AXPD-11927 - Breaking for monitoring: the alert name, the
service,job,containerandapp.kubernetes.io/namelabels and the OpenTelemetryservice.nameand scope all follow the rename, so dashboards, alert routes and silences stop matching -
AXPD-11927 - Fix the OpenShift Route, which carried no traffic: it asked for a
targetPortthe Service does not have, androute.pathdefaulted to/api, which reaches no handler.route.pathand the Ingress path now default to/ -
AXPD-11927 - A working Route or Ingress publishes an unauthenticated service, so set
BASIC_AUTH_ENABLED=truewith a username and password before you enable either -
AXPD-11927 - Docker mode containers are labelled
axual.io/managed-by=runtime-provisioner, and the bulk/statusmatches that label on its key, so containers left by 0.7.x stay listed and none has to be recreated -
AXPD-12019 - Add
GET /kafkaconnect/logs/download, which streams every worker’s current log file as one plain-text attachment in labelled sections, capped byCONNECT_DOWNLOAD_CAP_BYTES, 64 MiB by default -
AXPD-12019 - Breaking for trace queries: the log download reports its spans under the shared scope
axual.io/otel/runtime-provisionerinstead ofkafkaconnect-download, matching every other handler. The span name is unchanged -
AXPD-12019 - The connector filter matches a log entry on its
connector.contextfield instead of searching the line text, ending false positives such as a filter forsinkmatchingorders-sink -
AXPD-12019 -
GET /kafkaconnect/logsacceptsminLevel, keeping only lines at that severity or above in both modes, with stack trace lines kept alongside the line they belong to -
AXPD-12019 - The connector filter no longer matches a connector whose name is a hyphen-separated suffix of another, so
sinkstops matchingconnector-producer-orders-sink-0 -
AXPD-12019 - A log stream reports when its history reached the start of what the workers still hold, so a short window is told apart from a log that has ended
-
AXPD-11974 - A Kafka Connect log stream sends the history a batch request answers with and then follows the log, from one read per worker, so
tailLinesmeans the same in both modes -
AXPD-11974 - Breaking for API clients:
GET /kafkaconnect/logsanswers withentries, one time-ordered list carryingoriginandtimestampper line, pluslinesRead,linesReturnedanddegraded. The per-workerlogsfield is gone -
AXPD-11974 - A filtered request reads a wider raw window, so
tailLinesbounds the answer and not the read. AddsCONNECT_FILTER_WINDOW_PERCENT(default2000) andCONNECT_FILTER_WINDOW_MAX_LINES(default50000), and the matching chart values -
AXPD-11974 - Breaking for dashboards: the metric
kafkaconnect.log.stream.errorsis renamed tokafkaconnect.log.read.errorsand gains amodeattribute ofbatchorstream -
AXPD-11974 - Log streaming sends
Waiting for output from the application…after 2 seconds of silence, and KSML reports a not-started container as starting instead of failing with a 500 -
AXPD-11856 - Add
GET /kafkaconnect/logs, merging one Kafka Connect cluster’s worker pod logs into a single JSON or NDJSON answer, with theCONNECT_NAMESPACES,CONNECT_CONTAINER_NAMEandMAX_CONCURRENT_LOG_STREAMSsettings -
AXPD-11856 - Add
/healthzand/readyz, serve the KSML endpoints under/ksml/beside the flat paths, and addKSML_ENABLEDto switch KSML support off -
AXPD-11856 - Breaking for the chart: liveness and readiness now use
/healthzand/readyzinstead of a port check, so do not pinimage.tagbelow this chart version or the container restarts in a loop -
AXPD-11856 - KSML log streaming can answer 503 when
MAX_CONCURRENT_LOG_STREAMSis reached, because the cap is shared with Kafka Connect log streaming -
AXPD-11873 - Add
DOCKER_KSML_RUNNER_DEFAULTS_FILEto Docker mode, aksml-runner.yaml-shaped defaults file deep-merged under every generated runner config, giving Docker Compose whatCUSTOM_VALUES_FILEgives Kubernetes mode -
Update dependencies (Helm 3.22.0,
golang.org/x/crypto0.57.0 with a security fix, OpenTelemetry OTLP gRPC exporter 1.46.0, and related modules) -
Update default KSML CHART_VERSION to
1.3.2
0.7.0 - 2026-06-24
-
Add Docker runtime mode (
DEPLOYMENT_TARGET=docker) to deploy KSML applications as Docker containers on the local Docker daemon, intended for local development and testing via Docker Compose -
Introduce
RuntimeServiceinterface abstraction so all REST handlers are runtime-agnostic; Kubernetes and Docker runtimes are both implementations of this interface -
Migrate Docker SDK from
github.com/docker/dockerv27 tomoby/mobyv29 split modules, resolving known CVEs in the previous dependency -
Fix
schemaRegistriesJSON tag casing so deployments sending the conventional lowercase key are accepted correctly -
Fix inverted error-type check in
/metricsthat could misclassify a missing Metrics Server as a missing deployment -
AXPD-11468 - Add restart information for pods on
/statusendpoint -
Update dependencies to latest patch versions (Helm 3.21.2, Kubernetes 0.36.2, golang.org/x/*, oras-go 2.6.1, and related transitive modules)
-
Update default KSML CHART_VERSION to
1.3.0
0.6.1 - 2026-04-08
-
AXPD-11353 - Fix
DNS-1035compliance for hashed KSML names by using{tenant}-{instance}-{hash16}-ksmlformat instead of a plain hash, ensuring the name always starts with an alphabetic character -
Update default KSML CHART_VERSION to
1.2.0
0.6.0 - 2026-03-24
-
AXPD-11200 - Add streaming logs support for deployments
-
AXPD-9774 - Generated KSML Name can exceed k8s max length
0.5.0 - 2025-12-22
-
AXPD-10246 - Add
/metricsendpoint to retrieve CPU and memory usage for KSML deployments. New permissions/roles needed for metrics collection -
AXPD-10689 - Add optional
cleanupparameter in/undeployto delete PVCs -
AXPD-10244 - Add
stateStoresPersistenceconfiguration support in/deploy -
Update default KSML CHART_VERSION to
1.1.1
0.4.0 - 2025-09-22
-
Add configurable schemaRegistries in
/deploy -
Update default KSML CHART_VERSION to
1.1.0 -
Removed
.Capabilities.APIVersionschecks from PrometheusRule and ServiceMonitor templates; resources are now created whenever their enabled values are set