feat: virtual thread support by csviri · Pull Request #3664 · operator-framework/java-operator-sdk

csviri · 2026-10-27T13:21:33Z

feat: virtual thread support

Adds a single flag that switches the framework's concurrent work over to virtual threads:

Operator operator = new Operator(override -> override.withUseVirtualThreads(true));

It can also be set with josdk.use-virtual-threads. It is off by default, so nothing changes for
existing operators.

Why

Reconcilers spend almost all their time blocked on the Kubernetes API server or on external
systems. Virtual threads make that kind of blocking cheap, so concurrency limits can be raised far
above what is reasonable with platform threads, without sizing and tuning thread pools.

What changes when the flag is on

Area	Behavior
Reconciliations	Run on virtual threads. `withConcurrentReconciliationThreads(n)` still caps how many run at once (enforced with a `Semaphore`).
Dependent workflows	Same: virtual threads, still capped by `withConcurrentWorkflowExecutorThreads(n)`.
Internal housekeeping	The unbounded caching executor (starting informers, etc.) uses a virtual thread per task.
Scheduled tasks	Polling event sources and retried or rescheduled reconciliations run on virtual threads. See the next section.
Kubernetes client	The default client created by the framework runs its internal task executor (informer event dispatch, watch event delivery) on virtual threads. A client provided by the user is used as is.
JVM lifetime	Virtual threads are always daemon threads, so they don't keep the JVM alive. If `main` returns after starting the operator and nothing else keeps the JVM running, call `Operator.keepAlive()` (already available, see the docs). The framework does not start such a thread on its own.

Java 25 only

Virtual threads are supported on Java 25 and later (JEP 491
removed the carrier pinning inside synchronized blocks, which the codebase uses extensively; on
Java 21 to 24 it can starve the carrier pool and deadlock startup). This is checked at runtime: on
an older JVM a warning is logged and platform threads are used, so one configuration works on every
Java version the operator runs on. The compilation target stays Java 17.

Scheduled executors (the larger part of this PR)

Previously PollingEventSource and TimerEventSource each created a java.util.Timer, and
PerResourcePollingEventSource created its own ScheduledThreadPoolExecutor. Each cost a thread
that the operator neither sized nor shut down. They now use scheduled executors that are created
and shut down in a defined way:

Polling event sources get an executor of their own. PollingEventSource and
PerResourcePollingEventSource do not share an executor, so a slow poll of one source never
delays another. Unless one is provided with withExecutorService on the configuration builder,
the source creates a single-thread daemon executor when it starts and shuts it down when it stops
(a new one on every start). With useVirtualThreads on, that executor runs the polls on virtual
threads. A provided executor is used as is, is never shut down by the source, and may be shared
deliberately.
ConfigurationService.getRetryAndRescheduleExecutorService() backs the retry and reschedule
timer of every controller. It is managed by ExecutorServiceManager, sized with
concurrentRetryAndRescheduleThreads() (default 2) or
josdk.retry-and-reschedule.concurrent-threads, and can be replaced through
ConfigurationServiceOverrider. The operator shuts it down on stop and creates a new one when
started again.
The default pools use daemon threads, drop tasks scheduled for later on shutdown (so far-out
reschedules don't delay termination), remove cancelled tasks eagerly, and let idle threads time
out after 60 seconds.
Event sources resolve their executor on every start() rather than at construction. The polling
and timer sources let a poll or handler already in flight finish when stopped, as the Timer
they replace did. Note that the timer source does not schedule again the timers that were pending
on an executor the operator replaced (which only happens when the operator is started again after
a failed start), they are lost with the old executor.
ExecutorServiceManager.scheduledExecutorService() stays available for compatibility,
deprecated, as a general purpose executor created on first use. The framework itself no longer
uses it.

`VirtualThreadScheduledExecutor`

The JDK has no scheduled executor backed by virtual threads, so with the flag on the defaults are a
new package-private VirtualThreadScheduledExecutor: one daemon platform thread does the timing
and hands each fired task to a virtual-thread-per-task executor.

The number of scheduled tasks running at once is not limited in this mode, because they are
typically blocking calls. withConcurrentRetryAndRescheduleThreads only applies to platform
threads.
It keeps the semantics of the default platform executor: waiting and periodic tasks are discarded
on shutdown(), and tasks that have already started are let finish. shutdownNow() interrupts
running tasks.
Fixed-rate tasks make up missed executions after an overrun, as ScheduledThreadPoolExecutor
does. Fixed-delay tasks re-arm from completion time.

Other changes

Executor helpers live in an internal ExecutorUtilsInternal.
ConfigLoader and ConfigurationServiceOverrider expose the new settings
(josdk.use-virtual-threads, josdk.retry-and-reschedule.concurrent-threads).
The webpage and mysql-schema sample operators now run with virtual threads.
spotbugs exclusion for the VirtualThreadScheduledExecutor handles. They are ordered by remaining
delay but must stay distinct by identity, as the JDK's own scheduled futures.

Documentation

operations/configuration.md: new "Virtual Threads" section, which points to "Keeping the JVM
Alive" for operators whose main returns, and the new configuration keys.
eventing.md: "Threading of the polling event sources", with the per-source executor and who
owns it.
migration/v5-7-migration.md: the changes to the executors of the polling and timer event
sources, notably that an executor provided to a PerResourcePollingEventSource is not shut down
by it anymore, plus the new deprecations.

Testing

Unit tests, with the virtual thread ones gated by @EnabledForJreRange(min = JAVA_25):

VirtualThreadScheduledExecutorTest: virtual thread execution, cancellation, fixed delay and
fixed rate, overrun catch-up, shutdown semantics, unbounded concurrency.
VirtualThreadsTest: bounded virtual thread executor and fallback on old JVMs.
VirtualThreadsStartupDeadlockTest: startup with virtual threads does not deadlock.
ExecutorServiceManagerTest, ConfigurationServiceOverriderTest: lifecycle and overriding.
PerResourcePollingEventSourceTest, PollingEventSourceTest, TimerEventSourceTest: each source
owning its executor (and not sharing it), provided executors left running, virtual threads,
restart, stop and start ordering, polling resilience.
ExecutorUtilsInternalTest: idle pool threads time out and delayed tasks still run.

Integration test: VirtualThreadsIT runs a full operator with the flag on.

Known limitations and follow-ups

Only Java 25 and later is officially supported for virtual threads.

coderabbitai · 2026-10-27T13:21:41Z

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

Run configuration

Source mode: Auto
Repository: Unavailable
Branch: Unavailable
File: No configuration file from this repository applied
Commit: Unavailable
Layers: defaults
Review profile: CHILL
Plan: Advanced
Run ID: 21d39387-3725-4387-b833-afd6b479d802

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

Trigger review

Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Tip

🎯 Customize your risk assessments

Guide how CodeRabbit assesses merge risk and security architecture risk. Highlight sensitive changes and explain your security boundaries using Merge Risk Instructions and Security Risk Instructions in your review settings. Security architecture risk assessments require the Advanced plan or above.

✅ Automatic approvals, with your rules

Let CodeRabbit approve pull requests once required threads are resolved, pre-merge checks pass, and your risk limits are met. Enable Automatic approvals in review settings, with overrides for teams or file paths.

Copilot

🟡 Changes recommended

Unresolved executor lifecycle, polling startup, and event-validation issues can lose work or prevent events from being recorded.

4 open issues

What changed in this PR

Adds opt-in virtual threads to Java Operator SDK while preserving concurrency limits and Java 17 fallback. The PR also centralizes scheduling and includes event-recording and desired-state fixes.

Changes:

Adds virtual-thread configuration, executor implementations, client integration, documentation, and tests.
Moves polling and retry scheduling to separate configurable executors with lifecycle management.
Updates event aggregation, long event names, and desired-state aspect handling.

File	Description
sample-operators/webpage/src/main/java/io/javaoperatorsdk/operator/sample/WebPageOperator.java	Enables virtual threads in the sample.
operator-framework/src/test/java/io/javaoperatorsdk/operator/dependent/multiplemanagedexternaldependenttype/MultipleManagedExternalDependentResourceReconciler.java	Supplies context for shared polling.
operator-framework/src/test/java/io/javaoperatorsdk/operator/config/loader/ConfigLoaderTest.java	Tests new configuration keys.
operator-framework/src/test/java/io/javaoperatorsdk/operator/baseapi/virtualthreads/VirtualThreadsTestReconciler.java	Records thread type and concurrency.
operator-framework/src/test/java/io/javaoperatorsdk/operator/baseapi/virtualthreads/VirtualThreadsIT.java	Tests reconciliation threading and limits.
operator-framework/src/test/java/io/javaoperatorsdk/operator/baseapi/virtualthreads/VirtualThreadsCustomResource.java	Provides the integration-test resource.
operator-framework/src/main/java/io/javaoperatorsdk/operator/config/loader/ConfigLoader.java	Binds virtual-thread and scheduling settings.
operator-framework-core/src/test/java/io/javaoperatorsdk/operator/processing/event/source/timer/TimerEventSourceTest.java	Tests executor ownership and restart.
operator-framework-core/src/test/java/io/javaoperatorsdk/operator/processing/event/source/polling/PollingEventSourceTest.java	Tests shared scheduling and cancellation.
operator-framework-core/src/test/java/io/javaoperatorsdk/operator/processing/event/source/polling/PerResourcePollingEventSourceTest.java	Tests polling restart behavior.
operator-framework-core/src/test/java/io/javaoperatorsdk/operator/processing/dependent/DesiredStateAspectAuditTest.java	Adds desired-state aspect regression cases.
operator-framework-core/src/test/java/io/javaoperatorsdk/operator/api/event/EventRecorderAuditTest.java	Adds event-recording regression cases.
operator-framework-core/src/test/java/io/javaoperatorsdk/operator/api/event/DefaultEventRecorderTest.java	Tests collision-resistant name shortening.
operator-framework-core/src/test/java/io/javaoperatorsdk/operator/api/config/VirtualThreadsTest.java	Tests executor and client threading.
operator-framework-core/src/test/java/io/javaoperatorsdk/operator/api/config/ExecutorServiceManagerTest.java	Tests scheduler separation and lifecycle.
operator-framework-core/src/test/java/io/javaoperatorsdk/operator/api/config/ConfigurationServiceOverriderTest.java	Tests threading configuration overrides.
operator-framework-core/src/test/java/io/javaoperatorsdk/operator/api/config/BoundedVirtualThreadExecutorAuditTest.java	Tests limits and pending-task shutdown.
operator-framework-core/src/main/java/io/javaoperatorsdk/operator/processing/event/source/timer/TimerEventSource.java	Replaces timers with supplied or owned executors.
operator-framework-core/src/main/java/io/javaoperatorsdk/operator/processing/event/source/polling/PollingEventSource.java	Adds shared scheduling and executor ownership.
operator-framework-core/src/main/java/io/javaoperatorsdk/operator/processing/event/source/polling/PollingConfigurationBuilder.java	Adds custom polling executor configuration.
operator-framework-core/src/main/java/io/javaoperatorsdk/operator/processing/event/source/polling/PollingConfiguration.java	Stores the optional polling executor.
operator-framework-core/src/main/java/io/javaoperatorsdk/operator/processing/event/source/polling/PerResourcePollingEventSource.java	Resolves executors on startup and cancels tasks.
operator-framework-core/src/main/java/io/javaoperatorsdk/operator/processing/event/source/polling/PerResourcePollingConfiguration.java	Defers default executor selection.
operator-framework-core/src/main/java/io/javaoperatorsdk/operator/processing/event/EventSources.java	Accepts a retry scheduler supplier.
operator-framework-core/src/main/java/io/javaoperatorsdk/operator/processing/event/EventSourceManager.java	Connects retries to the managed scheduler.
operator-framework-core/src/main/java/io/javaoperatorsdk/operator/processing/dependent/external/PollingDependentResource.java	Passes context to polling sources.
operator-framework-core/src/main/java/io/javaoperatorsdk/operator/processing/dependent/BulkDependentResourceReconciler.java	Passes the user dependent to aspects.
operator-framework-core/src/main/java/io/javaoperatorsdk/operator/api/reconciler/DefaultContext.java	Handles reentrant desired-state aspect lookups.
operator-framework-core/src/main/java/io/javaoperatorsdk/operator/api/event/DefaultEventSink.java	Updates fields during named-event aggregation.
operator-framework-core/src/main/java/io/javaoperatorsdk/operator/api/event/DefaultEventRecorder.java	Adds event timestamps and hashed name suffixes.
operator-framework-core/src/main/java/io/javaoperatorsdk/operator/api/config/VirtualThreads.java	Implements virtual executors and concurrency bounds.
operator-framework-core/src/main/java/io/javaoperatorsdk/operator/api/config/Utils.java	Adds named daemon scheduler utilities.
operator-framework-core/src/main/java/io/javaoperatorsdk/operator/api/config/ExecutorServiceManager.java	Selects executors and manages scheduler lifecycle.
operator-framework-core/src/main/java/io/javaoperatorsdk/operator/api/config/ConfigurationServiceOverrider.java	Exposes threading and scheduler overrides.
operator-framework-core/src/main/java/io/javaoperatorsdk/operator/api/config/ConfigurationService.java	Defines threading defaults and client integration.
docs/content/en/docs/documentation/operations/configuration.md	Documents virtual threads and configuration keys.
docs/content/en/docs/documentation/eventing.md	Documents shared polling and executor ownership.

_{Review effort: Balanced}

Copilot

🟡 Changes recommended

Unresolved scheduler contract violations and cancellation/lifecycle races can skip fixed-rate executions, run cancelled work, and leave duplicate or stale polling chains.

7 open issues

4 resolved since last review

_{Review effort: Balanced}

Copilot

Copilot encountered an error and was unable to review this pull request. You can try again by re-requesting a review.

Copilot

🟡 Changes recommended

Unresolved lifecycle races can resurrect or duplicate polling tasks, while premature non-daemon thread startup can prevent process exit after initialization failure.

14 open issues

_{Review effort: Balanced}

Copilot

🔵 Needs a closer look

Critical polling lifecycle races can schedule deleted resources, prevent polling after recreation, and allow stale fetches to create duplicate polling chains after restart.

14 open issues

_{Review effort: Balanced}

Copilot

🟡 Changes recommended

Operator construction prematurely starts a non-daemon keep-alive thread, preventing JVM exit if the operator is never started or reconciler registration fails.

7 open issues

8 resolved since last review

_{Review effort: Balanced}

Copilot

🟡 Changes recommended

Unresolved polling lifecycle races can schedule deleted resources, prevent polling after resource recreation, and create duplicate polling chains after restart.

7 open issues

1 resolved since last review

Operator.init() obtains this manager, whose constructor calls start(), so merely constructing…

_{Review effort: Balanced}

…s an override changes their settings Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

…ecutors when started again while running - a per resource registration that races stop() is buffered until the next start instead of leaving a stale entry that blocks polling after a restart - the timer and per resource polling event sources resolve their executor again when started while running, which happens when the operator is started again after a failed start and replaces its executors Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

…s, and let in-flight work finish when stopping owned executors - PollingEventSource schedules its polling task again when started while running after the operator replaced its executors - TimerEventSource schedules the timers pending on the replaced executor again for their remaining delay - stopping a polling or timer event source that owns its executor no longer interrupts a poll or handler in flight Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

Signed-off-by: Attila Mészáros <a_meszaros@apple.com>

…se races in the polling event sources - ExecutorServiceManager keeps the public scheduledExecutorService() of previous versions, deprecated, created on first use and replaced after a restart - PerResourcePollingEventSource drops the result of a fetch from a previous run instead of overwriting the cache of the current one - PollingEventSource ends start() cleanly when stopped during the initial fetch instead of failing on the shut down executor it owns Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

…ck in the fetching task Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

…g the JVM alive Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

…when the executor is replaced Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

…tor supplier Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

…stom retry executor Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

…utor and drop the executor supplier only constructor of TimerEventSource - the timers pending on an executor the operator replaced are not scheduled again - the constructor taking just an executor supplier is removed, the event source is mostly for internal usage Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

…source polling starts instead of buffering registrations - the registrations seen before start, or racing stop, are dropped, start registers every resource of the primary cache - the registration racing the end of start is not handled anymore Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

…ll found it gone from the cache - a poll that finds its primary resource gone from the cache removes its registration, so a recreated resource is registered again instead of being skipped for the stale one - onResourceDeleted is synchronized with the registration and scheduling of the polls Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

… test name - remove BoundedVirtualThreadExecutorAuditTest and the AUDIT_N override of VirtualThreadsStartupDeadlockTest - describe in the Javadoc of VirtualThreadsStartupDeadlockTest what is guaranteed on which Java version - rename the test checking that executeAndWaitForAllToComplete respects the bound of the executor Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

…ges in the migration guide - the example of an executor with more threads uses PerResourcePollingConfigurationBuilder, a PollingEventSource runs a single periodic task - the migration guide describes the executors of the polling event sources, the provided executor that is not shut down anymore and the deprecations Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

Copilot

🟡 Changes recommended

Separately sampling remaining delays can violate the Comparable contract for handles with similar deadlines, potentially misordering futures in sorted collections or delay queues.

2 open issues

_{Review effort: Balanced}

…ng their remaining delays Sampling the remaining delay of each future separately could make a.compareTo(b) and b.compareTo(a) disagree for close deadlines, violating the Comparable contract. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

Copilot

🔵 Needs a closer look

Human review is needed to assess concurrency and lifecycle risks involving untracked polling tasks, executor recreation after shutdown, and repeated cancellation that can unexpectedly interrupt running work.

1 open issue

Align operator keep-alive documentation with implementation

1 resolved since last review

Compare Handle deadlines directly to preserve Comparable ordering

_{Review effort: Balanced}

…#3665) Correctness fixes of the event recorder, found while reviewing the virtual thread support (#3664). - Aggregating onto a named event now updates its type, reason and action along with the count, last timestamp and message, so a Warning repeated under the name of a Normal event is no longer reported as the Normal one. An occurrence without an action clears the earlier action, except on an event that carries an eventTime, where the reason stands in for it, as the API server requires an action for such events. The API server only makes these fields immutable for events.k8s.io/v1, not for the core API the sink writes to. - Supplied event names longer than the Kubernetes limit are shortened with a hash of the full name instead of being plainly truncated, so distinct names sharing a long prefix no longer merge into one event. Trailing '-' and '.' are stripped before the hash is appended, for supplied and generated names alike, so a name is never cut into an invalid "...-.<hash>". - Events about cluster scoped resources in the kube-system namespace carry an eventTime and an action (defaulting to the reason), which is what the API server requires to accept them there. Events in the default namespace are unchanged. The API server rejects such events in any other namespace, so the javadocs and docs no longer suggest the namespace the operator runs in. Tests: regression and unit tests, a kube-api-test based DefaultEventRecorderIntegrationTest recording events against a real API server, and ITs for cluster scoped resources in default and kube-system.

Add the useVirtualThreads flag (ConfigurationServiceOverrider. withUseVirtualThreads, josdk.use-virtual-threads) that runs the framework's concurrent work on virtual threads. It is off by default and only supported on Java 25 or later (JEP 491); on an older JVM a warning is logged and platform threads are used. With the flag on: - reconciliations and dependent workflows run on virtual threads, still capped by concurrentReconciliationThreads and concurrentWorkflowExecutorThreads - the caching executor used for internal housekeeping uses a virtual thread per task - the Kubernetes client created by the framework runs its internal task executor on virtual threads, a provided client is used as is - scheduled tasks run on virtual threads through the new VirtualThreadScheduledExecutor, a daemon timing thread firing each task on a virtual thread, with the semantics of the platform default (waiting and periodic tasks discarded on shutdown, running tasks let finish, fixed rate tasks catch up after an overrun) Scheduled work no longer relies on java.util.Timer: - PollingEventSource and PerResourcePollingEventSource poll on an executor of their own, provided through the configuration builder or created, and shut down, by the event source, on virtual threads when the flag is on. A provided executor is never shut down by the source. They can be started again after a stop, and a poll in flight is not interrupted by it. - PerResourcePollingEventSource registers the resources of the primary cache when it starts, drops the result of a poll of a previous run, and forgets the registration of a resource that is gone from the cache so that it is polled again if recreated - TimerEventSource schedules on the retry and reschedule executor managed by ExecutorServiceManager, see ConfigurationService.getRetryAndRescheduleExecutorService and concurrentRetryAndRescheduleThreads. Timers pending on an executor the operator replaced are not scheduled again. - the default pools use daemon threads, discard delayed tasks on shutdown and let idle threads time out Behavior and API changes, described in the v5.7 migration guide: - an executor provided to PerResourcePollingConfigurationBuilder is not shut down by the event source anymore, and the configuration returns a null executor when none was provided - PollingEventSource(Class, PollingConfiguration), the TimerEventSource(String, boolean) constructor and ExecutorServiceManager.scheduledExecutorService() are deprecated - ConfigurationServiceOverrider rebuilds the default retry executor when the flag or its thread count changes, a custom executor of the wrapped configuration is then replaced, see the Javadoc Also: configuration keys in ConfigLoader, documentation (Virtual Threads, polling event source threading, migration guide), the webpage and mysql-schema samples run with virtual threads, VirtualThreadsIT, and a spotbugs exclusion for the scheduled executor handles. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

…#3665) Correctness fixes of the event recorder, found while reviewing the virtual thread support (#3664). - Aggregating onto a named event now updates its type, reason and action along with the count, last timestamp and message, so a Warning repeated under the name of a Normal event is no longer reported as the Normal one. An occurrence without an action clears the earlier action, except on an event that carries an eventTime, where the reason stands in for it, as the API server requires an action for such events. The API server only makes these fields immutable for events.k8s.io/v1, not for the core API the sink writes to. - Supplied event names longer than the Kubernetes limit are shortened with a hash of the full name instead of being plainly truncated, so distinct names sharing a long prefix no longer merge into one event. Trailing '-' and '.' are stripped before the hash is appended, for supplied and generated names alike, so a name is never cut into an invalid "...-.<hash>". - Events about cluster scoped resources in the kube-system namespace carry an eventTime and an action (defaulting to the reason), which is what the API server requires to accept them there. Events in the default namespace are unchanged. The API server rejects such events in any other namespace, so the javadocs and docs no longer suggest the namespace the operator runs in. Tests: regression and unit tests, a kube-api-test based DefaultEventRecorderIntegrationTest recording events against a real API server, and ITs for cluster scoped resources in default and kube-system.

Copilot AI balanced review requested due to automatic review settings October 27, 2026 13:21

openshift-ci Bot requested review from metacosm and xstefank October 27, 2026 13:21

This was referenced Oct 27, 2026

feat: virtual thread support #3603

Closed

feat: schedule polling and retries on managed executors instead of Timers #3620

Closed

Copilot started reviewing on behalf of csviri October 27, 2026 13:26

View session

Copilot AI reviewed Oct 27, 2026

View reviewed changes

csviri linked an issue Oct 27, 2026 that may be closed by this pull request

Support for Virtual Threads #3538

Closed

csviri requested a balanced review from Copilot October 28, 2026 10:00

Copilot started reviewing on behalf of csviri October 28, 2026 10:01

View session

Copilot AI reviewed Oct 28, 2026

View reviewed changes

csviri requested a balanced review from Copilot October 28, 2026 13:57

Copilot started reviewing on behalf of csviri October 28, 2026 13:57

View session

Copilot AI reviewed Oct 28, 2026

csviri requested a balanced review from Copilot October 28, 2026 14:20

Copilot started reviewing on behalf of csviri October 28, 2026 14:20

View session

Copilot AI reviewed Oct 28, 2026

View reviewed changes

csviri requested a balanced review from Copilot October 29, 2026 08:22

Copilot started reviewing on behalf of csviri October 29, 2026 08:23

View session

Copilot AI reviewed Oct 29, 2026

View reviewed changes

csviri requested a balanced review from Copilot October 29, 2026 11:18

Copilot started reviewing on behalf of csviri October 29, 2026 11:18

View session

Copilot AI reviewed Oct 29, 2026

View reviewed changes

Comment thread ...core/src/main/java/io/javaoperatorsdk/operator/api/config/ConfigurationServiceOverrider.java Resolved

csviri requested a balanced review from Copilot October 29, 2026 15:08

Copilot started reviewing on behalf of csviri October 29, 2026 15:09

View session

Copilot AI reviewed Oct 29, 2026

View reviewed changes

Comment thread ...mework-core/src/main/java/io/javaoperatorsdk/operator/api/config/ExecutorServiceManager.java Resolved

csviri requested a balanced review from Copilot October 30, 2026 08:15

Copilot started reviewing on behalf of csviri October 30, 2026 08:15

View session

csviri and others added 13 commits November 4, 2026 14:01

fix: keep the scheduled executors of the original configuration unles…

0ad0137

…s an override changes their settings Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

docs: correct the since version of the scheduled executor API to 5.7.0

b47c719

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

update how polling event sources are handled

5560043

Signed-off-by: Attila Mészáros <a_meszaros@apple.com>

fix: synchronize on the outer class method instead of exposing the lo…

6436309

…ck in the fetching task Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

docs: point the virtual threads note to Operator.keepAlive for keepin…

81ed9ae

…g the JVM alive Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

fix: do not schedule again the timers whose handler is still running …

d57d8f2

…when the executor is replaced Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

refactor: deprecate the TimerEventSource constructor without an execu…

bb1fd8a

…tor supplier Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

docs: document that overriding the virtual threads flag replaces a cu…

c061f09

…stom retry executor Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

csviri force-pushed the virtual-threads-and-shared-executors branch from c1a98c7 to 7939ed2 Compare November 4, 2026 22:01

csviri and others added 4 commits November 4, 2026 14:45

fix: set the root log level of the webpage sample back to info

83d9e11

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

csviri requested a balanced review from Copilot November 4, 2026 23:04

Copilot started reviewing on behalf of csviri November 4, 2026 23:05

View session

Copilot AI reviewed Nov 4, 2026

View reviewed changes

Comment thread ...ore/src/main/java/io/javaoperatorsdk/operator/api/config/VirtualThreadScheduledExecutor.java Resolved

csviri requested a balanced review from Copilot November 4, 2026 23:26

Copilot started reviewing on behalf of csviri November 4, 2026 23:26

View session

Copilot AI reviewed Nov 4, 2026

View reviewed changes

csviri merged commit 07fb4e3 into next Nov 5, 2026
30 checks passed

csviri deleted the virtual-threads-and-shared-executors branch November 5, 2026 00:07

Conversation

csviri commented Oct 27, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

feat: virtual thread support

Why

What changes when the flag is on

Java 25 only

Scheduled executors (the larger part of this PR)

VirtualThreadScheduledExecutor

Other changes

Documentation

Testing

Known limitations and follow-ups

Uh oh!

coderabbitai Bot commented Oct 27, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Review skipped

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

🟡 Changes recommended

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

🟡 Changes recommended

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

🟡 Changes recommended

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

🔵 Needs a closer look

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

🟡 Changes recommended

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

🟡 Changes recommended

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

🟡 Changes recommended

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

🔵 Needs a closer look

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

csviri commented Oct 27, 2026 •

edited

Loading

`VirtualThreadScheduledExecutor`

coderabbitai Bot commented Oct 27, 2026 •

edited

Loading