Issues

New issue

Synchronous and asynchronous compatibility within Prefect
#15008 · cicdw opened on Aug 19, 2024
4
Add pending-claim self-healing for flow runs stuck in PENDING
#22664 · desertaxle opened on Jul 29, 2026

MattermostWebhook silently ignores the configured botname
PrefectHQ/prefect#23282 · opened on Oct 5, 2026
Client crashes instead of retrying when Retry-After is an HTTP-date (or any non-numeric value)
PrefectHQ/prefect#23276 · opened on Oct 3, 2026
Task parameters that are dataclasses with init=False fields fail to resolve
PrefectHQ/prefect#23269 · opened on Oct 2, 2026
Concurrency limits page only fetches 200 limits
PrefectHQ/prefect#23267 · opened on Oct 2, 2026
Preferred way to cite prefect
PrefectHQ/prefect#23248 · opened on Sep 30, 2026
2
SIGTERM in reschedule mode un-cancels a flow run the server already marked Cancelled
PrefectHQ/prefect#23243 · opened on Sep 30, 2026
Worker healthcheck reports unhealthy during Prefect Cloud maintenance, so liveness probes restart workers for the whole window
PrefectHQ/prefect#23234 · opened on Sep 29, 2026
Add support for Armada Multi-Cluster Batch System as a worker pool type
PrefectHQ/prefect#23233 · opened on Sep 29, 2026
[UI] Allow sorting task runs by duration
PrefectHQ/prefect#23229 · opened on Sep 29, 2026
1
Subtracting an input from a configured cache policy drops key_storage / isolation_level / lock_manager
PrefectHQ/prefect#23223 · opened on Sep 28, 2026
prefect-kubernetes observer: _get_k8s_jobs raises UnboundLocalError when the Kubernetes client can't be created
PrefectHQ/prefect#23219 · opened on Sep 27, 2026
Automation edit form crashes with React error #130 for Proactive triggers using after + wildcard expect + for_each
PrefectHQ/prefect#23217 · opened on Sep 26, 2026
Daily IntervalSchedule anchored at 09:00 America/New_York returns 08:00 runs when get_dates starts after a DST change
PrefectHQ/prefect#23204 · opened on Sep 24, 2026
1
SQLAlchemy 2.1.0 breaks deployment scheduling on fresh installs
PrefectHQ/prefect#23199 · opened on Sep 24, 2026
1
Event persister drops a whole batch when one event contains a null byte (PostgreSQL)
PrefectHQ/prefect#23197 · opened on Sep 24, 2026
2
Workers are not honoring deployment concurrency limits during poll time
PrefectHQ/prefect#23196 · opened on Sep 24, 2026
1
[UI] Open flow run on click in flow runs timeline
PrefectHQ/prefect#23194 · opened on Sep 24, 2026
1
[UI] Improve flow runs timeline x-ticks
PrefectHQ/prefect#23193 · opened on Sep 24, 2026
Nested schema required list leaks onto sibling properties, so whether None validates depends on property order
PrefectHQ/prefect#23189 · opened on Sep 24, 2026
Releasing a concurrency slot doesn't wake waiters in the same process, so the slot sits idle until their Retry-After expires
PrefectHQ/prefect#23183 · opened on Sep 23, 2026
Cached task runs wait for a tag concurrency slot before the cache lookup
PrefectHQ/prefect#23182 · opened on Sep 23, 2026
Mapped children not yet started leave no task run when PrefectFutureList.result() raises the first failure
PrefectHQ/prefect#23138 · opened on Sep 17, 2026
2
Flow ends COMPLETED when it returns a list of .map() results containing a FAILED task (states nested one level are ignored)
PrefectHQ/prefect#23137 · opened on Sep 17, 2026
mark_deployments_ready updates every deployment on a queue on every worker poll
PrefectHQ/prefect#23129 · opened on Sep 15, 2026
Allow List vs Deny List for ACLs
PrefectHQ/prefect#23119 · opened on Sep 14, 2026

Navigation Menu

Issues

Issues

MattermostWebhook silently ignores the configured botname

Client crashes instead of retrying when Retry-After is an HTTP-date (or any non-numeric value)

Task parameters that are dataclasses with init=False fields fail to resolve

Concurrency limits page only fetches 200 limits

Preferred way to cite prefect

SIGTERM in `reschedule` mode un-cancels a flow run the server already marked Cancelled

Worker healthcheck reports unhealthy during Prefect Cloud maintenance, so liveness probes restart workers for the whole window

Add support for Armada Multi-Cluster Batch System as a worker pool type

[UI] Allow sorting task runs by duration

Subtracting an input from a configured cache policy drops key_storage / isolation_level / lock_manager

prefect-kubernetes observer: _get_k8s_jobs raises UnboundLocalError when the Kubernetes client can't be created

Automation edit form crashes with React error #130 for Proactive triggers using `after` + wildcard `expect` + `for_each`

Daily IntervalSchedule anchored at 09:00 America/New_York returns 08:00 runs when get_dates starts after a DST change

SQLAlchemy 2.1.0 breaks deployment scheduling on fresh installs

Event persister drops a whole batch when one event contains a null byte (PostgreSQL)

Workers are not honoring deployment concurrency limits during poll time

[UI] Open flow run on click in flow runs timeline

[UI] Improve flow runs timeline x-ticks

Nested schema `required` list leaks onto sibling properties, so whether None validates depends on property order

Releasing a concurrency slot doesn't wake waiters in the same process, so the slot sits idle until their Retry-After expires

Cached task runs wait for a tag concurrency slot before the cache lookup

Mapped children not yet started leave no task run when PrefectFutureList.result() raises the first failure

Flow ends COMPLETED when it returns a list of .map() results containing a FAILED task (states nested one level are ignored)

mark_deployments_ready updates every deployment on a queue on every worker poll

Allow List vs Deny List for ACLs