apache/dolphinscheduler · error · ServiceException

10145

10145

Error message

delete worker group by id fail, for there are {0} workflow instances in executing using it

What it means

Thrown in deleteWorkerGroupById when queryByWorkerGroupNameAndStatus finds non-terminal (running/pending) workflow instances whose workflow instance still uses this worker group. Deletion is blocked because live instances depend on it; the message reports how many.

Source

Thrown at dolphinscheduler-api/src/main/java/org/apache/dolphinscheduler/api/service/impl/WorkerGroupServiceImpl.java:333

    public void deleteWorkerGroupById(User loginUser, Integer id) {
        if (!canOperatorPermissions(loginUser, null, AuthorizationType.WORKER_GROUP, WORKER_GROUP_DELETE)) {
            throw new ServiceException(Status.USER_NO_OPERATION_PERM);
        }
        WorkerGroup workerGroup = workerGroupDao.queryById(id);
        if (workerGroup == null) {
            log.error("Worker group does not exist, workerGroupId:{}.", id);
            throw new ServiceException(Status.DELETE_WORKER_GROUP_NOT_EXIST);
        }
        List<WorkflowInstanceSummaryDto> workflowInstances = workflowInstanceDao.queryByWorkerGroupNameAndStatus(
                workerGroup.getName(),
                WorkflowExecutionStatus.NOT_TERMINAL_STATES);
        if (CollectionUtils.isNotEmpty(workflowInstances)) {
            List<Integer> workflowInstanceIds =
                    workflowInstances.stream().map(WorkflowInstanceSummaryDto::getId).collect(Collectors.toList());
            log.warn(
                    "Delete worker group failed because there are {} workflowInstances are using it, workflowInstanceIds:{}.",
                    workflowInstances.size(), workflowInstanceIds);
            throw new ServiceException(Status.DELETE_WORKER_GROUP_BY_ID_FAIL, workflowInstances.size());
        }

        checkWorkerGroupDependencies(workerGroup);

        workerGroupDao.deleteById(id);
        boardCastToMasterThatWorkerGroupChanged();

        log.info("Delete worker group complete, workerGroupName:{}.", workerGroup.getName());
    }

    /**
     * query all worker address list
     *
     * @return all worker address set
     */
    @Override
    public Set<String> getWorkerAddressList() {
        return registryClient.getServerNodeSet(RegistryNodeType.WORKER);

View on GitHub (pinned to 02eac45a1b)

Solutions

Wait for running workflow instances to finish or stop/kill them, then delete the worker group.
Reassign the affected workflow/schedule definitions to another worker group and let old instances terminate.
If instances are stale/leaked, fix their state (or let them time out) so no NOT_TERMINAL instances reference the group, then retry.

Example fix

// before
workerGroupService.deleteWorkerGroupById(loginUser, id); // 10145 while jobs run
// after: stop running instances first
workflowInstances using the group -> stop/kill via UI or API
workerGroupService.deleteWorkerGroupById(loginUser, id); // succeeds

Defensive patterns

Strategy: validation

Validate before calling

List<WorkflowInstanceSummaryDto> active = workflowInstanceDao
    .queryByWorkerGroupNameAndStatus(groupName, WorkflowExecutionStatus.NOT_TERMINAL_STATES);
boolean deletable = active.isEmpty();

Try / catch

try {
    workerGroupService.deleteWorkerGroupById(loginUser, id);
} catch (ServiceException e) {
    if (e.getCode() == 10145) { /* stop/wait running instances, retry */ }
    else throw e;
}

Prevention

Schedule worker-group retirement for windows with no active or scheduled workflows using the group.
Migrate workflow schedules to a new group and drain old instances before deleting.
Monitor for leaked non-terminal instances that permanently block deletion.

When it happens

Trigger: Calling DELETE /worker-groups/{id} while at least one workflow instance in a NOT_TERMINAL state (e.g. RUNNING_EXECUTION, WAITING) was scheduled/executed with the group's name.

Common situations: Deleting a worker group while long-running or scheduled workflows are still active; a stuck non-terminal instance blocks deletion even though nothing is really running; deleting during peak scheduler usage.

Understand the failure class

Background: "Invalid state transition" errors: "status must be X, actually Y", "already rejected/charging/uninstalled", "cannot ... while running" — what they mean when a library rejects your call — this error's family across 31 libraries.

Related errors

AI-assisted analysis of apache/dolphinscheduler@02eac45a1b (2026-09-06). Data as JSON: /api/errors/5a7210149197de3e. Report an issue: GitHub.