conductor-oss/conductor · error · IllegalArgumentException

Cannot archive workflow: %s with status: %s

Error message

Cannot archive workflow: %s with status: %s

What it means

Thrown by removeWorkflowIndex when archiveWorkflow=true but workflow.getStatus().isTerminal() is false. Archival copies the workflow into the index as the durable record, and Conductor only allows archiving workflows that have reached a terminal state (COMPLETED, FAILED, TERMINATED, etc.). Archiving a running workflow would freeze an incomplete snapshot as the 'final' record.

Source

Thrown at core/src/main/java/com/netflix/conductor/core/dal/ExecutionDAOFacade.java:437

        try {
            queueDAO.remove(DECIDER_QUEUE, workflowId);
        } catch (Exception e) {
            LOGGER.info("Error removing workflow: {} from decider queue", workflowId, e);
        }
    }

    private void removeWorkflowIndex(WorkflowModel workflow, boolean archiveWorkflow)
            throws JsonProcessingException {
        if (archiveWorkflow) {
            if (workflow.getStatus().isTerminal()) {
                // Only allow archival if workflow is in terminal state
                // DO NOT archive async, since if archival errors out, workflow data will be lost
                indexDAO.updateWorkflow(
                        workflow.getWorkflowId(),
                        new String[] {RAW_JSON_FIELD, ARCHIVED_FIELD},
                        new Object[] {objectMapper.writeValueAsString(workflow), true});
            } else {
                throw new IllegalArgumentException(
                        String.format(
                                "Cannot archive workflow: %s with status: %s",
                                workflow.getWorkflowId(), workflow.getStatus()));
            }
        } else {
            // Not archiving, also remove workflow from index
            indexDAO.asyncRemoveWorkflow(workflow.getWorkflowId());
        }
    }

    public void removeWorkflowWithExpiry(
            String workflowId, boolean archiveWorkflow, int ttlSeconds) {
        try {
            WorkflowModel workflow = getWorkflowModelFromDataStore(workflowId, true);

            try {
                removeWorkflowIndex(workflow, archiveWorkflow);
            } catch (NotFoundException e) {

View on GitHub (pinned to cf7c3e4a8a)

Solutions

1. Terminate the workflow first (e.g. via the terminate API) so it reaches a terminal status, then archive.
2. If you just want to remove it without archiving, call removeWorkflow with archiveWorkflow=false.
3. If the workflow is genuinely stuck and will not terminate, investigate the decider/sweeper rather than forcing archival.

Example fix

// before - archive a still-running workflow
facade.removeWorkflow(id, /* archiveWorkflow */ true, ttl);
// after - terminate first, then archive, or skip archival
facade.terminateWorkflow(id, reason);
facade.removeWorkflow(id, /* archiveWorkflow */ true, ttl);

// Ensure terminal status before archiving
if (!workflow.getStatus().isTerminal()) {
    throw new IllegalArgumentException(
        "Cannot archive workflow " + workflow.getWorkflowId()
            + " with non-terminal status " + workflow.getStatus());
}

try {
    facade.removeWorkflow(id, true);
} catch (IllegalArgumentException e) {
    // workflow not terminal -> terminate first or remove without archiving
}

When it happens

Trigger: Calling removeWorkflow(id, archiveWorkflow=true) while the workflow is still RUNNING, PAUSED, or in any non-terminal status. The check is purely on workflow.getStatus().isTerminal().

Common situations: A cleanup/sweeper job attempts to archive and remove a long-running or stuck workflow. A manual operator script calls the archive path on a workflow that was never terminated. A race where the workflow transitioned out of terminal status (should not happen, but indicates a state bug).

Related errors

• Cannot archive task: {} of workflow: {} with non-terminal st
• Workflow can not be serialized to json
• Task %s of workflow %s can not be serialized to json
• Task %s not found
• Error creating workflow: %s

AI-assisted analysis of conductor-oss/conductor@cf7c3e4a8a (2026-08-14). Data as JSON: /api/errors/e57b3496b3def855. Report an issue: GitHub.