c8volt ops execute retention-policy
Execute process-instance retention cleanup
Synopsis
Execute process-instance retention cleanup.
The workflow discovers process instances older than the required retention age, freezes that candidate set, validates the delete plan, and then either reports the plan with –dry-run or submits deletion after confirmation. Use compatible process-instance filters to narrow discovery, –auto-confirm or –automation for unattended deletion, and –report-file to write an audit report.
c8volt ops execute retention-policy [flags]
Examples
./c8volt ops execute retention-policy --retention-days 90 --dry-run
./c8volt ops execute retention-policy --retention-days 90 --state completed --bpmn-process-id order-process --dry-run
./c8volt ops execute retention-policy --retention-days 90 --automation --json --dry-run
./c8volt ops execute retention-policy --retention-days 90 --auto-confirm
./c8volt ops execute retention-policy --retention-days 90 --auto-confirm --force --workers 4
./c8volt ops execute retention-policy --retention-days 90 --dry-run --report-file retention-report.md
./c8volt ops execute retention-policy --retention-days 90 --auto-confirm --report-file retention-report.json --report-format json
Options
-n, --batch-size int32 number of process instances to inspect per page (max limit 1000 enforced by server) (default 1000)
-b, --bpmn-process-id string BPMN process ID to filter process instances
--children-only discover only child process instances
--dry-run discover and validate retention cleanup without submitting deletion requests
--fail-fast stop scheduling validation or deletion work after the first error
--force force cancellation of the process instance(s), prior to deletion
-h, --help help for retention-policy
--incidents-only discover only process instances that have incidents
-k, --key strings unsupported explicit process-instance key selector
-l, --limit int32 maximum number of matching process instances to inspect across all pages
--no-incidents-only discover only process instances that have no incidents
--no-state-check skip checking process-instance state before deleting
--no-wait return after deletion requests are accepted without deletion confirmation
--no-worker-limit use all queued jobs as workers when --workers is unset
--parent-key string parent process instance key to narrow retention discovery
--pd-key string process definition key (mutually exclusive with bpmn-process-id, pd-version, and pd-version-tag)
--pd-version int32 process definition version
--pd-version-tag string process definition version tag
--report-file string write an audit report to the given path
--report-format string audit report format: markdown, json (default inferred from report-file extension)
--retention-days int required non-negative age in days for process-instance retention eligibility
--roots-only discover only root process instances
-s, --state string state to filter process instances: all, active, completed, canceled, terminated (default "all")
-w, --workers int maximum concurrent workers when validating the delete plan and deleting roots (default: min(targets, 2*GOMAXPROCS, 32))
Options inherited from parent commands
-y, --auto-confirm auto-confirm prompts for non-interactive use
--automation enable non-interactive mode for commands that explicitly support it
--config string path to config file
--debug enable debug logging
-j, --json output as JSON (where applicable)
--keys-only output keys only (where applicable)
--log-level string log level (debug, info, warn, error) (default "info")
--no-indicator disable transient terminal activity indicators
--profile string config active profile name to use (e.g. dev, prod)
-q, --quiet suppress output except errors
--tenant string tenant ID for tenant-aware command flows (overrides env, profile, and base config)
--timeout duration HTTP request timeout (default 30s)
-v, --verbose show additional output
SEE ALSO
- c8volt ops execute - Discover predefined operational playbooks