Scenario - RFM scenario, NSO ver 5.3.3, Commit queues enabled for target device. A kicker triggers the service redeploy, but because there is a concurrent request (entirely separate unrelated request by some other user / service pkg targeting the same device's config) in progress, the service redeploy fails.
<ERR> .... ncs[2547794]: devel-c prepare error {commit_queue_has_overlapping, ""} daemon id: 1
<ERR> .... ncs[2547794]: devel-c prepare/"ncs-internal-device-mgr" error {commit_queue_has_overlapping, ""}
........
<DEBUG> ... ncs[2547794]: ncs progress db=undefined usid=8810 thandle=1147447: ncs: service xxxx: service re-deploy error
So far this is expected as we have the check integrity flag set to true -
"devices global-settings commit-queue check-integrity true".
Some time later after the other request's commit q task has completed - the log shows a second attempt to re-deploy again.
<DEBUG> .... ncs[2547794]: ncs progress db=undefined usid=8817 thandle=1147469: ncs: service xxxx: service re-deploy...
Question - What controls or triggers this seemingly automatic second attempt to redeploy the service ? Is there any setting or flag that can be used to disable the retry behaviour ? How many times will NSO retry if the same overlapping error occurs again ? Setting retry values to zero for commit related settings have no impact on this behaviour - "devices global-settings commit-queue retry-attempts 0" and "devices global-settings commit-retries attempts 0".
