Device deleted successfully with Operator Console (OC) Deep Clean but robot processes still running
Article ID: 380433
Updated On:
Products
Issue/Introduction
Devices are deleted successfully using the Operator Console (OC) Deep Clean option,
But...robot processes remain running on the server.
ps -elf | grep nim
0 S root 2237043 177580 0 80 0 - 133788 - Oct13 ? 00:00:03 nimbus(controller)
0 S root 2237046 2237043 0 80 0 - 98903 - Oct13 ? 00:00:01 nimbus(spooler)
0 S root 2237049 2237043 0 80 0 - 57760 - Oct13 ? 00:00:01 nimbus(hdb)
Environment
- Redhat Enterprise (RHEL 8)
Cause
Delete devices (Deep Clean) cleanup process failure may vary due to one or more of the following:
- robot version not 23.4.2
- robot communication issues
- robot not up and running
- robot hung
- one or more of the robot processes (controller, hdb, or spooler) is down/red hung
- robot is a hub not just a robot
- networking issues
- access/permissions/ACL
- server/robot already decommissioned
- environmental
- other/unknown causes
In this particular case, the spooler process was down/red.
Resolution
'Deep Clean' devices for robot decommissioning:
Prerequisites from DX UIM v23.4 CU2
Ensure that you meet the following prerequisites for bulk on-boarding/decommissioning of devices:
- Controller and Hub must be v23.4.2 or later
- Robot_update v23.4.2 or later
- UIMAPI v23.4.2 or later (when using the UIMAPI)
- automated_deployment_engine v23.4.2 or later
Consideration for Bulk On-boarding/Decommissioning of Devices (From DX UIM v23.4 CU2)
Robot decommissioning is supported on Windows and Linux only.
Robot decommissioning is not supported for robots installed as 'non-admin.'
Note that hub queues need to be manually cleared from the respective hubs to clean up the queues for the decommissioned robot.
When a device or robot is not reachable or down, robot decommissioning does not work. You must re-trigger the device decommissioning when the robot is up and running.
Halting hub robot uninstallation - This happens if you try to do a deep clean on a hub. The state for this job will go to Paused.
Halting oc robot uninstallation - This happens if you try to do a deep clean on an OC robot. If there is a desire to decomm a secondary OC, than OC need to be uninstalled first. The state for this job will go to Paused.
UnblacklistDevices as Delete Robot has failed - This happens if you do a deep clean on a version 23.4.1 or lower robot, where blacklist is not selected. (and deep clean is not supported). This can also happen on a robot that is 23.4.2 or greater where deep clean is selected, but the robot is not running. The robot must be running to perform deep clean.
Stage = Revoking BlacklistDevices
Received status (11) on response (for sendRcv) for cmd = '_uninstall_robot' - This happens if you do a deep clean on a version 23.4.1 or lower robot, where blacklist was also selected. (deep clean not supported)
Received status (4) on response (for sendRcv) for cmd = 'nametoip' name = '/my_domain/my_hub/my_robot/controller' - This happens when a 23.4.2 or greater robot is shut down and both blacklist and deep clean are selected. The robot must be running.
Stage = Delete Robot
Note: This will put the system in the blacklist tables, but not delete the nimsoft robot on the deleted system as there is no way to communicate with it to run the _uninstall_robot callback which will remove the robot from the device. The _uninstall_robot callback can be run after the fact if the robot is started back up. It may make more sense for development to change when the robot is actually removed from the machine.
Delete devices (Deep Clean) cleanup process failure may vary due to one or more of the following:
- robot version not 23.4.2
- robot communication issues
- robot not up and running
- robot hung
- one or more of the robot processes (controller, hdb, or spooler) is down/red hung
- robot is a hub not just a robot
- networking issues
- access/permissions/ACL
- server/robot already decommissioned
- environmental
- other/unknown causes
Feedback
