Rollback failing with the following error:
All Environments
The output of the 'show upgrade status' command shows that for the Service Engine Group : 'se-gp--xx-x-x' precheck fails.
+------------------+--------+---------------+----------+---------------------------+---------------+-------------------------------+---------------------------------------------+--------------------------------------------------------------+| Name | Tenant | Cloud | SE Group | State | Operation | Image | Patch | Reason |+------------------+--------+---------------+----------+---------------------------+---------------+-------------------------------+---------------------------------------------+--------------------------------------------------------------+| Default-Group | admin | Default-Cloud | - | UPGRADE_PRE_CHECK_WARNING | EVAL_ROLLBACK | 30.2.4-9071-20250626.074225 | 30.2.4-9002-2p2-20250914.033554 | Use 'show upgrade status detail filter pre_check_status' || se-gp--xx-x-x | admin | Default-Cloud | - | UPGRADE_PRE_CHECK_ERROR | EVAL_ROLLBACK | 30.2.4-9071-20250626.074225 | - | Use 'show upgrade status detail filter pre_check_status' |+------------------+--------+---------------+----------+---------------------------+---------------+-------------------------------+---------------------------------------------+--------------------------------------------------------------+
Further, if we run the command, 'show upgrade status detail filter serviceenginegroup se-xx--xx-x-x', we can see the service engine group defaults to the current controller version and does not have a previous image prior to upgrade. It means that the service engine group was created after upgrade.
| checks[2] | |The error is expected as the service engine was created after upgrade.
| check_code | SYSERR_MC_ROLLBACK_NOT_POSSIBLE |
| description | Rollback not possible since there is no previous image. |
| details[1] | Node(se-gp--xx-x-x/serviceenginegroup-#######-####-####-####-############) does |
| | not have a previous image. |
| state | UPGRADE_PRE_CHECK_ERROR |
| start_time | 2025-10-07 16:38:07 |
| end_time | 2025-10-07 16:38:07 |
| duration | 0 sec |
| checks[6] | |
| check_code | SYSERR_MC_IMAGE_VALIDATION_ERR |
| description | Invalid image state across the Cluster members. |
| details[1] | Previous image is not present in upgradestatusinfo for Node: se-gp--xx-x-x/servic|
| | eenginegroup-#######-####-####-####-############ |
| state | UPGRADE_PRE_CHECK_ERROR |
| start_time | 2025-10-07 16:38:09 |
| end_time | 2025-10-07 16:38:09 |
| duration | 0 sec |
1. Delete the SE group that was created after upgrade and then perform the rollback
OR
2. Split the rollback in 2 operations. First, rollback only the upgraded SE-Groups and then rollback the controller.
https://techdocs.broadcom.com/us/en/vmware-security-load-balancing/avi-load-balancer/avi-load-balancer/22-1/vmware-avi-load-balancer-administration-guide/nsx-advanced-load-balancer-controller-operations/upgrade-and-patches/rollback/rollback-for-avi-controller.html
https://techdocs.broadcom.com/us/en/vmware-security-load-balancing/avi-load-balancer/avi-load-balancer/22-1/vmware-avi-load-balancer-administration-guide/nsx-advanced-load-balancer-controller-operations/upgrade-and-patches/rollback/rollback-for-se-groups.html