Error "Client-side error: Invalid state ERROR of loadbalancer resource <Load Balancer UUID>" is received while deleting a load balancer

Products

• VMware Integrated OpenStack

Issue/Introduction

Unable to delete a load balancer in OpenStack. It fails with an error similar to below 

2025-05-08 10:21:42.310 164 DEBUG vmware_nsx.services.lbaas.octavia.octavia_driver [-] vmware_nsx.services.lbaas.octavia.octavia_driver.NSXOctaviaDriverEndpoint method update_loadbalancer_status called with arguments ({'client_timeout': None},) {'status': {'loadbalancers': [{'id': '###############-####-####-####-###############', 'provisioning_status': 'ERROR', 'operating_status': 'ERROR'}]}} wrapper /usr/lib/python3.7/site-packages/oslo_log/helpers.py:66
2025-05-08 10:34:27.654 159 DEBUG vmware_nsx.services.lbaas.octavia.octavia_driver [-] vmware_nsx.services.lbaas.octavia.octavia_driver.NSXOctaviaDriverEndpoint method update_loadbalancer_status called with arguments ({'client_timeout': None},) {'status': {'loadbalancers': [{'id': '###############-####-####-####-###############', 'provisioning_status': 'ERROR', 'operating_status': 'ERROR'}]}} wrapper /usr/lib/python3.7/site-packages/oslo_log/helpers.py:66
2025-05-08 10:47:44.572 1 INFO octavia.api.v2.controllers.load_balancer [req-########-####-####-####-############### - #################### - ################### b###################] Invalid state ERROR of loadbalancer resource ###############-####-####-####-###############
2025-05-08 10:47:44.574 1 DEBUG wsme.api [req-cb023fe2-####-####-############### - #################### - ################### ###################] Client-side error: Invalid state ERROR of loadbalancer resource ###############-####-####-####-############### format_exception /usr/lib/python3.7/site-packages/wsme/api.py:222

Environment

7.x

Cause

The pool member associated with the load balancer has a reference to a stale subnet. Hence, the deletion process requires the stale subnet to delete the neutron port corresponding to the LB interface

Resolution

Delete the stale Subnet from the Octavia database. 

1. Confirm there is a member who has an IP address but does not have an associated subnet and make a note of the pool_id.
   
   openstack loadbalancer pool show <Load Balancer UUID>
   
   
2. Access the Maria Database
   
   osctl exec -ti mariadb-server-0 -- mysql --defaults-file=/etc/mysql/admin_user.cnf
3. Log in to the Octavia database and access the marida db pod:
   
   mysql --defaults-file=/etc/mysql/admin_user.cnf --host=localhost octavia -e 'select * from load_balancer';

kubectl -n openstack exec -it mariadb-server-0 bash
   
   
4. Delete the member entry from Octavia DB. The table name should be member :

select * from member \G

DELETE FROM members WHERE pool_id= '<pool_id>';
   
   
5. In NSX-V, remove the member from the pool on LB edge.


6. Retry the cascade deletion of the load balancer.
   
   
   

Note: Please take a backup of the database before making any changes Refer to Page 160, Section 10 Backup and Restore in VMware Integrated Openstack Administration Guide