Force deleting a Transport node in NSX-T UI, does not release TEP IP addresses from pools used

Products

• VMware NSX Networking

Issue/Introduction

Symptoms:

• A transport node is in a failed state and part of the cleanup operation you preformed a FORCE delete operation in the NSX-T UI.
• Checking the IP pools allocations after the force delete shows no IP addresses were released back into the pool:

GET /policy/api/v1/infra/ip-pools/<pool-uuid>/ip-allocations/

• If there are not enough free IP addresses in the pool, adding the host back again fails, as there is no free IP addresses in the pool, which can be seen in the NSX-T manager log /var/log/proton/nsxapi.log:

ERROR L2HostConfigTaskExecutor2 IPAMUtils 4433 FABRIC [nsx@6876 comp="nsx-manager" errorCode="MP8212" level="ERROR" subcomp="manager"] Failed to get a valid IP from IpPool IpPool/<IP-Pool-UUID> with cidr null, {}.

...

ERROR L2HostConfigTaskExecutor5 IPAMServiceImpl 4433 POOL-MGMT [nsx@6876 comp="nsx-manager" errorCode="MP5109" level="ERROR" subcomp="manager"] Insufficient Resource Exception : no free IP found to allocate from the pool IpPool/<IP-Pool-UUID>
 

Note: This issue can also occur on NSX edge node.

Environment

VMware NSX-T Data Center 3.x
VMware NSX-T Data Center

Cause

When a FORCE delete operation is carried out, the IP addresses in the TEP pool were not released back.

Resolution

This issue is resolved in VMware NSX 4.1.1, available at VMware downloads.

Workaround:

• Manually release the IP addresses from the TEP pool using the below API's:

POST https://<nsx-ip>/api/v1/pools/ip-pools/<pool-id>?action=RELEASE

{

"allocation_id": "192.168.0.3"

}

Note: 192.168.0.3 is an IP address in the IP pool subnet and is showing as allocated, replace the address with an IP address you wish to free from the IP pool.

• Use the below API call in order to confirm that the Vtep IPs are released,

GET https://{{nsx-mgr}}/api/v1/pools/ip-pools/<IP-Pool-UUID>/allocations