Hi,
we have a UCSD deployment with automated ESXi staging. We are facing problems with one particular server (hostname/IP combination) which we cannot re-deploy.
The server has been deployed and then the workflow has been rolled back. Now, re-deployment always fails with the following error message:
Aug 30, 2016 08:32:51 UTC Executing custom action ESXi Deployment (PXE Boot)
Aug 30, 2016 08:32:51 UTC Entering critical section with name :PXE-IP-Lock
Aug 30, 2016 08:32:51 UTC Released 1 critical section locks
Aug 30, 2016 08:32:51 UTC Task: ESXi Deployment (PXE Boot) failed with error - Failed to setup PXE boot in PXE Boot.IP Address pool exhausted, All addresses are already assigned or used in previous PXE reqests. Please rollback any failed PXE request to release the address., selectedContext=<None>
Further informtion:
- Other servers (hostname/IP combinations) can be deployed
- It's independent on the physical blade
- It's independent on the MAC address used for PXE
- Only one specific hostname/IP combination is having problems
- No "hanging" PXE requests in UCSD visible (Physical>Compute>PXE Boot Requests)
- No entry in UCSD BMA DHCP DB (has been cleared manually)
- There are plenty of free addresses in DHCP pool
Questions:
- Any idea what cloud cause the problem observed?
- Is there a way to get more information besides the inframgr-log?
Thanks in advance!
