- Posts: 7
- Thank you received: 1
vAPI endpoint status changes from green to yellow constantly in VMware 6.0
- David Hovey
- Topic Author
For all the other admins out there who have ever come across this issue I figured I'd give a quick write up about the behavior it can cause in regards to NETLAB+, current work arounds, and of course the solution to this problem.
First the potential NETLAB+ problems. This vAPI error causes a substantial slow down when it comes to API calls, and when a large number of API calls are made from the NETLAB+ systems (e.g. a set of 5, 7 VM, pods starting at exactly the same time) some of the API calls will timeout and fail. This can present itself in much the same way as a reservation that was cancelled in progress so expect to see a number of "Error code: E_MBUSC_TIMEOUT" messages, but the key difference is that more than a few restarts will be needed for the reservation to finally start, and reservations can take upwards of 30 minutes to start (timing not applicable to Cisco equipment pods which are highly dependent on physical equipment boot times). If this sounds similar to your use case then it may be pertinent to log into your vSphere system and check to see if the vAPI error is present.
The current work around is to restart the vAPI service, which is a fairly easy task assuming you have some form of console connectivity (direct or remote). When logged into the vSPhere system bring up a shell instance and run these two commands:
service-control --stop vmware-vapi-endpoint service-control --start vmware-vapi-endpoint
The solution is to update your vSPhere system to at least 6.0 U2, in which this issue was fixed. (I might do a separate write up on updating the vSPhere system but at this time google is your friend)
relevant VMware Knowledge Base article: