a general system error occured: internal error

[zfinlayb]

I have a problem adding a new host to a Datacentre in VC. When I go to add the Host I enter in my Username and Password and I get "a general system error occurred: internal error". I am able to login to the ESX server directly using VI. This started happening when I decided to trash the DC and start again. I have configured this before with no hassles. Can anyone point me in the right direction or will I have to re-install and start again. This is a new install so I can start again but I would like to learn from my mistakes. Thanks in advance.

[chrisau]

I have the same problem.

If I look at the Recent Tasks window in the host while it's attempting to be added, I get the following messages (in chronological order):

1)Set Entity Permission Rules - The requested change could leave the system without full administrative privileges for any user or group

2)Update User - ha-folder-root - Completed

3)Create User - ha-folder-root - The specified key, name, or identifier already exists.

What is VirtualCenter trying to do here?

Thanks

[VirtualKenneth]

Have you tried to restart the mgmt-vmware[/i] services on the ESX host that you want to add. I had this same problem and I can remember that restarting this allowed me to re-insert the host back.

(Try restarting the VirtualCenter Service as well)

[conyards]

Probably a Dumb question but are you adding this into a HA or DRS cluster? If so are all the shortnames for you're hosts configured in DNS, or does the host you're adding in have these listed in /etc/hosts ?

[zfinlayb]

I have restarted both the ESX server and Virtual Centre and get the same problem. Could it be something to do with the VPXUSER? lost permissions?

BTW this looks like the same problem as chrisau

To answer conyards question:- No it is not in HA or DRS, However i Did at one stage have it configured for DRS and decided to undo it. Maybe that had something to do with it?

[zfinlayb]

Also yes I have configured the shortnames in DNS. That was one of my first issues... When trying to move a server between ESX Servers using vmotion etc it would fail if you had added the server in with only an IP address. When I realised this... I configured /etc/resolv.conf to point to my DNS Server. So now all my ESX servers (All 2 of them) are pointing to my DNS Servers instead of using the hosts file. I have just checked and the ESX Servers can still resolve via shortnames

[VirtualKenneth]

I'm not sure but i thought that FQDN (and not short names) was the only thing supported for HA.

When (re)adding the host to VC the system (re)creates the vpxuser however you could try and delete the account before readding it.

Isn't something logged in the /var/log/vmware/vpx/vpxa.log ? (that's the basic logging from VC to the ESX server.)

Eventually you can also check the /var/log/vmware/hostd.log

[zfinlayb]

Thanks Vliegenmepper

On the esx host that I was trying to add to Virtual Center... I deleted the vpxuser and did a service mgmt-vmware[/i] restart. I went back into Virtual Center, removed the host and re-added the host no problems. BTW the tip for looking in the /var/log/vmware/vpx/vpxa.log helped as it said vxpuser already exists[/i] Thanks once again

[VirtualKenneth]

Great

no problem!

[chrisau]

This solved my problem also. Thanks!

[jrtjr]

I too was hitting this issue with one of my 5 ESX servers.

I removed the Host from the VC

I removed the vpxuser (userdel -r vpxuser)

I rebootws the ESX machine.

Problem solved.

Thanks for the help in this post!

[metroPCS_Admin]

I am having the same issue only with adding a VMware Server 1.0.1 host to my Virtual Center.

[krugster]

I was using VC 1.x and everything was working fine. I upgraded to VC 2.x and now I can't add my VMware Server 1.x in. I keep getting a "a general system error occurred: internal error" message.

Please help!

[Eddy]

I have seen then once in a blue moon, and what has worked for me after checking all the DNS entries, hosts files, etc... I simply issued an ipconfig /flushdns from my VC server; and it all worked from that point..

Also try hoping on the VC server and attemting to run the VI client locally from there, these will rule out any wan/lan latency issues communicating to VC...

[EricChampagne]

Hi,

There is the workaround

1) Delete the ESX object in VC

2) Log with SSH on ESX server

3) Delete VPXuser Account (userdel -r vpxuser)

4) Restart Management services (serivces mgmt-vmware restart)

5) Go back on VC

6) Add the ESX host again,

Works for me.