VC 2.5 and Web Access Issues

[gary1012]

Hello all,

I'm running into a snag with my VC 2.0.2 to 2.5 upgrade and cannot seem to get the web client to work any longer. I can get to the login page on the web client but when the credentials are entered, I get a message that the web service is unavailable. Furthermore, when I applied the March 2008 patches to a ESX host, it can no longer communicate with VC. All other hosts and VMs are OK. Here's what I've checked/tested thus far:

1. Checked the VirtualCenter server and ensure that Tomcat was up. Netstat -a shows that 8086 is listening.

2. Checked the ESX host that had the patches applied and ensured that mgmt-vmware was running, successfully performed a restart JIC, and check to see if it was accessible by VIC and a direct web client connection; which it was.

3. Dropped it from the VC inventory and re-added the host. Got an dialog box error attached below.

4. Checked the proxy.log file on the VC server and found these entries at the end of the log:

2008-03-19 08:07:16.251 'HTTP server /var/lib/vmware/hostd/docroot' 12999600 warning HandlePost: Not supported HTTP method for given URI

2008-03-19 08:07:16.301 'HTTP server /var/lib/vmware/hostd/docroot' 20368304 warning HandlePost: Not supported HTTP method for given URI

5. Excerpts from the hostd.log file show after attempting to register the host into VC:

2008-03-19 08:07:16.251 'HTTP server /var/lib/vmware/hostd/docroot' 12999600 warning HandlePost: Not supported HTTP method for given URI

2008-03-19 08:07:16.301 'HTTP server /var/lib/vmware/hostd/docroot' 20368304 warning HandlePost: Not supported HTTP method for given URI

6. When accessing the WebClient on VirtualCenter (fails and states the Web Service Not Available)

com.vmware.webcenter.exceptions.ActionException: Web service unavailable.

at com.vmware.webcenter.control.action.LoginAction.execute(Unknown Source)

7. Uninstalled Web Access from VC and reinstalled it. Got the same errors above.

I googled the HandlePost message but wasn't able to find anything on the topic. Does anyone have any suggestions before I open an incident with VMware?

[mike_laspina]

Hello,

This error is more common to a DNS problem or port not open. Make sure that you can resolve the VC and the host using a ping to the FQDN's to each other.

[gclinch]

Not sure I can help with the web client upgrade, but I had a similar issue with the March patches to ESX 3.0.2 - when trying to reconnect to the esx host VC gave an error along the lines of "Either the server software is not responding or there is a network issue". The fix appears to be to remove the vpxa package from the ESX host (it is reinstalled by the VC server when connecting), like so:

Get the version of vpxa that is currently installed:

# rpm -qa | grep vpxa

VMware-vpxa-2.5.0-64192

then remove it:

# rpm -e VMware-vpxa-2.5.0-64192

Then try reconnecting the host from VC.

[gary1012]

Thx for the tip on vpxa. That worked great. As for DNS resolution, it's working both ways.