Issue 1 - OVF 6.5.0.0 Bug:
OVF template deployment fails with:
This version of vCenter Server does not support Deploy OVF Template using this version of the vSphere Web Client. To Deploy OVF Template, login with version 6.5.0.0 of vSphere Web Client
The dialog should not be shown if the the VCSA version is 6.5 U1 or higher.
Cause and Remediation:
The cause may be that the vmware-content-library is not loaded.
To find out:
1. Use SSH to get a console to the VCSA (vCenter appliance)
2. Enable shell and check library status:
> shell.set --enabled true
> shell
3. See if library is enabled:
service-control --status vmware-content-library
4. If the service returns a message that it is not running, then proceed below. We will back-up the old config file, and enable the new one.
# cd /etc/vmware-content-library/config/
Is there a rpmnew config file?
# ls ts-config.properties.rpmnew
If the file exists, then proceed to the next step. Otherwise, then an old config not being deployed is not the issue. It's something else.
Backup:
# cp ts-config.properties ts-config.properties.old
# cp ts-config.properties.rpmnew ts-config.properties.rpmnew.bkup
Replace/overwrite:
# mv ts-config.properties.rpmnew ts-config.properties
Start the library service:
# service-control --stop vmware-content-library;service-control --start vmware-content-library
Click the vSphere Web Click refresh icon, and see if it works.
---
Issue 2 - Network Does Not Connect and Starts Disconnected:
OVF or VM template deployment is successful, but the network is disconnected.
Failure Message 1:
The /var/log/vmware-vmsvc-root.log shows the error:
DeployPkgInGuest failed, error 5
Remediation 1:
Add the following entry to /etc/vmware-tools/tools.conf:
[deployPkg] enable-custom-scripts = true
No Failure Message:
Was the template edited w/o Power being enabled?
Check the .vmx file of the template, and verify the following is in the template:
ethernet0.startConnected = "TRUE"
Failure Message 2:
Reconfigure virtual machine --> Failed to connect virtual device ethernet0
The message is likely found in the /var/log/hostd.log and the /var/log/vpxa.log
Remediation 2a:
If the server version is 7.0.0 and the NSX version is 6.4.x, this is likely the dvfilters max bug.
See technote, 318707.
Steps:
a. Put the host in maintenance mode:
vSphere client --> datacenter --> host --> Actions --> Maintenance Mode --> Enter Maintenance mode.
b. Update the max setting:
# esxcli system module parameters set -a -p dvfiltersMaxFilters=8192 -m dvfilter
c. Reboot the host via the vSphere client:
vSphere client --> datacenter --> host --> Actions --> Power --> Reboot
d. Take host back out of maintenance mode:
vSphere client --> datacenter --> host --> Actions --> Maintenance Mode --> Exit Maintenance mode.
Remediation 2b:
If you found the template had the NIC disconnected, you can connect the NIC by:
a. vSphere --> datacenter --> VM --> Edit Settings
b. In settings, change the NIC to another portgroup on the same DV switch, and check the box Connected. Click OK.
c. Return to: VM --> Edit Settings, and confirm the NIC is still Connected.
d. If still says Connected the DV is likely okay. Switch the VM back to the original portgroup, confirmed Connected. Click OK.
e. Return to: VM --> Edit Settings, and confirm the NIC is still Connected.
Remediation 2c:
If the server version is 7.0.x and less than Update 2c, and the vm log shows "no more ports available", and you know that you are not exceeding the number of ports configured for the DV Switch and port group, you should upgrade to the latest 7.0.x update (at least 7.0 U2c). This specific error is caused by a memory leak for the pktlist in portset world data. See technote: 832338.
Temporarily, this can avoided by rebooting the ESXi host.
However, the reboot will require evacuating the VMs and entering maintenance mode, so it generally makes sense to just upgrade it "now".
previous page
|