Troubleshooting VMware Web Client issues
In order to troubleshoot web client issues, we need to fetch logs from our plug-in appliance and from vCenter.
Collect the plug-in information by running:
wget --no-check-certificate https://172.16.91.206/vwc/vwc.zip
- Collect the logs of the vSphere Web Client service
Troubleshooting start-up issues for Host PowerTools for VMware
When the Host PowerTools for VMware starts it attempts to register itself in the vCenter server. To do so, it needs:
- IP address
- vService definition
- IP connectivity to vService (see 202319392 TCP and UDP ports required for INFINIDAT products)
Troubleshooting deploying issues
Deployment of the OVA when the VMware VirtualCenter Management WebServices is not running
When deploying the OVA in Windows-based installations with vCenter Server, the VMware VirtualCenter Management WebServices must be running. When it is not, the deployment fails with the following error message:
To fix this, start that service and then deploy the OVA.
Troubleshooting network settings
Network interface eth0 does not have an IP address
By default, the appliance is configured to get an IP address from a DHCP server. If an IP address is not configured, the appliance will print the following message to the console:
Network interface eth0 does not have an IP address.
If this happens, make sure that the network adapter of the VM is connected and configured to connect at power on, and that is it attached to the correct network:
If you need to change the IP address from DHCP to manual or vice-versa, you can do this from the options tab:
Troubleshooting OVF issues
OVF file /opt/vmware/etc/vami/ovfEnv.xml <error message>
The vService provides the appliance, the connection parameters and credentials to vCenter. The vCenter information is exposed to the appliance in XML form via VMware Tools. When the vService is not configured or there is a problem with is binding through VMware Tools, the appliance will print one of the following messages to the console:
OVF file /opt/vmware/etc/vami/ovfEnv.xml does not exist
OVF file /opt/vmware/etc/vami/ovfEnv.xml is corrupted
OVF file /opt/vmware/etc/vami/ovfEnv.xml is empty
OVF file /opt/vmware/etc/vami/ovfEnv.xml is not a valid XML
And verify that the OVF is configured as follows:
Troubleshooting for Recovering from the error about missing/invalid OVF after Upgrade using CD drive
1. Power off the VM
VIA the VWC UI:
Disconnect the CD/DVD drive
2. Edit Settings --> Virtual Hardware--> CD/DVD drive1
3. Change to "client device"
4. Press OK
Reconfigure the ovfenv.xml
5. Edit Settings-->vApp Options-->Authoring section-->OVF settings-->OVF environment transport
6. Uncheck the "ISO image" and "VMware Tools" checkboxes
7. Press OK
8. Edit Settings-->vApp Options-->Authoring section-->OVF settings-->OVF environment transport
9. Check (v) the "ISO image" and "VMware Tools" checkboxes
10. Press OK
11. Power ON the vm
Troubleshooting connectivity issues
vCenter configuration is invalid: vCenter URL/IP is <error message>
The appliance expects to find the vCenter fully qualified domain name and the IP address in the OVF.
If the appliance fails to find the vCenter information, one of the following message will be printed to the console:
vCenter configuration is invalid: vCenter URL/IP is missing in OVF
vCenter configuration is invalid: vCenter URL/IP is empty in OVF
In case the vCenter information exists but yet unreachable, the following message will be printed to the console:
Connection to vCenter failed: <reason>.
This can happen when:
- The vCenter address does not exist in the OVF
- There's an IP connectivity error between the appliance and the vCenter server
First, check the Managed IP Address and Name in the vCenter Server Settings, as show in the following image:
If one of those parameters are not configured or are with the wrong value, then:
- Set the right value and press OK
- Power off the appliance and power it on again
If the vCenter settings are correct, please check that the appliance can access the vCenter in the TCP ports mentioned here.
No InfiniBoxes Detected
If the ESX host does not have an appropriate zoning to the InfiniBox FC ports, the following error message will be shown:
Troubleshooting registration issues
Plug-in registration has failed with <reason>
When the appliance fails to register itself as an extension in vCenter, the following message will be printed to the console:
Plug-in registration has failed.
First attempt to register at <url> failed with <reason>.
Second attempt to register at <url> failed with <reason>.
Please check that the vCenter IP address is correct, and that the appliance can access the vCenter in the TCP ports mentioned here.
First attempt to register at https://<>/vsm/extensionService failed with HTTP error 400: Bad Request.
Plug-in Registration has failed with HTTP error 400: Bad Request
If the IP address in the Managed IP Address field in the vCenter Server Settings is not the actual vCenter address, but rather of another server (the ESXi, for example), the following error message will appear on the console:
Plug-in registration has failed.
First attempt to register at https://<ip>/vsm/extensionService failed with HTTP error 400: Bad Request.
Second attempt to register at https://<ip>/vsm/extensionService failed with HTTP error 400: Bad Request.
When this happens, go bak to the vCenter Server Settings, fix the Managed IP Address field with the correct vCenter address, and re-start the appliance.
Address already in use
In case the appliance fails to start and the console shows the error "Address already in use", reboot the appliance and it will start up on the second boot.
Troubleshooting Upgrade issues
One of the cases we saw of failed upgrade is not enough disk capacity.
Although we have a script to check the disk capacity prior to the upgrade this wouldn't help in all cases.
All versions which are smaller then 2.0.4 has a disk size of 2.0 GB.
This leaves for the upgrade only a bit of space, in it the upgrade downloads and extract the upgrade packages.
In the script, we rotate the logs and delete files from /opt/infinidat/host-power-tools-for-vmware/data/logs/
Take into account that the script operate after the initial upgrade file was downloaded.
Therefore there might be couple of cases it wouldn't work:
- There wasn't enough space to download the initial upgrade package.
- The cleanup didn't free enough space to allow the upgrade.
In both of those cases please ssh to the sa using du and df find what can be deleted without causing damage and restart the upgrade.