Disabling host monitoring
If as per above you are carrying out any planned network maintenance activities that would affect the host’s management network connectivity best to disable host monitoring. Main reason for this is to prevent unnecessary alerting of the host isolation response configured for the HA cluster. To do need to edit the cluster settings
The following describes how to disable host monitoring:
- Logon to your vCenter instance
- Go to the Home inventory and navigate to the Hosts and Clusters view.
- Select the cluster you are performing maintenance and go to Manage | Settings | vSphere HA, and click on Edit.
- Deselect the Enable Host Monitoring checkbox and click on OK:
5. To re-enable simply repeat steps 1-3 above and this time for step 4. tick the box beside Enable Host Monitoring and click on Ok
The primary communication mechanism citrix Provisioning services (PVS) uses is Kerberos authentication. It registers the components against AD through the Service Principal Name (SPN) and permitting the AD DC’s to identify the accounts, which manage the running services.
From time to time registration issues can occur, if that happens you will probably find your PVS service could fail.
To avoid this happening, you have to use the
setSpn command in order to give the right permissions to the account that manage the earlier described services (such as the PVS Soap or streaming service for example) by running the following command:
setSpn –a PVSSoap/PVS_Server_FQDN <username_managing_service>
As Microsoft do recommend you do run their BPA for hyper-v its perhaps a good idea to do it – perhaps what is interesting is they suggest you run it every month.
That being so did take the step to automate via a SCO runbook using the below cmdlets so I could take a “hands off” approach to it.
To start a scan using the Hyper-V best practice analyzer, type the following command:
Invoke-BpaModel –BestPracticesModelId Microsoft/Windows/Hyper-V
After running the Hyper-V BPA, you can use the
Get-BPAResult command to analyze the results. The following command shows the BPA scan results:
Get-BpaResult –BestPracticesModelId Microsoft/Windows/Hyper-V
Voila! your good to go
Recently was working on an edgesight console and noted really poor performance running pretty much any type of report. One way of mitigating this is to limit data uploads from the agents to the EdgeSight Server database by deselecting the type of performance data you are not interested in gathering, will be surprised by how much stuff you probably don’t ever need / use to monitor
To configure these settings, navigate to Configure | Server Configuration | Data Maintenance | Upload Configuration
Hitting hostd unresponsive issues?
The VMware-hostd service as you probably know is the main communicator between the host and the VMkernel. Handles most of the management tasks, including interfacing connections from vCenter and remote command-line connections with esxcli or PowerCLI.
If the hostd service fails, the ESXi host becomes unavailable for management. Crucially though, ESXi is still aware of all the running VMs, datastore mounts, so important to note the hostd failure does not mean virtual machines are taken down. At the same time, nearly all your administration will be done through vCenter so vital for the administrator hostd is up and running again.
In some cases, when the hostd service does not respond and cannot be restarted by normal means, vSphere 5/6 offers a new utility called localcli which can help overcome this issue.
As this tool effectively bypasses hostd, it makes it ideal to make changes in emergency situations. Any changes made with the tool will not be reflected in the user’s interface or hostd internal state.
As localcli is an equivalent to esxcli the same syntax applies.
Tool should be used carefully as localcli can bring hosts into an inconsistent state so should be used only as a means to recover from failure. In unsure VMware support can provide assistance with the tool.
How to configure your ESXi Host to Redirect Their Core Dumps
The primary method for configuring hosts to redirect their core dumps to your newly setup ESX Dump Collector is to use the esxcli command-line tool:
- Logon via ssh tool to your esxi host of your choice eg putty
- To run a check on your existing Dump Collector configuration:
esxcli system coredump network get
- Then need to configure the host’s dump redirection settings (will need management VMkernel interface and collector’s IP and port):
esxcli system coredump network set -v vmk0 -i 192.168.0.1 -o 6500
- Switch on dump redirection:
esxcli system coredump network set -e true
- To verify the settings are configured correctly:
esxcli system coredump network get
In enterprise environments might want to use the host profiles feature to configure to save you the bother of individually configuring each one – details – https://pubs.vmware.com/vsphere-50/index.jsp?topic=%2Fcom.vmware.vsphere.install.doc_50%2FGUID-E5669933-106E-4E6B-8D5A-6780415A12FD.html
As you may know in any of the 7.x variants of XenApp there is 3 types of potential databases that can be setup as part of your XenApp site, they are:
I) Site database – containing information relating to the objects in your XenApp 7.x site
II) Logging database – containing information relating to historical changes made in the
III) Monitoring database– containing desktop director information
By default all the information from above is held in one database, the site database.
To move logging and monitoring off to their own databases its very very straightforward simply click on the relevant datastore in the Databases section within the Studio console and choose Change Database from under the actions pane