vsphere

This one is just a quick post regarding an issue I came up against just before Christmas. To get straight to the point, I found an issue where in multiple separate environments, when cloning a virtual machine via the VMware vCenter HTML5 Web client interface, if you select the tick box to “customize this virtual machine’s hardware” and change any of the settings of the VM, the cloned virtual machine’s VMDK file(s) will be pointed to the VMDK file(s) of the source machine you used for the clone.

I was recently involved in an issue in an evnironment that started out looking like a simple SSL certificate replacement and ended with multiple hours of troubleshooting with GSS and redeployment of vSphere Replication. Fun times, right?! Let’s take a step back. This customer has a straight forward environment. Two physical sites, one vCenter Server and external PSC per site (6.5) in the same single sign on domain. Each site had one vSphere replication appliance and one Site Recovery Manager Server, version 8.

We had a customer recently that upgraded from vSphere 6.5 to 6.7 and following the upgrade to 6.7, virtual machines were sporadically being reset by vSphere HA with the following message in the vCenter event log: Error message on <VM Name> on <Host Name> in <Cluster>: VMware ESX unrecoverable error: (vcpu-0) vmk: vcpu-0:Unable to decompress BPN(0x100018003) from frameIndex(0x3d4aa2) block 1 for VM(2103599) We got this logged with GSS quickly and after being sent on a bit of a goose chase with firmware and driver upgrades (the versions we were running we compatible, but not latest), GSS noted that 5 other cases has been logged over the past couple of days with this same issue.

Recently I was playing around with NSX-V in the lab and getting myself more familiar with the automation options available. During the deployment of one of my environments, I couldn’t get the NSX Controllers to come up without failing. NSX Manager would deploy the controller VM, but it sat in a “deploying” state for around 45 minutes and then failed with a message “no route to host”. I checked out the console of one of the controllers directly after deployment and as soon as it booted up it went to the following GRUB boot loader screen:

[toc] Introduction Starting from VMworld US 2017, VMware announced their plans to support PSC replication and Enhanced Linked Mode with the vCenter Embedded deployment model, simplifying overall deployment and management requirements for a multi-vCenter environment. This feature is now included in the release of vSphere 6.7. In this post, I go over a little history for common deployment models, as well as what the new “vCenter Embedded Linked Mode” brings to the table.

It’s back, finally! A new feature with vCenter 6.7 is the ability to repoint a vCenter Server to another Platform Services Controller node, that resides in an entirely different vSphere SSO domain. This functionality is huge for domain consolidation, and also domain splitting (which admittedly is a less required use case from what I’ve seen, but something that still can be a useful use case). Edit: As per a comment from Rupak, I believe this feature is only available on the vCenter Server Appliance and is not available for the Windows deployment of vCenter 6.

Overnight VMware announced a new technical resource for vSphere administrators called vSphere Central. Check out the blog post here. In conjuntion with the above, Adam Eckerle released a blog post to announce the availability of a new tool named vSphere 6.5 Topology and Upgrade Planning Tool. The blog post can be found here. I don’t want to repeat much of the post here, but in short to quote Adam; “This tool aims to help customers plan and execute both upgrades to vSphere 6.

I was recently chatting to a few guys regarding powering on and off VMs within a vSphere lab and using a script to do so in some sort of controlled manner (or force power off everything if desired). A few of us had a quick search, and couldn’t see anything that met the requirements of being able to power on and power off the lab, as well as place VMs into some sort of priority group.

Ahh, vSphere Host Profiles. My old friend. One of our customers recently had a hardware failure on one of their Dell M620 blade servers, and the server was repaired under warranty which included a motherboard swap. The server got inserted into the chassis and booted back up OK. The server was part of a 12 node cluster, so the customer tried applying the host profile, which would fail when it hit 22%.

Runecast Analyser was founded in 2014 and at a high level, is a product to provide proactive analysis for VMware environment. I’ve only recently been introduced to Runecast and I’ve had the chance to download and use it in my lab environment. I was lucky enough to meet Ched and Stanimir recently at the Melbourne VMUG. I thought I would take the opportunity to record a walkthrough of the product for those that may be interested to see what Runecast Analyser is all about.