Virtualisation

I’ve been playing with cloud platforms quite a lot recently, and in particular Juju ↗’s way of operations caught my particular attention. It has a very impressive feature set ↗ and makes deploying and linking “canned” apps very simple, whether you are using public, private or hybrid cloud instances. I wanted to set Juju up with minimal fuss to give it a quick spin, obviously deploying OpenStack etc is a little more involved for a simple POC than I would have liked and using AWS/Azure/HP Cloud/MAAS was just way overkill....

I recently tried to deploy Cisco VIRL on VMWare Workstation 10 - the install instructions ↗ are for v8 - there are a few differences I noted. It doesn’t account for the host-only network installed by default so increment all vmnets by 1 The labelling for VT-x/EPT has changed, it now lives under Settings -> Hardware -> Processors -> Virtualisation engine -> Preferred mode: You need to explicitly select Intel VT-x/EPT or AMD-V/RVI mode After this entering the sudo kvm-ok command in the VIRL CLI still output KVM acceleration can NOT be used....

VMWare, as of writing, has a nasty bug that means your backups that run utilising CBT (hint: if you have basically any enterprise backup product worth its salt, it’s got CBT enabled) it loses track of the changed blocks when the VMDK reaches any Power 2 value of 128GB (128, 256, 512, 1024, etc.) which may make your backup unrecoverable. The VMWare bug is in KB: kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=2090639 ↗ The remedy for this is to disable and re-enable (reset) CBT on the affected machines, this can be done with the machine powered off or with it turned on by running PowerCLI commands and a snapshot, we will be doing the latter, no one likes downtime ↗:...

I had recently come across the need to deploy an OVA file remotely and didn’t want to have to upload the file over VPN to the destination vCenter, the solution is to install OVFTool ↗ on a server that has local access to the vCenter and deploy it using the following syntax: ovftool.exe --acceptAllEulas -ds="[DATASTORE NAME HERE]" --net:"NAME OF OVA NETWORK"="NAME OF PORT GROUP" --prop:[PROPNAME]=[Value] \path\to\appliance.ova vi://vcsa.domain.com/DatacenterName/host/ClusterName The easiest way to get a list of objects you must reference for --net and --prop values is by running (in this example I am testing a vSphere Data Protection 5....

It’s fairly straight forward to update your Horizon Workspace vApp to the latest (this is an out-of-hours update due to downtime): Back up your vApp Ensure all VAs have connectivity to vapp-updates.vmware.com on port 80 Log into your configurator-va CLI with the root password you set up initially Run the following command to check for update: /usr/local/horizon/lib/menu/updatemgr.hzn check It should come back after checking all other appliances in the vApp with something like this:...

It’s sometimes necessary (say you’ve been working on a VM on your local workstation in either VMWare Fusion, or VMWare Workstation) to move the VM you’ve been playing with to an ESXi instance to either move into development or to have it properly backed up etc. The easiest way I find to do this is create a .ova file from the VM i’ve been working on. First install the VMWare OVATool found here on whatever flavour your OS is (sign in required): https://my....

So you’ve downloaded all the new VMWare 5.5 goodies ↗ and you want to upgrade your vCSA install to v5.5 - this is a little more involved than you may think, however it is very much worth the effort: In vSphere 5.5, the vCenter Server Appliance limitations have been extremely raised when using the embedded database: Previous to vSphere 5.5, the limits were: 5 vSphere Hosts 50 Virtual Machines With vSphere 5....

Recently had a problem were Veeam was giving bother on one VM that had a dedicated datastore, not allowing hot-add virtual appliance mode to work. I originally thought it was a problem with CBT (changed block tracking) so I disabled that, with no luck, as it transpires there were a few (all datastore formatting related) problems: The Veeam proxy’s datastore was formatted in VFMS-3 with a 2MB block size and upgraded to VMFS-5 (retaining its 2MB block size of course - otherwise a reformat would be needed)....