When I started building my VMware ESXi server, I did not have a Switch that could handle 10Gbe SFP+. Now that I have a Dell X1052, I figured I would cable up a 10Gbe DAC and get moving. Much to my surprise, I received a link light on the switch, but not on the motherboard.
Digging into the VMware side, I noticed that the 10Gbe NICs are not available, only the 1Gbe NICs.
A quick Google search brought me to a great site for VMware knowledge, tinkertry.com. It appears that the drivers for the 10Gbe are not loaded. So following the directions here, we open the SSH Console and enter the following command.
esxcli software vib install -v https://cdn.tinkertry.com/files/net-ixgbe_4.5.1-1OEM.600.0.0.2494585.vib –no-sig-check
We then reboot our host to get the new VIB to be loaded.
Low and behold on reboot we see the two 10gbe NICs.
If you are not familiar with v-front.de and you have a ESXi servers, you are really doing yourself a dis-service. v-front.de maintains a repository of all ESXi updates and makes it very easy for you to update your servers when the newest patches come out.
Lately I have been having a series of interested GUI crashes with ESXi no matter what OS or Browser I use. Knowing I was a bit behind on patches, I decided to update.
From their patching site I was able to grab the latest software profile and install it. The steps are easy to do this and I will detail it here.
- Enable the SSH Shell for your ESXi Host. At the Host Tab, click Actions -> Services – Enable Secure Shell (SSH).
- Going to the patching site and clicking on the latest update, v-front.de lists us the steps how to update our host.
esxcli network firewall ruleset set -e true -r httpClient
esxcli software profile update -p ESXi-6.5.0-20170404001-standard
esxcli network firewall ruleset set -e false -r httpClient
- Reboot and then verify that the vibs have updated
[root@esxi:~] esxcli software vib list | grep esx-base
esx-base 6.5.0-0.19.5310538 VMware VMwareCertified 2017-05-13
[root@esxi:~] esxcli software vib list | grep esx-ui
esx-ui 1.18.0-5270848 VMware VMwareCertified 2017-05-13
[root@esxi:~] esxcli software vib list | grep vsan
vsan 6.5.0-0.19.5310540 VMware VMwareCertified 2017-05-13
vsanhealth 6.5.0-0.19.5310541 VMware VMwareCertified 2017-05-13
Another VMware basics post and this time I will teach you how to set up Swap for an ESXi Host. The Swap here is for the VMKernel, and basically can create performance boost for those hosts that are heavily utilized. For myself, I created a RAID 1 Volume of two Intel DC3710 800GB SSDs to for this.
The steps to do this are easy. First create a datastore on your volume of choice. It is best to use a low latency SSD that is local to the hosts.
Then we go to Host-> Manager and under the System Tab select Swap.
Then we click Edit Settings, and change the Data Store to the one we just created.
That’s it, now we have the Cache there.
If we browse to the DataStore we can see there is a swap file created.
I have been in the process of rebuilding my Plex Server and restoring the data. I popped in four 4TB WD Red NAS Drives and attached them to my 12G Controller. I figured it would be easy to create a datastore across the Raid 5 Drive and contine on my way. Unfortunately that was not the case, and I kept recieving errors similar to the one seen below.
While rather annoying, what I believe happens is that the Drives had a previous partition on them that ESXi just can not read or write to. So what we have to do is do some configuration in the ESXi Shell.
First we need to enable SSH on this host, which is very simple. At the Host Tab, click Actions -> Services – Enable Secure Shell (SSH). ESXi will enable the service and pop up a reminder warning for you.
Next we SSH into our ESXi Host.
We then need to determine the Disk ID of the Device we wish to fix. For us this is easy, as ESXi appends this to the name of the Device under Storage ->Devices.
From there we go back to the Shell and cd to /dev/disks where we should see a bunch of disk IDs.
Then we run partedUtil mklabel /dev/disks/diskid msdos
Then we run through the steps to create the new Datastore and after a bit of a wait we should see the new Datastore created.
So there we have it, we have our Plex Datastore and I can create my new disk and start sharing my Media much to my fiancee’s enjoyment. She has really missed her Power Rangers. Also don’t forget to turn off SSH access for safety’s sake.
Another basics posts for those of us who have not had the experience of installing and configuring ESXi 6.5. Today we will configure our Datastores.
Once we have logged into our ESXi client web page, click on Storage and the Datastore Tab
The initial Datastore is the media that you installed your ESXi “OS” on. For me it was a 64GB SATADOM. I do not wish to place any VMs on this, but I do have a NVMe Drive installed in the machine that I want my vSphere and vRealize components to live on. The thinking is if the NVMe drive fails, I will have backups of those VMs and be able to purchase and install a new drive, and then restore the VMs.
To setup the new Datastore, click New datastore and select Create new VMFS datastore and Click Next
Then we would select our device to install the Datastore on and continue on our merry way. Wait though, does anyone see the NVMe drive? Why isn’t it there?
Now it’s possible that this is a quirk of the new Install or its a quirk of the HTML5 Client. I have noticed at least with vSphere in my job that you still have to fall back to the “Fat” Client from time to time. In this case we do not.
If we click on the Devices Tab and highlight the NVMe drive, we get the option to select New Datastore
When we click that we get the Option to Name Our Datastore, Partition it, and then complete the steps. It seems it is just a shortened version of the Datastore Creation Wizard, and that the Storage Datastore Creation Wizard doesn’t have the necessary code to recognize the NVMe Drive.
So first we name the drive.
Then we Partition it utilizing the entire drive
Accept the Warning and we are complete.
Going back to the Datastore Tab we see the two stores.
With my brand new VMware ESXi Installation, it takes some time to configure everything to your liking and to best practices. ESXi 6.5 is not shy about warning you when you do not have redundancy or something configured as it should.
With my setup I have two 1Gbe NICs but ESXi only configured one NIC as the second was not active. Now that it is cabled and active, lets add it to the Virtual Switch.
At the main ESXi home page, click on Networking and verify that each NIC you wish to use shows as Enabled and has a Link Speed correct for the Network Adapter
For my installation the only two NICs we worry about are the first two, as vmnic2 corresponds to a 10Gbe Mellanox Card that has yet to be configured.
If we click on vSwitch0 we notice that VMWare has a warning!
If we click Actions we can Choose “Add Uplink”
We can see that VMware has added vmnic1. Now wasn’t that simple! No digging for the correct setting, just nice and easy.
Once we save and refresh the view, we see the Warning is gone and that we now have the two adapters configured for the Switch.
If we look at the Monitor tab under the VMNetwork we can see in the Events that the warning has cleared and why.
For those of us familiar with the “FAT” Client that ESXi used to have, this will not be anything new. For someone that is new to VMware this might take them a while to figure out.
I have a brand new ESXi 6.5 install that is using the Free License. Luckily I am a VMUG user and have access to a full license.
The setup is very simple, access your ESXi HTML Web Console. For me I was greeted with a warning that the License would expire in 60 days.
To license the host, it is very simple . Select Manage and then Licensing
Then you click Assign License and Copy and Paste your License Key into the Window. Click Check License and then Assign License
Then you can see that your License will change from Evaluation Mode to the License that you have purchased.