Jeramy Thompson

How to get llama 2 up and running , in a VM, with no GPU, and limited memory on Ubuntu

2023-07-20T16:10:00.004-06:00

OK I decided to write this up after unsuccessfully being able to find all the required info I needed in one place.

In this setup we will be using Ubuntu Server 22.04.2 LTS as the OS.

I have this running on a home lab ESXi server 8, on a HP Compaq Pro 6300 SFF

CPU = Intel Core i7-3770

Installed Memory 16 GB

I have some 10K SAS drives installed for the VM's

If you have not already, navigate to Get Ubuntu Server | Download | Ubuntu

and download the 22.04.2 LTS ISO

Next Lets create our VM that we are going to run this in.

*Note Im using ESXi however you can probably do this in Workstation, Fusion, VirtualBox etc

The main things to note on the VM creation.

Set Ubuntu 64 bit as the guest OS
Set your CPU relevant to the physicals CPU, I will be starting with 4 VCPU
Set your Memory to as much as you can tolerate, I will be using 12
Disk Space - we are creating a 100G swap file, and the rest of the file can take up some room , so more is better if you can afford it
Dont forget to add the Ubuntu ISO

Boot your new VM and begin the install

Update the installer when prompted

Set a static IP (recomended)

Use the Entire Disk

Verify the ubuntu-vg volume

And install OpenSSH (not required but makes life easier)

Finish Installing, remove the media (ISO) and reboot

To get access to llama 2, you need to head over to Llama 2 - Meta AI

After Clicking the download link, fill out the form, and you will get an Email URL.

You will use this when its time to download the models

Next SSH into your server

This is where allot of tinkering and finding the right info helped.

This site Running llama.cpp on Linux: A CPU and NVIDIA GPU Guide - Kubito , is a huge help for this and we will be following it mostly.

The llama.cpp allows us to use just CPU processing.

First lets make sure we are up to date

$ sudo apt-get update

$ sudo apt-get upgrade

I have no idea why this happens but for some reason ubuntu does not give you all the space int the volume so you have to extend it.

Run the following

$df -h

If you see the folume only giving 98GB then you have to extend it

/dev/mapper/ubuntu--vg-ubuntu--lv 98G 21G 73G 22% /

$ sudo lvextend -l +100%FREE -r /dev/ubuntu-vg/ubuntu-lv

$df -h

/dev/mapper/ubuntu--vg-ubuntu--lv 391G 21G 353G 6%

Now cd to your home directory (or where ever you want to put the application) and download llama.cpp

$ git clone https://github.com/ggerganov/llama.cpp

$ git clone https://github.com/facebookresearch/llama.git

$ cd llama.cpp

$ sudo apt install make

$ wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh

$ chmod +x Miniconda3-latest-Linux-x86_64.sh

$ ./Miniconda3-latest-Linux-x86_64.sh

Accept the terms , and use the defaults to install miniconda

say yes when prompted to initialize miniconda

Exit the SSH session and reconnect, you should now see a (base) infront of your use name.

$ cd llama.cpp

$ sudo apt install build-essential

$ make

$ conda create -n llama-cpp python=3.10.9

Say yes to proceed

$ conda activate llama-cpp

Now your command should show (llama-cpp)

ok Now lets download some models!

$ cd ../llama

$ chmod +x download.sh

$ ./download.sh

Here is where you enter the URL you got earlier from META

If you have a massive amount of storage , just get all of them.

I do not so I will be selecting 7B-chat to start (its about 12.55 GB)

Take note of where it is saving it to as we will need this info later.

Once downloaded, lets get the swap file created

#check your current swap size

$ free -h

#turn off your current swap

$ sudo swapoff -a

#increase swap to 100GB to be able to offload the entire model from RAM to disk

$ sudo fallocate -l 100G /swapfile

#make sure swapfile permissions are set, then activate

$ sudo chmod 600 /swapfile

$ sudo mkswap /swapfile

$ sudo swapon /swapfile

#check new swap size (should say something like 97Gi)

$ free -h

now cd back into the llama.cpp directory

$ python3 -m pip install -r requirements.txt

And we are going to convert the model we just downloaded

$ python3 convert.py /home/jeramy/llama/llama-2-7b-chat/

This will take some time

Once it completes , it should show you where it wrote the .bin file. We will next quantize that file

Here we will target the file that was created , and create a new one:

$ ./quantize /home/jeramy/llama/llama-2-7b-chat/ggml-model-f32.bin /home/jeramy/llama/llama-2-7b-chat/ggml-model-q4_0.bin q4_0

That should be it, you should be able to run the model with the new file you created:

$ ./main --color --interactive --model /home/jeramy/llama/llama-2-7b-chat/ggml-model-q4_0.bin --n-predict 512 --repeat_penalty 1.0 --reverse-prompt "User:" --in-prefix " " -f prompts/chat-with-bob.txt

And here are some stats from top:

vCenter Services Dependency tree

2022-10-26T10:43:00.002-06:00

The other day I needed a list of dependency's for the vCenter services.

Found out you can do this with service-control --list-dependencies <service name>

Version 7.0.3U3h

So I started with service-control --status --all, created a list.

I started to do this one by one, but of course its more fun to script it so I put all of them and ran a FOR LOOP on it

for SERV in $(cat ./serv.txt); do echo $SERV "--" && service-control --list-dependencies $SERV; done >> ServiceDepend.txt

applmgmt --

statsmonitor

lookupsvc --

rhttpproxy

vmware-vpostgres

lwsmd --

observability --

observability-vapi --

lookupsvc

vpxd-svcs

pschealth --

vpxd

vlcm --

lookupsvc

vmware-vpostgres

vpxd

vmafdd --

lwsmd

vmcad --

vmafdd

vmdird

vmdird --

vmafdd

vmonapi --

lookupsvc

sts

vmware-analytics --

envoy

lookupsvc

vmafdd

vpxd

vmware-certificateauthority --

lookupsvc

sts

vpxd-svcs

vmware-certificatemanagement --

lookupsvc

sts

vpxd-svcs

vmware-cis-license --

envoy

lookupsvc

sts

vmware-content-library --

rhttpproxy

sca

vapi-endpoint

vmware-vpostgres

vpxd

vpxd-svcs

wcp

vmware-eam --

vmware-envoy --

vmware-hvc --

lookupsvc

vmon

vmware-certificatemanagement

vmware-stsd

vmware-trustmanagement

vmware-vapi-endpoint

vmware-vpostgres

vpxd-svcs

vmware-infraprofile --

lookupsvc

vpxd-svcs

vmware-perfcharts --

sca

vpxd

vsphere-ui

vmware-pod --

vmware-postgres-archiver --

vmware-vpostgres

vmware-rhttpproxy --

vmware-sca --

lookupsvc

rhttpproxy

vmware-sps --

vpxd

vpxd-svcs

vmware-statsmonitor --

vmware-stsd --

lookupsvc

vmware-topologysvc --

lookupsvc

sts

vpxd-svcs

vmware-trustmanagement --

lookupsvc

vmware-updatemgr --

vpxd

vmware-vapi-endpoint --

envoy

lookupsvc

vmware-vdtc --

vmware-vmon --

vmafdd

vmware-vpostgres --

vmware-vpxd --

lookupsvc

vmware-vpostgres

vpxd-svcs

vmware-vpxd-svcs --

lookupsvc

sts

vmware-vpostgres

vmware-vsan-health --

vmware-vpostgres

vpxd

vmware-vsm --

vpxd

vsphere-ui --

envoy

lookupsvc

rhttpproxy

sts

vpxd-svcs

vstats --

lookupsvc

vmware-vpostgres

vpxd

vpxd-svcs

vtsdb

vtsdb --

vmware-vpostgres

wcp --

cis-license

eam

lookupsvc

sts

trustmanagement

vapi-endpoint

vmware-vpostgres

vpxd

vpxd-svcs

vmcam --

vpxd

vmware-imagebuilder --

vpxd

vmware-netdumper --

vmware-rbd-watchdog --

vpxd

vmware-vcha --

Upgrading to vSphere 8 on unsupported hardware

2022-10-20T09:55:00.003-06:00

So im still running unsupported hardware, but perhaps this workaround will let me hold on to my lab hardware just a couple more years longer.

Enable SSH into the ESXi host

This is what i did that so far is working

vi /bootbank/boot.cfg

**add in allowLegacyCPU=true to kernelopt

vi /altbootbank/boot.cfg

**add in allowLegacyCPU=true to kernelopt

esxcli software vib remove -n synology-nfs-vaai-plugin

** this vib comes up in the pre-check so it had to be removed

reboot

cat /bootbank/boot.cfg

cat /altbootbank/boot.cfg

**Make sure the allowLegacyCPU=true is still there

esxcli software profile update -p ESXi-8.0.0-20513097-standard -d /vmfs/volumes/340ffc1d-98386c9a/VMware/vSphere/8.x/VMware-ESXi-8.0-20513097-depot.zip --no-hardware-warning

cat /bootbank/boot.cfg

cat /altbootbank/boot.cfg

**Make sure the allowLegacyCPU=true is still there

reboot

cat /bootbank/boot.cfg

cat /altbootbank/boot.cfg

**Make sure the allowLegacyCPU=true is still there

reboot

**reboot one more time just to be sure

VMSA-2022-0024 CVE-2022-31676 How to update your VMware tools 12.1 and beyond

2022-08-23T12:21:00.010-06:00

With VMSA-2022-0024 just coming out, figured Id do a quick post on how to update your VMware tools.

First open up lifecycle Manager, and click Action 'Sync Updates':

If you VCSA does not have internet access or the depo is not updated , you can manually import updates (as I had to).

Create (or edit ) a new Base Line-

Select Patch

Select VMware as the Vendor

Any , Any, Any

Release date After 8/1/2022

Click Matched and make sure they are selected

Click Next, Next Finish

Attach base line to your ESXi host, and check compliance

Remediate as soon as possible (no reboot required).

Once All ESXi host are updated, you can then right click the VM and upgrade VMware tools.

Optionally if you have them set to auto update tools on reboot then simply shutdown and power back on.

Configure windows (and MAC) to point to a specific DNS server for Specific domain

2022-08-05T09:29:00.004-06:00

I have probably searched off an on for this solution for about 10 years at this point, and always found some other way to work around the issue.

Well today I was searching for some other DNS settings and found this:

https://docs.microsoft.com/en-us/powershell/module/dnsclient/add-dnsclientnrptrule?view=windowsserver2022-ps

And it actually works!

So here is my example:

Add-DnsClientNrptRule -Namespace ".universe.lab" -NameServers "10.10.1.1"

Ran this as an admin

I want any thing going to 'universe.lab' (dont forget to pre-pend the "." ) to go to 10.10.1.1

And then you can check what you have configured

PS C:\WINDOWS\system32> get-DnsClientNrptRule

Name : {2942DF55-16A3-4EED-B71B-7274B18F4FED}

Version : 2

Namespace : {.universe.lab}

IPsecCARestriction :

DirectAccessDnsServers :

DirectAccessEnabled : False

DirectAccessProxyType :

DirectAccessProxyName :

DirectAccessQueryIPsecEncryption :

DirectAccessQueryIPsecRequired :

NameServers : 10.10.1.1

DnsSecEnabled : False

DnsSecQueryIPsecEncryption :

DnsSecQueryIPsecRequired :

DnsSecValidationRequired :

NameEncoding : Disable

DisplayName :

Comment :

Fantastic right?!

And just to add in , the command for macOS is below just change the 10.10.1.1 to your DNS server and change universe.lab to your specific domain.

sudo cat 'nameserver 10.10.1.1' > /etc/resolver/universe.lab

VCSA Partition size by Storage Selection Size

2022-06-24T08:52:00.002-06:00

Ran into an issue upgrading a VCSA where it ran out of storage in the /storage/seat partition.

Figured it would be easy enough to select a larger options during deployment, however I wanted to know what my options were as I didn't want to go over kill on the partition.

Interestingly I was only able to find info on a Tiny deployment, no one seams to have any additional info on the additional deployment sizes.

So I just deployed the appliance over and over again and here are the results:

XL Storage Size

Large Storage Size

Default

So lets create a matrix:

Mounted	Default	Large	Extra Large
/storage/core	25GB	25GB	25GB
/storage/log	9.8GB	9.8GB	9.8GB
/storage/db	9.8GB	9.8GB	9.8GB
/storage/dblog	15GB	15GB	15GB
/storage/seat	9.8GB	541GB	1.4TB
/storage/netdump	985MB	985MB	985MB
/storage/autodeploy	9.8GB	9.8GB	9.8GB
/storage/imagebuilder	9.8GB	9.8GB	9.8GB
/storage/updatemgr	98GB	98GB	98GB
/storage/archive	49GB	49GB	49GB
/storage/vtsdb	9.8GB	541GB	1.4TB
/storage/vtsdblog	4.9GB	15GB	15GB
/storage/lifecycle	98GB	98GB	98GB
/storage/lvm_snapshot	174GB	492GB	984GB

Upgrade VMware tools to the latest version , automatically

2022-06-16T10:13:00.001-06:00

Wanted to put out a quick post about updating VMware tools to the latest version, in this case 12.0.5 on all VMs in an environment.

For vSphere 7.x this is really easy.

First download the "VMware Tools Offline VIB Bundle".

https://customerconnect.vmware.com/en/downloads/details?downloadGroup=VMTOOLS1205&productId=1259&rPId=88838

Upload this to your vSphere Lifecycle Manager by clicking on three bars on the left, then in the Lifecycle Manager page, click Actions > Import Updates

Select the file and it will upload

This should add it to the Non-Critical Patches (Predefined) baseline.

If not you can create a new baseline and just add in the tools package

Attach the redefined or custom baseline to your ESXi host and remediate the host , the tools update did not require a reboot.

Once all the hosts are updated, you can set your VMs to automatically upgrade on next power on \ reboot

https://docs.vmware.com/en/VMware-vSphere/7.0/com.vmware.vsphere-lifecycle-manager.doc/GUID-AA1EAFA8-E386-48D6-9F7B-D06399742AAA.html

For the Windows VMs, I noticed that once they reboot / power on, they upgrade tools , vsphere will show this as in progress, then it will reboot again

Once rebooted, you are good to go.

vSphere 7.0 U3c released

2022-01-27T20:19:00.023-07:00

vSphere U3c finally released.

TL;DR

https://docs.vmware.com/en/VMware-vSphere/7.0/rn/vsphere-vcenter-server-70u3c-release-notes.html

Here is my lab process:

ESXi Host

I am not using vLCM as these hosts are unsupported.

Well maybe not to night , Ill check again in a little while.

Dates appear to be a bit off:

Wow... alot of updates

Here we go.......

Well... That was un event full

No option to update vCenter... guess ill check tomorrow.

Hmm nothing this morning...

Nothing found in the logs

2022-01-28T13:07:47 PM UTC [2136]DEBUG:vmware.appliance.update.update_functions:runCommandAndCheckResult failed: '--2022-01-28 13:07:47-- https://vapp-updates.vmware.com/vai-catalog/valm/vmw/8d167796-34d5-4899-

be0a-6daade4005a3/7.0.3.00200.latest/manifest/manifest-latest.xml\nResolving vapp-updates.vmware.com... 69.192.208.24\nConnecting to vapp-updates.vmware.com|69.192.208.24|:443... connected.\nHTTP request sent, a

waiting response... \n HTTP/1.1 404 Not Found\n Accept-Ranges: bytes\n Content-Length: 10\n Server: AkamaiNetStorage\n Date: Fri, 28 Jan 2022 13:07:47 GMT\n Connection: keep-alive\n2022-01-28 13:07:47 ERRO

R 404: Not Found.\n\n'

2022-01-28T13:07:47 PM UTC [2136]ERROR:vmware.appliance.update.update_b2b:

Got exception while trying discover at URL https://vapp-updates.vmware.com/vai-catalog/valm/vmw/8d167796-34d5-4899-be0a-6daade4005a3/7.0.3.00200.latest: NotFound(messages=[{'id': 'com.vmware.appliance.update.inv

alid_url', 'default_message': 'Check the URL and try again.', 'args': ['']}], data=None, error_type='NOT_FOUND') 'Traceback (most recent call last):\n File "/usr/lib/applmgmt/update/py/vmware/appliance/update/u

pdate_b2b.py", line 1513, in _discoverUpdateAt\n tempFolder)\n File "/usr/lib/applmgmt/update/py/vmware/appliance/update/update_functions.py", line 581, in wgetWrapper\n "Certificate error at target URL")

)})\n File "/usr/lib/applmgmt/update/py/vmware/appliance/update/update_functions.py", line 373, in runCommandAndCheckResult\n raise exception\nvmware.appliance.update.update_functions.LocalizableException: {

\'id\': \'com.vmware.appliance.update.download_failed\', \'default_message\': \'Download failed\', \'args\': []}\n\nDuring handling of the above exception, another exception occurred:\n\nTraceback (most recent c

all last):\n File "/usr/lib/applmgmt/update/py/vmware/appliance/update/update_b2b.py", line 1776, in processURLUpdates\n header = _discoverUpdateAtUrl(url,\'latest\')\n File "/usr/lib/applmgmt/update/py/vmw

are/appliance/update/update_b2b.py", line 1701, in _discoverUpdateAtUrl\n versionFolder)\n File "/usr/lib/applmgmt/update/py/vmware/appliance/update/update_b2b.py", line 1519, in _discoverUpdateAt\n vapiN

otFound(messageInvalidUrl(\'\'))\n File "/usr/lib/applmgmt/update/py/vmware/appliance/update/update_functions.py", line 166, in vapiNotFound\n raise ErrorFactory.new_not_found(messages=messages)\ncom.vmware.

vapi.std.errors_provider.NotFound: {messages : [{\'id\': \'com.vmware.appliance.update.invalid_url\', \'default_message\': \'Check the URL and try again.\', \'args\': [\'\']}], data : None, error_type : NOT_FOUN

D}\n'

Guess we will download the update and do it manually

7GB!

While thats downloading, gives me time to power down the VC, and take a powered off Snapshot.

Ok well it finds it on the CD ROM:

And here we go:

Just waiting at 'Converting data as part of post install'....

Looks like a post-patch.sh script is running:

2022-01-28 14:03:55,828.828Z last_component:Patch INFO vmware_b2b.patching.utils.reporting_utils Setting global ReportingFactory with identifier - last_component:Patch

2022-01-28 14:03:55,830.830Z last_component:Patch INFO extensions Found patch hook <module 'last_component' from '/storage/updatemgr/software-update8graktlm/stage/scripts/patches/payload/components-script/last_component/__init__.py'>:Patch'

2022-01-28 14:03:55,830.830Z last_component:Patch INFO extensions Executing patch hook '<module 'last_component' from '/storage/updatemgr/software-update8graktlm/stage/scripts/patches/payload/components-script/last_component/__init__.py'>:Patch' with context <patch_specs.PatchContext object at 0x7f2c14f63c10>.

2022-01-28 14:03:55,831.831Z last_component:Patch INFO last_component Running post-patch script /storage/updatemgr/software-update8graktlm/stage/scripts/post-patch.

No updates in the UI, but progress in the logs:

2022-01-28 14:06:19,233.233Z last_component:Patch INFO last_component Starting all VMware services... The immediate command stdout is redirected to file /var/log/vmware/applmgmt/startAllServices.out.log and stderr is redirected to file /var/log/vmware/applmgmt/startAllServices.err.log

No UI changes... but movement in logs:

So in my case about 41 minutes install time...waiting for services to startup now

All up and running.

So Lets check a Nested environment.

Nope, no option for U3c

CVE-2021-44228

2021-12-11T06:42:00.002-07:00

Looks like VMware is already on it

https://www.vmware.com/security/advisories/VMSA-2021-0028.html

vSphere 7.0 U3b Released - and then removed

2021-11-17T12:22:00.003-07:00

Update --

https://kb.vmware.com/s/article/86398

On 11/12 VMware released 7.0 U3B for bot ESXi and vCenter

However vCenter appears to have been pulled as its completely missing

https://docs.vmware.com/en/VMware-vSphere/7.0/rn/vsphere-vcenter-server-70u3b-release-notes.html

https://docs.vmware.com/en/VMware-vSphere/7.0/rn/vsphere-vcenter-server-70u3a-release-notes.html

This appears to be an issue with HA:

https://kb.vmware.com/s/article/86191

IMPORTANT NOTE: vCenter Server 7.0 U3b is no longer available for download due to further complications with HA configuration post-upgrade. If a vCenter has already been upgraded to vCenter 7.0 U3b(or earlier) and is experiencing the symptoms mentioned above, please follow the steps in the Workaround section below.

Synology NAS performance issue after updating to DSM 7

2021-10-29T06:28:00.010-06:00

Looks like Synology had a significant oversight when they allowed customers of the DS416Play to upgrade to the latest DSM 7 OS.

The default configuration comes with 1GB of memory, this was adequate for the version 6 DSM, however it appears the version 7 clearly has a larger memory footprint and does not leave enough room for file operations.

In my case I purchased a 8GB stick of memory:

https://www.amazon.com/G-Skill-Ripjaws-204-Pin-SO-DIMM-F3-1600C9S-8GRSL/dp/B00EOUFEG2

Or This one has been reported to work as well:

https://www.amazon.com/Crucial-Single-PC3-12800-Unbuffered-204-Pin/dp/B006YG8X9Y

Followed this YouTube tutorial on disassembling the NAS

https://www.youtube.com/watch?v=zE6P7F16X7Q

And booted back up. It now runs like it did previously and my backups function again.

More info here.

https://community.synology.com/enu/forum/1/post/146281?page=1&reply=463033

I wish i would have taken a screen shot before the upgrade, but here is TOP after the upgrade:

Unifi and UNMS \ UISP on the same ubuntu server

2021-10-07T11:08:00.002-06:00

Needed a place to put my notes for installing Unifi an UNMS on an ubuntu server.

Install Ubuntu with only openssh

SSH in

Setup the Java 8 environment that is required and remove default java:

--------------------------------------------------------------------

sudo apt-get autoremove default-jdk openjdk-11-jdk

sudo apt install openjdk-8-jdk

export JAVA_HOME=/usr/lib/jvm/java-8-openjdk-amd64

--------------------------------------------------------------------

Next lets get some updates, and add ui to the apt sources

--------------------------------------------------------------------

sudo apt-get update && sudo apt-get install ca-certificates apt-transport-https

echo 'deb https://www.ui.com/downloads/unifi/debian stable ubiquiti' | sudo tee /etc/apt/sources.list.d/100-ubnt-unifi.list

sudo apt-key adv --keyserver keyserver.ubuntu.com --recv 06E85760C0A52C50

--------------------------------------------------------------------

Put a hold on openjdk-11 files

--------------------------------------------------------------------

sudo apt-mark hold openjdk-11-*

--------------------------------------------------------------------

Install Unifi

--------------------------------------------------------------------

sudo apt-get update && sudo apt-get install unifi -y

--------------------------------------------------------------------

And Start the unifi service and check the status

--------------------------------------------------------------------

sudo service unifi restart

sudo service unifi status

--------------------------------------------------------------------

Lastly install unms via script

--------------------------------------------------------------------

curl -fsSL https://unms.com/v1/install > /tmp/unms_inst.sh && sudo bash /tmp/unms_inst.sh

--------------------------------------------------------------------

Windows 11 GA Instant Clone Lab with unsupported hardware workaround \ test deploy

2021-10-06T15:07:00.002-06:00

Windows 11 has finally GA'ed so now its time to do some testing.

If you don't have the required hardware no problem, we can apply some work arounds.

In our Setup we will be using the following Setup

Horizon 8

vSphere 7.0U3

Windows 11 GA

First download the Windows 11 ISO, put it on a datastore

Create a new VM for the parent VM.

I chose Windows 10 as the Guest OS

Confirm that Secure Boot is enabled

And left everything as Default

Now Console into your VM and power it on, Once you get to the 'Windows Setup' screen press Shift +F10

This will bring up a command prompt, from here type regedit and hit enter

From this point navigate to

HKEY_LOCAL_MACHINE\SYSTEM\Setup

Right-click on Setup key and select New -> Key option.

Name it LabConfig

Select LabConfig key and then create the following 3 new DWORDs and set their values to 1 to bypass or disable TPM 2.0, Secure Boot and RAM amount checks

BypassTPMCheck

BypassSecureBootCheck

BypassRAMCheck

Close regedit, and the command prompt and continue with the install as usuall.

Once this completes you should be ready to start configuring the VM.

After you rename your PC it does a reboot

Make sure you select Sign-In Options (unless you want to use a MS account or SSO)

And Choose Domain join instead

I Just choose the default here

At this point it will reboot and auto login the user

At this point you could just use the admin account and install Horizon.

I prefer to use the Administrator account, delete the newly created account before deploying

Enable it and set a password, then reboot

Like This

After a reboot you should be able to delete it

At this point Eject the Windows ISO if you haven't already and install VMware tools

Reboot Shutdown Take a snapshot

Power on and install the agent

After installing the agent and rebooting, you are now required to use CTL ALT DEL to log in

Shutdown and take a new snapshot

And thats it, you IC should deploy and you can connect with the client.

vSphere \ vCenter 7.0 Update 3 7.0U3 - 7.0.3.00000 - Authentication Failed, Lifecycle Manager server could not be contacted

2021-10-06T05:45:00.014-06:00

-----------Update 3------------

And it looks like they are working on it:

https://kb.vmware.com/s/article/85962

---------------------------------------------------

Upgraded last night So far Lifecycle manager appears not to be working.

"Authentication Failed, Lifecycle Manager server could not be contacted"

A quick look at the logs says it thinks im not authenticated

[#18] 2021-10-06T11:38:51.077Z warning vmware-vum-server[13587] [Originator@6876 sub=SimpleConnection] [simpleConnection 85] Connection requested for ><, returning default connection

[#18] 2021-10-06T11:38:51.077Z info vmware-vum-server[13587] [Originator@6876 sub=VcIntegrity] Session: [sessionId masked]; user jthompson@HOME.LAB

[#18] 2021-10-06T11:38:51.079Z info vmware-vum-server[13587] [Originator@6876 sub=VcIntegrity] Active? false

[#18] 2021-10-06T11:38:51.079Z error vmware-vum-server[13587] [Originator@6876 sub=SessionMgr] [sessionMgr 189] Login: User's VC session is not active; User: jthompson@HOME.LAB; VcSession id: [52a0acb4-5bc9-f0fc-69ba-06d9838561ca]

[#18] 2021-10-06T11:38:51.081Z error vmware-vum-server[13587] [Originator@6876 sub=SessionMgr] [sessionMgr 189] [backtrace begin] product: VMware Update Manager, version: 7.0.3, build: build-18649303, tag: vmware-vum-server, cpu: x86_64, os: linux, buildType: release

[#18] --> backtrace[00] libvmacore.so[0x0037E963]

[#18] --> backtrace[01] libvmacore.so[0x002C8917]: Vmacore::System::Stacktrace::CaptureFullWork(unsigned int)

[#18] --> backtrace[02] libvmacore.so[0x002D7AB7]: Vmacore::System::SystemFactory::CreateBacktrace(Vmacore::Ref<Vmacore::System::Backtrace>&)

[#18] --> backtrace[03] libvci-vcIntegrity.so[0x00786BC7]: Integrity::SessionMgrImpl::Login(std::string const&, std::string const&, Vmacore::Optional<std::string> const&, Vmacore::Ref<Integrity::UserSession>&)

[#18] --> backtrace[04] libintegrity-types.so[0x001CF878]

[#18] --> backtrace[05] libvmomi.so[0x0013F1AE]

[#18] --> backtrace[06] libvmomi.so[0x0013F80B]

[#18] --> backtrace[07] libvmacore.so[0x00235A97]

[#18] --> backtrace[08] libvmacore.so[0x0023B03F]

[#18] --> backtrace[09] libvmacore.so[0x00377398]

[#18] --> backtrace[10] libpthread.so.0[0x00007F87]

[#18] --> backtrace[11] libc.so.6[0x000F35EF]

[#18] --> backtrace[12] (no module)

[#18] --> [backtrace end]

-------------------------Update-------------------

Administrator@vsphere.local appears to have no issues

------------Update 2----------------

Tried clearing browser cache... no dice

Checked permissions, the users is part of administrators, dont see any new groups added.

Im also using windows Integrated auth, should not be an issue, but it is being depreciated.

When in doubt ,reboot!

Lets see where this takes us

No Joy after reboot

Step By Step upgrade to VCSA 7.0 Update 2 D (7.0U2d)

2021-09-22T09:20:00.003-06:00

Found this link that is a simple walkthrough through the upgrade

https://communities.vmware.com/t5/vSphere-Upgrade-Install/Step-by-step-procedure-to-update-vCenter-sever-7-0-appliance/ta-p/2848957

bypass the password already used in vCenter appliance (and most linux distros)

2021-09-01T09:01:00.005-06:00

It gets quite annoying when you need to deploy a lab and need to keep the same password (or your just too lazy to change\ remember a new one).

to clear the password history in a VC appliance, simply type:

echo “” > /etc/security/opasswd

this will clear your password history , now you can change your password to any previous password you had before.

Sleep Progress Bar for Powershell \ Power-CLI scripts function

2021-08-27T07:53:00.003-06:00

I used this to update my lab power shell scripts so that during the sleep cycle , I have some indication of where we are at. You can read about these scrpits here https://jeramythompson.blogspot.com/2021/08/powercli-script-creating-instant-clone.html https://jeramythompson.blogspot.com/2021/08/vmware-vsphere-vcenter-and-esxi-home.html

The End result will look like this

This is achivied by creating a simple function at the top of the script and calling it later:

----------Here is the Function -------------------

function Get-SleepTimer{

param (

[int]$Time

)

$Lenght = $Time / 100

For ($Time; $Time -gt 0; $Time--) {

$min = [int](([string]($Time/60)).split('.')[0])

$text = " " + $min + " minutes " + ($Time % 60) + " seconds left"

Write-Progress -Activity "Watiting for..." -Status $Text -PercentComplete ($Time / $Lenght)

Start-Sleep 1

}

-----------------End Function---------

And we can simply call this function and pass the number of seconds we want to wait

----------------calling the timer function--------------

Get-SleepTimer 600

----------------End calling the timer function--------------

Script was originally found here

VMware vSphere vCenter and ESXi Home Lab shutdown script with PowerCLI

2021-08-18T10:00:00.001-06:00

Some times you need to shut down your lab in as efficient way as possible.

I do this quite often as I don't like leaving my home lab running during various events (thunderstorms, weekends , vacation etc). And I really dont like having to click , wait, click, wait, click.... wait some more to get every thing down.

So I wrote this script to power down my lab, Ill walk you through what it does and the completed version will be at the end of this post. (I'm also including a start-up script I use as well)

First we need to import our power CLI modules and tell the script to ignore any self-signed or Invalid certificates ( If you want more info on setting up powerCli see the beginning of this post for more info)

-------------------------------------------------------------------

Import-Module VMware.PowerCLI

Set-PowerCLIConfiguration -InvalidCertificateAction Ignore -Confirm:$false

-------------------------------------------------------------------

Now we need to setup our credential variables, one will be for vCenter, and the other is the host that everything will be migrated to and then eventually shut down on. We are going to use a feature in power CLI that allows us to store our password so we don't have to have it in plain text in our script.

You should do this on the computer you will always run this script on.

For the ESXi host, we will use root.

From the command prompt run the following replacing the -Host with something you can identify as your host, and -Password with your password

New-VICredentialStoreItem -Host AlphaHost -User root -Password "SuperSecretP@$$w0rd"

Next lets store the password for your vcenter, I will be using administrator@vsphere.local as it does not require AD to be up and running.

New-VICredentialStoreItem -Host VCSA -User administrator@vsphere.local -Password "SuperSecretP@$$w0rd"

And now the lines we put in the script are:

-------------------------------------------------------------------

$VCSAPass = get-VICredentialStoreItem -Host VCSA -User administrator@vsphere.local

$rootPass = get-VICredentialStoreItem -Host AlphaHost -User root

-------------------------------------------------------------------

And now we connect to that vCenter

-------------------------------------------------------------------

connect-VIServer vcenter.home.lab -User 'administrator@vsphere.local' -Password $VCSAPass.Password

-------------------------------------------------------------------

The next step is we are going to create the vmservers variable that gets a list of all VMs that are powered on, except for our vcenter, domain controllers and the vCLS vms, and then shutdown the guest OS of the VM's. We do this as we need the DC for DNS resolution, and the vCLS vms will be powered off in a later step by vCenter (if they are powered off now, they just get auto started back up)

-------------------------------------------------------------------

$vmservers=Get-VM | Where-Object {$_.powerstate -eq ‘PoweredOn’ -and $_.name -ne 'vcenter.home.lab' -and $_.name -ne 'dc1.home.lab' -and $_.name -ne 'dc2.home.lab' -and $_.name -cnotlike 'vCLS*' }

$vmservers| Shutdown-VMGuest -Confirm:$false

-------------------------------------------------------------------

Now I have a 3 node cluster, so I will be putting 2/3 into maintenance mode before shutting them down. Since i have DRS enabled this will move the remaining VMs to a single host (makes startup script work as well)

-------------------------------------------------------------------

Get-VMHost -Name charlie.home.lab | set-vmhost -State Maintenance

Get-VMHost -Name bravo.home.lab | set-vmhost -State Maintenance

-------------------------------------------------------------------

Now that our hosts are in maintanence mode, we are going to set retreat mode for vCLS vms, this will cause vCenter to delete them(to find your domain-c<number> see the following VMware KB to get that info and replace it in the line below https://kb.vmware.com/s/article/80472):

-------------------------------------------------------------------

Get-AdvancedSetting -Entity vcenter.home.lab -Name 'config.vcls.clusters.domain-c26.enabled' | Set-AdvancedSetting -Value 'false' -Confirm:$false

start-sleep 60

-------------------------------------------------------------------

Now lets shut down those hosts

-------------------------------------------------------------------

Get-VMHost -Name charlie.home.lab | Stop-VMHost -Confirm:$false

Get-VMHost -Name bravo.home.lab | Stop-VMHost -Confirm:$false

-------------------------------------------------------------------

Our tasks for vCenter are done, we now disconnect from the vCenter , and connect directly to the last ESXi hosts that still has VMs running by IP:

-------------------------------------------------------------------

Disconnect-VIServer vcenter.home.lab -confirm:$false

connect-VIServer 192.168.0.252 -User 'root' -Password $rootPass.Password

-------------------------------------------------------------------

Now we power off the last set of VM's(dc's and vCenter should be all thats left):

-------------------------------------------------------------------

$morevmservers=Get-VM | Where-Object {$_.powerstate -eq ‘PoweredOn’}

$morevmservers| Shutdown-VMGuest -Confirm:$false

do { $vmstatus = (get-vm $morevmservers).PowerState ; Start-Sleep -Seconds 5} while ($vmstatus -eq "PoweredOn")

-------------------------------------------------------------------

And put this host into maintenance mode, then power it off and disconnect

-------------------------------------------------------------------

set-vmhost -State Maintenance

start-sleep 60

Stop-VMHost -Confirm:$false

Disconnect-VIServer 192.168.0.252 -confirm:$false

-------------------------------------------------------------------

Lastly I put in a final check to verify that all the ESXi hosts are no longer ping-able (powered off)

-------------------------------------------------------------------

#Check servers are down

do{}Until (!(Test-Connection 192.168.0.250 -Quiet -Count 1))

Write-Host -ForegroundColor Black -BackgroundColor Green "Host Charlie.Home.Lab Is Down"

do{}Until (!(Test-Connection 192.168.0.251 -Quiet -Count 1))

Write-Host -ForegroundColor Black -BackgroundColor Green "Host Bravo.Home.Lab Is Down"

do{}Until (!(Test-Connection 192.168.0.252 -Quiet -Count 1))

Write-Host -ForegroundColor Black -BackgroundColor Green "Host Alpha.Home.Lab Is Down"

-------------------------------------------------------------------

That's it, so as mentioned here is the full script:

#-------------------------------------------------------------------

#-------------------------vSphere Lap Power Off Script------

#--------------------------------------------------------------------

#------------By Jeramy Thompson------------------------------

#-----------https://jeramythompson.blogspot.com/----------

#-------------------------------------------------------------------

Import-Module VMware.PowerCLI

Set-PowerCLIConfiguration -InvalidCertificateAction Ignore -Confirm:$false

$VCSAPass = get-VICredentialStoreItem -Host VCSA -User administrator@vsphere.local

$rootPass = get-VICredentialStoreItem -Host AlphaHost -User root

connect-VIServer vcenter.home.lab -User 'administrator@vsphere.local' -Password $VCSAPass.Password

$vmservers=Get-VM | Where-Object {$_.powerstate -eq ‘PoweredOn’ -and $_.name -ne 'vcenter.home.lab' -and $_.name -ne 'dc1.home.lab' -and $_.name -ne 'dc2.home.lab' -and $_.name -cnotlike 'vCLS*' }

$vmservers| Shutdown-VMGuest -Confirm:$false

Get-VMHost -Name charlie.home.lab | set-vmhost -State Maintenance

Get-VMHost -Name bravo.home.lab | set-vmhost -State Maintenance

Get-AdvancedSetting -Entity vcenter.home.lab -Name 'config.vcls.clusters.domain-c26.enabled' | Set-AdvancedSetting -Value 'false' -Confirm:$false

start-sleep 60

Get-VMHost -Name charlie.home.lab | Stop-VMHost -Confirm:$false

Get-VMHost -Name bravo.home.lab | Stop-VMHost -Confirm:$false

Disconnect-VIServer vcenter.home.lab -confirm:$false

connect-VIServer 192.168.0.252 -User 'root' -Password $rootPass.Password

$morevmservers=Get-VM | Where-Object {$_.powerstate -eq ‘PoweredOn’}

$morevmservers| Shutdown-VMGuest -Confirm:$false

do { $vmstatus = (get-vm $morevmservers).PowerState ; Start-Sleep -Seconds 5} while ($vmstatus -eq "PoweredOn")

set-vmhost -State Maintenance

start-sleep 60

Stop-VMHost -Confirm:$false

Disconnect-VIServer 192.168.0.252 -confirm:$false

do{}Until (!(Test-Connection 192.168.0.250 -Quiet -Count 1))

Write-Host -ForegroundColor Black -BackgroundColor Green "Host Charlie.Home.Lab Is Down"

do{}Until (!(Test-Connection 192.168.0.251 -Quiet -Count 1))

Write-Host -ForegroundColor Black -BackgroundColor Green "Host Bravo.Home.Lab Is Down"

do{}Until (!(Test-Connection 192.168.0.252 -Quiet -Count 1))

Write-Host -ForegroundColor Black -BackgroundColor Green "Host Alpha.Home.Lab Is Down"

#-------------------------------------------------------------------

As Mentioned I also have a startup scrip. At this point its should be self explanitory.

#-------------------------------------------------------------------

#-------------------------vSphere Lap Power ON Script------

#--------------------------------------------------------------------

#------------By Jeramy Thompson------------------------------

#-----------https://jeramythompson.blogspot.com/----------

#-------------------------------------------------------------------

Import-Module VMware.PowerCLI

Set-PowerCLIConfiguration -InvalidCertificateAction Ignore -Confirm:$false

$VCSAPass = get-VICredentialStoreItem -Host VCSA -User administrator@vsphere.local

$rootPass = get-VICredentialStoreItem -Host AlphaHost -User root

#Connect To Alpha

connect-VIServer 192.168.0.252 -User 'root' -Password $rootPass.Password

#If its in Maintence mode, exit it:

set-vmhost -State connected

$StartServers=Get-VM | Where-Object {$_.name -eq 'vcenter.home.lab' -or $_.name -eq 'dc1.home.lab' -or $_.name -eq 'dc2.home.lab'}

$StartServers| Start-VM -Confirm:$false

start-sleep 60

Disconnect-VIServer 192.168.0.252 -confirm:$false

Start-Sleep 600

connect-VIServer vcenter.home.lab -User 'administrator@vsphere.local' -Password $VCSAPass.Password

Get-AdvancedSetting -Entity vcenter.home.lab -Name 'config.vcls.clusters.domain-c26.enabled' | Set-AdvancedSetting -Value 'true' -Confirm:$false

Get-VMHost -Name charlie.home.lab | set-vmhost -State Connected

Get-VMHost -Name bravo.home.lab | set-vmhost -State Connected

#Start the vApp that contains other essentials

Start-VApp -VApp Infra -Confirm:$false

Disconnect-VIServer vcenter.home.lab -confirm:$false

#-------------------------------------------------------------------

PowerCLI Script creating Instant Clone Pools in Horizon 8 Walk through

2021-08-05T13:10:00.003-06:00

One task I run into quite often is deploying Instant Clone pools to test one thing or another.

I dont always leave my lab running, so all the pools get deleted , and then recreated when needed.

If you have ever created an instant you know you have to click about 20 settings, while not a monumental task, it is a bit reparative.

So we are going to go through the process of creating our own script , so we only need to click once to deploy our instant clone pool.

You are going to need a couple of things:

Internet access

PowerCLI installed

PowerCLI Example Scripts

(Optional) A dedicated workstation to store your password on.

Ok lets start with installing PowerCLI.

Open power shell as an administrator

And run the following command to install PowerCLI

Install-Module VMware.PowerCLI

Type A and hit Enter

Wait for the progress bar to complete the install of the modules

Once Complete run the following to verify it was installed and get the Version number if you need it at some point

Get-Module -Name VMware.PowerCLI -ListAvailable

PowerCLI is now installed, that's it!

Next We need the example scripts zip file. The zip file includes a specific module for horizon that is not built into the module we just installed.

Open your browser to https://github.com/vmware/PowerCLI-Example-Scripts , click code, and download as Zip

Once this is downloaded, open the Zip, and we are going to extract the VMware.Hv.Helper folder from

the Modules directory.

You can extract this any where you like, however for our example we will Extract it to C:\Scripts for ease of use

At this point we are ready to write our script!

If you have your own powershell IDE then you can just use that, for this post we are going to use ISE.

Click start and search for powershell ISE:

You should now have a blank space to start

Right off the bat we are going to import our modules, PowerCLI , vmware.vimautomation.horizonview (part of PowerCLI), and the VMware.HV.Helper that we downloaded Earlier.

We will also tell PowerCLI to ignore Invalid certificates in case we are using Self Signed certs, we dont want it to sto for that

Import-Module VMware.PowerCLI

Import-Module vmware.vimautomation.horizonview

Import-Module -Name "C:\scripts\VMware.Hv.Helper"

Set-PowerCLIConfiguration -InvalidCertificateAction Ignore -Confirm:$false

Next we are going to setup our variables, this will make it easier if the say the Parent VM, or Snapshot change in the future.

#The Name of the Instant Clone Pool

$HZPool = "W11-IC"

#The display name of the Instant Clone Pool

$HZDisplay = "W11-IC"

#The description of the pool displayed in the horizon Admin

$HZDescription = "Windows 11 Insider Preview Instant Clone Pool"

#How are the users assigned to the desktops for this pool

$HZAssign = "FLOATING"

#What is the name of the parent VM

$HZParent = "W11-Insider-Instant-Clone-Parent"

#The Snapshot we need to use

$HZSnap = "UEM-2103 Installed"

#The folder you want to place the VMs in, I Always put my in the root and let it create the sub folder , so I use HomeDatacenter as its the default

$HZFolder = "HomeDatacenter"

#The name of your vSphere cluster you are deploying to

$HZCluster = "AMD"

#If you have resource pools specify this here, else use the name of your cluster

$HZResourcePool = "VDI"

#How are you naming your VMs?

$HZNamingMethod = "PATTERN"

#The naming pattern of your Instant Clones

$HZPattern = "W11-IC-{n}"

#The Datstore you will deploy the VMs to

$HZDatastore = "SYNFS2"

#Your instant clone admin domain you use to join the domain

$HZDomain = "HOME"

#you instant clone admin username

$HZICAdmin = "hzncomposer"

(Optional) Next we are going to use a feature in power CLI that allows us to store our password so we don't have to have it in plain text in our script.

From the command prompt run the following replacing the -Host with your host, -User with your Horizon Admin user and -Password with your password

New-VICredentialStoreItem -Host hacs1.home.lab -User ViewAutoBot -Password "SuperSecretP@$$w0rd"

I put the password in " as powershell uses $ so its the only way to pass these

This will complete any your password is now saved.

You can validate this by running the following and your password should be returned in plain text:

get-VICredentialStoreItem -Host hacs1.home.lab -User ViewAutoBot | Select-Object Password | foreach {$_.password}

So now our next line is to set a variable for our password:

$HZPass = get-VICredentialStoreItem -Host hacs1.home.lab -User ViewAutoBot | Select-Object Password | foreach {$_.password}

And our line to connect to the connection server utilizing the password

*Note if you did not setup the VICredentialStore, the replace the $HZPass variable with your password in plane text

Connect-HVServer -Server hacs1.home.lab -User ViewAutoBot@home.lab -Password $HZPass

Now lets take all of this data we just setup and get the command setup to use it (I formatted it this way for readability).

New-HVPool -InstantClone -PoolName $HZPool `

-PoolDisplayName $HZDisplay `

-Description $HZDescription `

-UserAssignment $HZAssign `

-ParentVM $HZParent `

-SnapshotVM $HZSnap `

-VmFolder $HZFolder `

-HostOrCluster $HZCluster `

-ResourcePool $HZResourcePool `

-NamingMethod $HZNamingMethod `

-NamingPatter $HZPattern `

-Datastores $HZDatastore `

-NetBiosName $HZDomain `

-DomainAdmin $HZICAdmin

2 more lines, the last thing that is done when deploying a pool is assigning users, because this script runs really fast, I have put a 60 sec sleep statement and then Assign the "Domain Users" to the pool.

Start-Sleep -s 60

New-HVEntitlement -Type Group -User "HOME\Domain Users" -ResourceName $HZPool -Confirm:$false

Thats All, now lest run it !

Now you can make a short cut on your desktop with the following to run it at the click of a button:

C:\Windows\System32\WindowsPowerShell\v1.0\powershell.exe -noexit -ExecutionPolicy Bypass -File "C:\Scripts\W10IC.ps1"

Here is the full script:

Import-Module VMware.PowerCLI

Import-Module vmware.vimautomation.horizonview

Import-Module -Name "C:\scripts\VMware.Hv.Helper"

Set-PowerCLIConfiguration -InvalidCertificateAction Ignore -Confirm:$false

#The Name of the Instant Clone Pool

$HZPool = "W11-IC"

#The display name of the Instant Clone Pool

$HZDisplay = "W11-IC"

#The description of the pool displayed in the horizon Admin

$HZDescription = "Windows 11 Insider Preview Instant Clone Pool"

#How are the users assigned to the desktops for this pool

$HZAssign = "FLOATING"

#What is the name of the parent VM

$HZParent = "W11-Insider-Instant-Clone-Parent"

#The Snapshot we need to use

$HZSnap = "UEM-2103 Installed"

#The folder you want to place the VMs in, I Always put my in the root and let it create the sub folder , so I use HomeDatacenter as its the default

$HZFolder = "HomeDatacenter"

#The name of your vSphere cluster you are deploying to

$HZCluster = "AMD"

#If you have resource pools specify this here, else use the name of your cluster

$HZResourcePool = "VDI"

#How are you naming your VMs?

$HZNamingMethod = "PATTERN"

#The naming pattern of your Instant Clones

$HZPattern = "W11-IC-{n}"

#The Datstore you will deploy the VMs to

$HZDatastore = "SYNFS2"

#Your instant clone admin domain you use to join the domain

$HZDomain = "HOME"

#you instant clone admin username

$HZICAdmin = "hzncomposer"

$HZPass = get-VICredentialStoreItem -Host hacs1.home.lab -User ViewAutoBot | Select-Object Password | foreach {$_.password}

Connect-HVServer -Server hacs1.home.lab -User ViewAutoBot@home.lab -Password $HZPass

New-HVPool -InstantClone -PoolName $HZPool `

-PoolDisplayName $HZDisplay `

-Description $HZDescription `

-UserAssignment $HZAssign `

-ParentVM $HZParent `

-SnapshotVM $HZSnap `

-VmFolder $HZFolder `

-HostOrCluster $HZCluster `

-ResourcePool $HZResourcePool `

-NamingMethod $HZNamingMethod `

-NamingPatter $HZPattern `

-Datastores $HZDatastore `

-NetBiosName $HZDomain `

-DomainAdmin $HZICAdmin

Start-Sleep -s 60

New-HVEntitlement -Type Group -User "HOME\Domain Users" -ResourceName $HZPool -Confirm:$false

Formatting Annotations.WelcomeMessage for the ESXi host so its not one long line 7.0

2021-07-29T13:41:00.022-06:00

Lets say your company needs a security banner on your ESXi host, but you need your text to go to the next line like the examples below:

Edit the settings, and place a /n (linux escape char) in the line you want to start new:

Now you have your text on a new line

vSphere 7 - Specify a vmkernel adapter to use for NFS datastores Step By Step \ Walkthrough on a DVS

2021-07-12T10:21:00.002-06:00

Recently I needed to setup a NFS server but needed it to specifically be on a specific vmkernel adapter.

I found this KB https://kb.vmware.com/s/article/50112854 however there were some things missing so I decided to write this walkthrough incase it helps any one else.

First thing , remove all NFS mounted datastores if you haven't already.

I ran through this on 7.0.2 however is probably the same for other versions.

Step 1:

This is going to be on a DVS, so we want to go ahead and create a port group to silo off our NFS traffic.

To make life easy, we will just call it NFS

Static binding and elastic port allocation were the defaults so I let these there. If you are VLAN'ing your traffic modify this info here, and complete the wizard.

Next put your host into Maintenance Mode

We are going to add a New TCP/IP stack to the host. First check that you have not already created one in the UI by opening a browser to the ESXi host, navigate to networking > TCP/IP Stacks

SSH into your ESXi host and run the following command to add a new netstack called nfs

localcli network ip netstack add --netstack=nfs

Run the next command to verify the new netstack exists and reboot (it does not show up in the UI untill a reboot has been done)

esxcli network ip netstack get --netstack=nfs

After the reboot you can check the UI and verify you now see the new TCP/IP stack

The next set of commands will stop and start services (if not already stopped) and bind NFS traffic to our newly created TCP\IP stack 'nfs'. reboot after running

/etc/init.d/nfsgssd stop

vmkload_mod -u nfs41client

vmkload_mod -u nfsclient

vmkload_mod -u sunrpc

localcli system module parameters set -m sunrpc -p netstackName=nfs

localcli system module load -m sunrpc

localcli system module load -m nfsclient

localcli system module load -m nfs41client

/etc/init.d/nfsgssd start

Once the host comes back up you can now use vCenter to add in a new vmkernel adapter

Choose VMkernel NEtwork Adapter

Select the DVS port group we created at the begining

Now change the drop down to the new stack we just created

Complete the setup

Once it is created you can now mount your NFS share with the vmkernel IP you specified and have the NFS traffic go through that vmk.

Windows 11 Insider Preview Instant Clones Horizon 8 2103 walkthrough - Part 2 Pool Setup and Deploy

2021-07-05T14:48:00.002-06:00

In this section we will deploy an Instant Clone Poo on Horizon 2103 with the Windows 11 Insider Preview.

If you missed it , In this post we showed how to get the Preview, and how to setup the image here.

Log into your Horizon Admin, select Desktops and click on Add.

Walk through the Wizzard:

Lets Get connected

Ok here are some quick inital performance shots:

Blast:

PcoIP:

Printer Redirection working :

Client Drive redirection working as well

Windows 11 Insider Preview Instant Clones Horizon 8 2103 walkthrough - Part 1 Image setup

2021-07-05T14:18:00.013-06:00

Just a reminder , non of this is supported by Microsoft, VMware, or any other software vendor at the time of this writing, any thing you do is for testing purposes. Enjoy!

----------Updated Guide for Windows 11 GA build can be found here https://jeramythompson.blogspot.com/2021/10/windows-11-ga-instant-clone-lab-with.html ------

Now that you have your Windows 11 Insider Preview up and running (if not see this post) we can get an Instant Clone pool up and running.

First thing we need to do is enable the local administrator account, and setup a password if not done already. Open Computer management (run compmgmt.msc).

Next log out of your temp admin you setup originally and log in as Administrator.

You will get some prompts for the new user setup.

Open up Computer Management and delete the old user:

Go ahead and reboot.

Once you have rebooted log back in as the local Administrator Profile.

Right click the Windows Icon and choose System:

In System Click on Advanced System Settings

click on User Profiles, and Delete the old users profile (should say Account Unknown )

At this point you are in a good spot to shutdown the VM and take a snapshot.

Next Lets get the OS Optimization tool and the Horizon 2103 Agent copied to the desktop.

You can find the OS Optimization tool here:

https://flings.vmware.com/vmware-os-optimization-tool

First We will install the Horizon Agent.

We will go through the process, and select every option in the agent.

After you have installed and rebooted for the Horizon Agent you will notice you are now prompted to press CTL ALT DEL to login instead of swiping up

Click Analyze

In our case we just want to get up and running as fast as possible so click apply. If this was a production setup you would definitely want to go through every option to make sure they match your companies requirements.

Lastly click the Finalize Tab, and click Finalize on the bottom Right

It seamed like it took an hour, but when finished you will get the following:

Shutdown the VM, and take your snapshot.

We are ready to deploy our Instant Clone pool, follow along in part 2 Here.

Windows 11 Insider Developer Preview in a VM - How to get up and running and test out Windows 11 in vSphere 7

2021-07-05T12:11:00.003-06:00

It is extremely easy to get the preview of Windows 11. We are going to be doing some testing Later with Horizon View so we will need a base VM.

To get started Create a NEW VM in vSphere, Select the VMXNET3 adapter and name it something easy to remember , in this Case we are using W11.

Next Install the latest version of Windows 10 you have hanging around, In my case I has a 1903 ISO so we just did a typical install.

Created a default user call 'Admin' (we will change this later), then installed VMware tools.

Reboot, you are now ready to start configuring the VM for the Windows 11 Insider Preview.

Click on Start and type Update to get to your windows update settings.

On the bottom Left You will see a link for the "Windows Insider Program" , click this.

In the Right Panel you will need to sign in to a Microsoft account (or work or school account if you have one setup with MS already) To join the program. When prompted select the "Dev Channel" to get the Windows 11 build.

You will be prompted to reboot. Once you reboot log back in and verify the following is the same.

Now navigate to the top right and click Windows Update, then check for update.

Windows will begin updating, in this case you should see 'Windows 11 Insider Preview' and some build number available.

Optional

As mentioned earlier this will be used in Horizon View, so to start the cleanup process I have enabled the local Administrator account and set a password. We will be deleting the "Admin" account once we are up and running.

At this point we are simply waiting for Windows 11 to complete the install.

Once it completes, you will be prompted to reboot

At this point you may want to re-install VMware tools, but you have a working copy of Windows 11!

I would recommend taking a snapshot before you got fully into the OS so you can restore to this point if you need to.

Extract remote DNS to local hosts file

2021-04-30T06:38:00.004-06:00

So with the whole work from home process, my company recently made changes to its VPN policy to block local DNS queries. This has completely screwed up my work flow for testing.

So with a bit of google-ing and duct taping some code I have come up with a simple powerslell script to extract the DNS zone from my lab to my local host file.

Enjoy!

#VARIABLES

$scriptpath = $MyInvocation.MyCommand.Path

$dir = Split-Path $scriptpath

$CSV = "$dir\DNS_A_records.csv"

#parameters

$DNSServer = "dc1.home.lab"

$Zone1 = "home.lab"

#$Zone2 = "me.myzone.local"

#SCRIPT MAIN

clear

$DNS_Zones = @()

$DNS_Zones += $Zone1

#$DNS_Zones += $Zone2

$hosts = @()

$DNS_Zones | % {

$zone = $_

Write-Host "Getting DNS A records from $zone"

$DNS_A_records = @(Get-WmiObject -Class MicrosoftDNS_AType -NameSpace Root\MicrosoftDNS -ComputerName $DNSServer -Impersonation 3 -Credential HOME\administrator -Filter "ContainerName = `'$zone`'")

$DNS_A_records | % {

$hostA = "" | select "IPAddress" , "hostname"

$hostA.hostname = $_.OwnerName

$hostA.IPAddress = $_.IPAddress

$hosts += $hostA

}

$hosts = $hosts | Sort-Object @{Expression={[Version]$_.IPAddress}}

$hosts | Export-Csv $CSV -NoTypeInformation -Force

$conv = Get-Content $CSV

$conv.Replace('","',",").TrimStart('"').TrimEnd('"') | % {$_ -replace ',',' '} | Out-File .\hosts.txt -Force -Confirm:$false

Get-Content .\hosts.txt | select-string -pattern "IPAddress hostname" -notmatch | Out-File C:\Windows\System32\drivers\etc\hosts -Force -Confirm:$false