World of EUC

I have one group with 2 catalogs and trying to work out sessions per catalog seems impossible

Not sure if this is what you need but I'd do this:
Get-BrokerSession | Select UserFullName,CatalogName | Sort-Object -Property CatalogName
Of course you can add a filter to select only the catalog you want:
Get-BrokerSession | Where { $_.Catalogname -eq "MyCatalog"} | Select UserFullName,CatalogName | Sort-Object -Property CatalogName

is there a function that can sum the lines and just give me the number?

Get-BrokerSession | Where { $_.Catalogname -eq "MyCatalog"} | Select UserFullName | Measure-Object

it’s the historical data I need though, I can get that live count from studio

Yep, you'll have to talk to OData API. <https://www.citrix.com/blogs/wp-content/uploads/2014/01/users-connected-to-Delivery-group-for-the-day.txt> Don't know if you can find info for the catalog, but I'd start from there

I don't know if you can do this, but it's opening pandora's box, you'll probably have to change properties of the VMs on Citrix side to align with Hypervisor's connection as well as a lot of other things (Provisionning schema, VMID, and so on).
Honestly, I would recreate all the VMs.

I’d imagine the way to do this (we run Citrix Hypervisor though) would be to leave the existing machines and hosting _as is_, create new hosts in the delivery controllers and a new machine catalog referencing the new hosts.  Image could be the same image, you could use the existing delivery group. Spin up as many new VMs in the new catalog as needed to start a transition.

Do your cleanup after the transition.

<@ULLRRKZ8E> Makes sense. Simply leave the existing as is and recreate new ones. I assume if these were persistent you would be screwed then.

It would be the same process, just more tedious.

Like convert to full clone as an example and either reimport for power management or change UID using PS for the hosting connection

If i had to move a bunch of persistent machines i’d schedule an outage.

• Power them down.  
• Remove from the existing MC, ensuring i don’t select delete the machines from the hypervisor :)
• At Hypervisor level do a vMotion to new cluster.
• Back in Citrix import them into the new MC.
• Add back to the delivery group, fix any 1-1 user mappings.

Full clones persistent just flip hypervisor connection using PowerShell.

<https://support.citrix.com/article/CTX140486>

Possibly try this - Support statement on storage migration of MCS machines (<http://citrix.com|citrix.com><https://support.citrix.com/article/CTX216179|)>

It may just be easier recreating MCS machines and adding them back to Delivery Groups.

We did this and just moved all of it, worst that happens is the power state goes unknown and you can’t power manage them or deploy new machines to those catalogs.

and if you want to mess about in the database changing IDs you can get the power state back

I posted my process on reddit a couple years back when we moved to a new vcenter we had mcs provisioned non persistent SBC machines the move went well without a hitch <https://www.reddit.com/r/Citrix/comments/6mufxx/vcenter_move_mcs/?utm_medium=android_app&amp;utm_source=share|https://www.reddit.com/r/Citrix/comments/6mufxx/vcenter_move_mcs/?utm_medium=android_app&amp;utm_source=share>

Process was basically remove the machines in the DG and delete from machine catlog removing everything the VM /AD Account/Remove from XD DB  , Delete the Machine catalog/DG, setup new hosting connection to new Vcenter and cluster in studio installing new vcenter cert and selecting the new datastores and networks, swing over the ESXI host hosting your MCS master image VM to the new Vcenter the host has to use a standard v switch once it over you can change it to use a distributed switch port group, recreate the machine catalog using the new hosting connection, recreate your DG adding the machines from the new machine catalog, restore DG config (published apps /assigned users ) i used powershell to do this, re do DHCP mac reservations as there new vms job done

make sure you’re on the latest CU if you’re on 7.15LTSR because it doesn’t like the 6.7 certificates pre CU3 and you can’t do it via the GUI

I’ve got some scripts that handle persistent machines being moved into manual catalogs and all that fun - on GitHub if you need/want them - few could use some cleanup at some point 

<@U010KKUE4HH> Got it. That was my initial guess, that it would be simply easier to recreate everything

we're using app-v for a small number of apps.  any app that is easily patched remotely (pdq deploy for us) i have in the image and masked w/ fslogix for anyone not in a specific ad group and those that AREN'T easily patched (ones i have to do manually ) i've tried to work into app-v so i can patch it once and deploy

Stopped using app-v when app layering was bundled with our Citrix entitlement. 

App-V for as much as possible, because we can update them on-the-fly. I blogged about it here <https://james-rankin.com/features/appventix-fslogix-super-slick-application-delivery/>

<@UL5TCSW3D> your blogpost is the reason why i‘m asking :-D

But for me it‘s important to know which apps your packaging

All of them except Office, browsers, platform tools and anything with Office add-ins

We don't have any CAD tools, but I would do them if we did, yeah

Interesting, why not browsers? Thats the shit with the most updates and most work in img update

We rebuild our entire farm every month so unless there's a critical out-of-band update inside that, it doesn't matter

Okay, i think that depends on the environment. 

Everyone should do it. No patching, instant security response capability, no mucking around with PVS or MCS - it's the future

Yes, but for me as external consultant its not possible to manage that and the most internal it Departments do not have the man power or knowledge for that

Yet the have the manpower for gold image updates and patching? Give them the infrastructure-as-code, and it becomes very portable. It's much simpler than you think. I had exactly the same reaction initially - now I am taking this approach to all my customers

Some years ago only one customer want to build the farm with sccm instead of any citrix provisioning. But today he is using mcs because its easier to manage for him. And i also have some horizon customers... here it should not possible to work without instantclones (and i really hate them) in vdi... why does vmware push such a big shit

No, they not have the manpower for packsging and software deployment. So we have some images with hundreds off applications and really special addons and extra software developments. There are not possible to packagae. Thats the point where it breaks for me

Anything is possible to package, and once packaged, they become modular. The old way of doing things - MCS, PVS, SCCM, etc. - is very much the past

I‘m totally compare with you, maybe germany need some more time toget there ;)

100% agree, and you can get started for free. You can also automatically build App-V and MSIX packages. Automation Framework Community Edition (<http://euctraining.com|euctraining.com><https://courses.euctraining.com/|)>

Finally! Windows Image Acquisition scanner redirection

WIA scanner jumped out at me as well.  Eager to test with *.wfh scanners

Turns out WIA Redirection is prohibited by default, here's the policy for it:

image.png

I should add that this is happening on Windows 10 in Azure using NVv4 AMD image and running 1912 CU2 VDA

Hey. Dont know if this is any help but: <https://support.citrix.com/article/ctx237608|https://support.citrix.com/article/ctx237608>

Doesn't this use DRS if enabled? If not I think round robin. Need to double check.

I think we turned DRS off because we ran out of resources. I guess it’s round robin

This would be power on. Creation would be round Robin.

Hey Brian. DTLS is just an encryption protocol. It's got nothing to do with SR. You can use SR with or without EDT.
If you do not want to use EDT yet, you can disable it via Citrix policy by disabling "Adaptive Transport".
Alternatively, you can disable DTLS on the Gateway vserver and that will force all connections to use TCP since DTLS won't be available to encrypt UDP-based traffic (i.e. EDT). With this option, your direct/internal connections would use EDT (assuming UDP 2598 isn't blocked in your network).

You dont need DTLS enabled on the Netscaler GW Vserver for session reliability to work. As long as you got it enabled within the citrix polices and its enabled on the NS GW appliance in storefront your good. Just make sure 2598 is open from the SNIP to VDA subnet

sessionreliability.JPG

so many times i have seen customers enable session reliability but only have 1494 open on their internal facing firewall so double check from the SNIP to VDA subnet 2598 is open. Another good option if your using a HA pair is to enable "session reliability on HA failover"

Thanks guys. This is what I suspected but wanted to verify. I don't have access to the netscaler as they are owned by the network team :disappointed:

tell them it has a citrix badge on it :slightly_smiling_face: they are normally all too happy to let the citrix teams manage them

The docs make it sound like ANY device that supports WIA.

Windows Image Acquisition scanner redirection

This Citrix solution allows support for the Windows Image Acquisition (WIA) framework for image acquisition devices that utilize it. WIA exposes a framework that enables scanning/imaging applications access to a variety of features from the scanning device itself. The WIA API gives access to various functionality of image acquisition devices, such as enumeration of all available devices, connections to multiple devices, querying various properties of devices, acquiring device data, and providing notifications of device events using event handlers.

Thanks Carl. I already found it and nothing more. WIA feature wars requested by me more than 2 years ago and it was delayed 12 months because of some implementation issues based on my source from Citrix.

Unfortunately description from “what’s new” this is very little to understand how WIA works and if any pre-requirements are needed. 

got a 6.5 farm that any good or got to be 6.0 ?

Got to be 6.0 sadly. I just need to verify if you can run 6.0 apps using the latest Workspace App. 6.5 I know is fine with 2010 CWA

let me check might have a 6.0 farm knocking about

Bildschirmfoto 2020-12-15 um 09.49.56.png

Bildschirmfoto 2020-12-15 um 10.12.18.png

OK, to clarify… This was a VDA 2009 with WEM agent 2011.1.0.1 from WEM cloud service. Seems that the VDA didn’t notice that there is already a WEM agent installed because there was no option in the GUI to configure WEM. If this is true we have a problem here, because WEM cloud agent and WEM on prem agent differ in the version!

gosh, now I need to test this with on prem version to see how it acts

Yea, Ask them why we always find out all the issues after the release to GA?

haha, Chances are they will be like, ummm idk

yeah this is all pretty basic so far, I’m only doing it because the exam is so awful and I need to know what madness they used to come up with the questions

Bildschirmfoto 2020-12-15 um 16.20.52.png

Bildschirmfoto 2020-12-15 um 16.19.30.png

Bildschirmfoto 2020-12-15 um 16.20.24.png

I’ve had that when update services are disabled 

Almost all of ours take around 400MB. My current session looks like this:

with that, Edge/teams my vdis are at 12GB RAM for reg users

Ouch. This is on Server 2016, XA. I don't think we have any larger than that.

What uses that executable? Is it dependent upon screen resolution?

Are you using 1912 or 1912CU1? It’s already fixed in CU2

One more option: ask Citrix support for fix - you will be able to apply on CU0 or CU1 - it’s just one file.

We're on 1912 CU1. Looking to go to CU2 shortly.

Upgrade your VDA agent to CU2 and memory leak should be gone. 

Another thread stated they had to get a private fix for another issue. So just FYI, if they go unregisterd.

1912 CU2 has a problem with VDAs unregistering? Sounds fantastic!

<https://worldofeuc.slack.com/archives/CK6TCV3JN/p1606150044279600>

<@ULHA2PS2H> Keep us updated on the CU2 VDA on your testing. I am curious to see the results. Please and Thank you :slightly_smiling_face:

1912.2 in Test, still seeing high CtxGfx.exe memory usage:

Tim: This is too high. Report it to Citrix Support. Maybe it’s new bug. Don’t forget to dump few processes and share with Citrix. They beed to investigate

Thanks <@ULL7R5KT6>. Contacted Citrix support, and they have several other reports of the same, and are investigating / gathering information. I'll update once I know more.

Just ask them for hotfix. It was the same with my support request. They investigate 2 weeks when fix was available since long time. Sometimes Citrix Support wasting your time... :man-shrugging::skin-tone-2:

That was a great 6.5 and lower feature, that AFAIK hasn't made it into 7.x.. I could be wrong though... i remember <@UM8LJ5G9L> was trying to do this at one point

I think you have to modified the default ICA file and put the app in there with the resolution.

Something like thisConfigure custom screen resolution for a Published Desktop on XenApp 7.x | Robert Steeghs (<http://wordpress.com|wordpress.com><https://robertsteeghs.wordpress.com/2018/06/26/configure-custom-screen-resolution-for-a-published-desktop-on-xenapp-7-x/|)>

Makes me wonder why they never add that back after all this time?

It's only one app requirement <@ULLH5JGFR>

This article is for published desktop, does it applies for published application?

There are all  I can find.
<https://helgeklein.com/blog/2015/05/citrix-desktop-viewer-screen-resolution-and-window-size/>

<https://support.citrix.com/article/CTX116357>
<https://www.jasonsamuel.com/2012/09/10/how-to-force-a-citrix-xenapp-application-to-open-in-a-custom-size-window/>

If none of these work, then I am not sure.

Remove them using <#CLCTBHBLH|>? :slightly_smiling_face:

Or whatever tool you prefer. AppX packages are a mayhem, there was good research from <http://go-euc.com|go-euc.com>

yea but that introduces more issues, picture viewer, calc, sticky notes.  I tried to remove them but users want them :disappointed:

Agree Cory, and with MSIX coming this is something we just need to start living with. Though the logon time penalty is less than 10 sec

It was supposed to get much better in 2004 build. I would typically leave "core" appx applications intact (picture viewer, snip &amp; sketch, calculator and sticky notes), but remove everything else

THIS Martin. My test is on W10 20H2 which behaves well

ok so time to update I guess :slightly_smiling_face:

Yeah we should start testing for IT Admins

Thank you for the answers :slightly_smiling_face:

stop using sticky notes and use outlook notes which are backed up in exchange

I had a client that was trying to use sticky notes as a customer database, it was a nightmare

Stickynotes will sync via Office 365 account these days, IIRC

I see a lot of persistent desktops. The customers that use them also use SCCM for management and update/upgrade tasks.

Cool, yeah, we have SCCM too (that's one of the many hats I wear), so I plan to treat them the same in terms of patching etc. But given that VDA upgrades are coming thick and fast, I'm curious if people upgrade that on a regular basis for persistent desktops. It's easier for non-persistent, as it's in the image. But my assumption that persistent desktops are treated akin to physical or non-Citrix VMs

we treat them like they are physical machines.  previously we didn't update the VDA unless we had to.  We have a software tool that helps us with updates now so not as big of a deal

<@UL5TCJ57D> great, thanks. What's the tool, if you don't mind me asking? And what's your process in updating the VDA?

BigFix, IBM's version of SCCM.  We just create a scripted install and push the package out

Do you remove the old one first, or just install over the top?

Hmmm, persistent desktops for admins.......high value targets

yeah, use LAPS, use WEM with privilege elevation, use applocker, those persistent machines will get popped

it’s always the persitent machines where they have local admin that turn out to be the source of malware infections

We're starting to embrace the PAW model. It's a long journey but it involves multiple layers of auth, use of CyberArk, removal of persistent profiles and the like. Interesting stuff

You aren't always given the choice.  Keep in mind it's not a purely technological choice

Customers will give you the option pretty sharp when they get smashed by the ICO for a breach :slightly_smiling_face:

I mean you always have the choice to get more flexibility at the risk of less security

local admin makes life easy for everyone, which is why so many people do it

how is it any less secure than a physical endpoint?

All endpoints for admins are high value targets. and should be secured as such

I read the question as persistent desktops FOR admins, not necessarily making them admins on the persistent desktop.  It's an access path.  If secured like a physical device, that's the main point

and that’s what we’re saying, secure them

Yep, any device that admins are adminning from, whether they are admins of the admin device or not, is still a high value target

Multiple admins logged on and using all sorts of tools to connect to stuff hackers love

and you can run mimikatz and just grab all those creds from ram

Slight thread hijack going on here, sorry, but it's a point I'm hearing a lot these days when talking about how to manage admins

We have a bunch of persistent machines for devs, and they are the ones CISO are always screaming about because that's where the malware infections are always found. We rebuild them every six months but that's not often enough for me. They're managed via SCCM (which gives me the shudders as well)

All good points. Not sure if they will be local admins (my preference would be no), but I don't make that choice. But they will be used for administrating other services mostly. But I take all your very well made points. What's up with SCCM? It essentially pays my mortgage :laughing:

It was used by the dinosaurs for managing their estates

T-Rex doesn't have long enough arms, or thumbs to manage SCCM.

It's like AppSense, if you know it, you can do some cool stuff.  But getting to know it sucks.  And then you can kill things massively really fast

Yeah. I'm gonna respectfully disagree on that one. Speaking as an SCCM admin..

We use MSP software and Chocolatey to manage the software.

We built our own set of PoSh scripts to keep our ~400 persistent desktops up-to-date, pretty much automatic, I don't upgrade VDA usually.

<@UL5TCJ57D> I have a security team that start persistent VDI is less secure because of the different set of security tools that are in use as compared to physical end point devices...

Don't forget us :slightly_smiling_face:

well when they are using security tools from the 80s

Don't make us ask for a rate increase..there's a pandemic going on. lol

I'm not in your service territory so jack the rates up :slightly_smiling_face:

I have to chip in too, SCCM is like carrying a very big square item, you can move it but its difficult and awkward. Just my opinion based on my experience. PDQDeploy, WSUS and APPV Scheduler always seems to work a treat. PDQ for adhocs apps and scripts, wsus for Windows (although pdq enterprise does it) , apps scheduler for apps obviously. Sounds disjointed but I've had more reliability with these tools than using sccm and they're cheaper.

Sure, it's not as ad-hoc as PDQ, but it's definitely doable. I haven't used PDQ, but people do tend to like it. I don't find SCCM to be difficult or awkward, but I've been an admin for some years now. In a world where we're 100% remote, and early stages of EUC, there's no other tool out there that does as good a job of managing our remote Windows endpoints. Its ties into Azure and a CMG means we can patch your device without being on our network/vpn, including 3rd party patching too. Plus, Microsoft practically gives it away with any of the Microsoft 365 licensing.

We have a few persistent desktops provisioned with MCS full copy. Then managed with SCCM for patches.

I appreciate everyone's insights. Given me a lot of food for thought. Truth be told, it seems like I was making a bigger deal about it than it probably needed to be.

We use them and are coverting to pooled. The biggest problem is we have to use sccm and they can get dirty fast. The master images isn't tied to them after they are deployed. You have to use powershell to update provisioning method on them machine catalogs to be able to have newly created machines created from any updates made on the master image.   Some people say blow them away and redo, sure but because they are persistent, the changes that are made are outside my awareness. There for I could impact the users. It servers it purpose and can be useful for crappy software that doesn't work on a non persistent approach. But overall they are ok and are useful. It's still faster the physical computers. I just have to make sure what we push out is pushed out to the MCS master images that is used to provision them out. Example I just spun one up today and it was almost a year old. I forgot to update the provisioned part with powershell. Which then would have created it with all the updates changes.

Ah, I was reading something like that yesterday. Are you referring to this?

<https://support.citrix.com/article/CTX129205|https://support.citrix.com/article/CTX129205>

Oh, ha, maybe it was that I was reading that then. Your site is in my RSS reader

<http://www.citrixirc.com/?p=1158|http://www.citrixirc.com/?p=1158>

Cool, I'll add this to my OneNote, cheers

Oh, do you have 2 sites? 

<https://citrixadvice.blogspot.com/?m=1|https://citrixadvice.blogspot.com/?m=1>

I post on Citrix IRC and that free google blog is my main site. Call it on a budget site lol

What I dont fully understand is why Citrix has not been about to update presistent machines from a machine snapshot. If you think about it, when you snapshot a device, It creates a different disk. Then it will merge the changes into the base disk on any give device. Why cant the same thing happen for MCS persistent? Push the snapshot out and have the change get merged into the presistent disk.

Yeah, In the beginning, I assumed that would be the default behaviour, to be honest. *If* that was the case, that was one of my concerns, we'd have to coordinate with folks who'd be using them at all hours. If you were able to update via pushing out the snapshot they'd be easier to manage in some regards. Instead, they become another machine having to be managed by 3rd-party tools (i.e. not Citrix).

Sorry, I didnt mean to say that is what happens, Which it does not on persistent. I just was saying why they haven been able to do that yet.

Sure, that was my interpretation of you said. Just had a bit of a ramble. Sorry!

the training guide is 800 pages, and the presentations in the training is extra material on top of the guide.

Has anyone here done this before? Am I missing something crucial that ties it all together?

I was on the course last year renewing my cce-v was disappointed labs and lab guides sucked

It's like 5k right?  I need to renew my cca-v and wanted to step it up to ccp-v Not saying that's the course I would take but it's disappointing to here this.  I don't do well on test, and need all the help I can get.

Not to make it about me. But I'm weak on the cloud stuff and getting a ccp would make me feel like it's fake. Even if I renewed my cca I have read it integrated with cloud stuff. So I have to learn it to past the test.

the professional doesn’t have that much cloud stuff from memory, unless they changed it in the last 2 years

the expert level is just brutal though, like you need to know that you can’t run PVS in the cloud (unless you’re on vmware cloud but the exam won’t list that as an option)

and then you need to fill in the blanks on these access matrixes which I’ve never seen anyone in the real world lay out the way the do in the training and the exam

why is profile management so crappy!! <@UK6S90JDR> when will Nutanix have something :slightly_smiling_face:

Not crappy just tool in the toolbox depending on need. Profile Management doesn’t have a filter driver like others.

Honestly the hardest part of VDI is the profile part, maybe just me but feels like I am always fighting with an issue.. not mounted, UWP apps, GPO not applying, OST.  Feels like I fix one just to have another issue.

<https://james-rankin.com/articles/quickpost-citrix-upm-profile-containers-feature/>

<@UL5TCSW3D> you have a blog for everything! Bravo! 

Yes reading your blog it’s not ready for prime time 

It'll do a job, but it's not as flexible as FSLogix. There may be new features coming that make it better, but until then, it is what it is

James does have a blog for almost everything, The guy is a beast

Try <https://discussions.citrix.com/topic/356373-xendesktopxenapp-7x-applications-exporting-importing/page/7/#comment-2071881>

Not terribly clear?  <https://www.usb.org/defined-class-codes>

or just "class=0e" if you want to cover all video devices, the "h", just means it's a hex code, I don't think it's needed in the citrix policy

Send me your testing, I can update my USB blog around your findings

Black screen can be caused by many things! The first thing I look for is the AD properties of a user account. Login  scripts, RDSH profile, profile, home drive, bad network drive....all of this crap can lead to black screen and slow logon since everything will be processed synchronously.

Login scripts -&gt; GPP/WEM
Roaming profiles -&gt; FSLogix/CPM
Folder Redirection -&gt; Use reg keys via GPP instead of GPO or redirect Known Folders to OneDrive/ShareFile

I put this in place for a client this week , we were noting increased logon times after an MCS update:

<https://support.citrix.com/article/CTX212905>

It only happens when I enable fslogix office containers 

<@ULLH59VQF>: This might be a case to run through using procmon or the Guy Leech logon analyzer script available from here (doesn't require ControlUP)
<https://www.controlup.com/script-library-posts/analyze-logon-duration/>
I ran it last year, and identified 10-15 seconds added logon time from old App-V package caching

Ran into this at the start of the pandemic which killed any attempt of rolling out RemotePC.

I just got hotfix for CU1 and will test next week. 

Wait what. Work is suppose to happen next week.

Unfortunately yes. It’s 24/7/365 business. 

Maybe something that I was experiencing. This is what I do: <https://www.jhouseconsulting.com/2019/05/16/best-practice-for-the-windows-time-w32time-service-for-rdsh-and-vdi-workloads-1948>

I added the below commands to our VDA startup script after we experienced some issues with DST :
`#region Windows Time service (W32Time)`
`&amp; "$env:SystemRoot\System32\w32tm.exe" /config /update`
`&amp; "$env:SystemRoot\System32\w32tm.exe" /resync /nowait`
`#endregion`
BrokerAgent is set to "startup delay"

if it can’t pick up the correct time from the domain then you run into all sorts of issues with group policy as it doesn’t think it’s properly domain joined because the time is too far out for kerberos.

I’ve seen it more with MCS than anything else

it will grab the time in UTC from vmware at boot and then have to change it once it can see the domain but it doesn’t always work

I’ve been meaning to write an article about it for years because NTP is one of the most misunderstood technologies

but I’m not sure I even understand it enough to explain exactly why it goes wrong sometimes

thanks everyone for the help!  ultimately would it make the most sense to just change it from manual (triggered) to automatic for everyone?  cant  imagine it would cause any more harm that way right?

Yep, that would be a good start. But as Mike says, it gets it's initial time reference from the hypervisor during boot and then gets itself a bit messed up with the way the image management works. The explanation I gave in my article was the closest I found to why this happens. Need some further help from Microsoft and/or Citrix support for a deeper dive.

thanks guys! we're a persistent desktop setup in AHV if that makes any difference but hopefully just the service type change helps since we dont get too many instances of this but i spun my wheels for a while until i stumbled on this one since even though the time service was stopped , the date/time was correct and even when i'd restart the citrix desktop service then it wouldnt matter

<https://kb.vmware.com/s/article/1189> check that out, by default (even if you uncheck the sync time to host box) esxi syncs time to the host “when you migrate the virtual machine using vMotion, take a snapshot, restore to a snapshot, shrink the virtual disk, or restart the VMware Tools service in the virtual machine (including rebooting the virtual machine).”

you can edit the vmx and turn it all off.
What I found in one client was that because esxi is set to UTC, the VMs were assuming that time was local time and were so far out that NTP would not resync them.

you can also set the time to whatever you want in that file with rtc.startTime = &lt;Epoch timestamp&gt;

thanks dude! only thing is we're Nutanix AHV so not sure those specifics apply but overall idea i'm sure does in one way or another.  i'll start  w/ the automatic setting and maybe i get lucky  but this is all great info!

so testing a bit here w/ set-service to change that service to automatic which works but its changing the service to Automatic (triggered start) rather than just automatic for some reason.