World of EUC

Hi, can confirm this, maybe you can click on "provide article feedback" and get an info to citrix, that also in newer versions than 7.7 the problem can exist, too. Thanks!

I can't believe I've never come across this before, but is there a limitation in Citrix with accessing optical drives in a double-hop scenario? I can access the CD-ROM drive in my XenDesktop session just fine, but attempting to open it in a published XA app within that desktop results in a "\\Client\E$ is not accessible" error, even though the same CDA policies are set.

Have you followed this? <https://support.citrix.com/article/CTX127872>

I opened the ICA reference PDF that was linked. Assuming that the NativeDriveMapping keys listed there need to be set for this to work, I've confirmed that they're there in my W10 VDA and am waiting for the XA admin to confirm them on her end.

You have to apply those changes on first hop (VDI), not on last hop (RDS)

Nope. I've had that enabled from the start.

Did you make registry changes or .ini changes?

It was a couple of years ago, but pretty sure it required registry changes, not changing any .ini files

And I believe it's all being handled by our Citrix policy in Studio

Nativedrivemapping should only be needed if you want the drives from the middle hop to show (since XA6.5 and up). Default unset it should map through client drives from the original client to the second hop, which it does in the environments where I use it. If you set nativedrivemapping on your VDI you're telling the next hop XA to map to your VDI, not your original client (even though in some environments the text on the drivemapping will have you think otherwise)

Well there you go, changing _NativeDriveMapping_ to FALSE worked.

in HKEY_LOCAL_MACHINE\SOFTWARE\Citrix\ICAClient\Engine\Configuration\Advanced\Modules\ClientDrive

thought so, I was going to type, if you have it set and don't have a E: drive natively on your VDI, you'd be seeing that issue

Now I need to figure out where that particular registry setting is coming from

remember that if you're using other citrix in citrix connections that they might be using the mapping in that way. A lot of HR applications that are used in Belgium work that way and we actually need the second hop to present client drives since the first hop could be a thin client etc.

<@ULKKN0B44> Would be worth a blog post (or CTX article)

Yeah. that'd be helpful. I was manually modifying that key in my W10 VDA, but if I change the "Client drive mapping" GPO by deselecting "Enable client drive mapping", none of the drives are mapped in the XA session.

yeah that gpo controls the cdm virtual channel from the client side. So turning that off is the same as applying a ctx policy to say no drive maps are allowed, just only for that client.
As far as I know, there is no managed gpo setting for nativedrivemapping so manual (gpp prefs etc.) is the way to go. It will take on any session after that, no reboots necessary.

<@ULGKFDUVD> having them fix the original article would probably go a long way, I'm pretty sure it used to contain something in the direction of that. Otherwise I'd gladly give some feedback on extra stuff they could mention in it.

Already opened a ticket to get it fixed...

We have been planning for a while to create a "Double-hop Best Practices" article for <http://techzone.citrix.com|techzone.citrix.com>

Whenever you write it, definitely throw a link to it somewhere on Slack.

I'll have to test setting that key as a GPP, because I'm not sure if the other policy enabling it will override that setting.

Any new content is automatically promoted on <#CLKSAPFUG|> channel :slightly_smiling_face:

Definitely page file unless you have a ton or ram.

This is why i like this site.  Thank you all.

WEM memory working set optimization basically flushes to pagefile every X mins so if you’re using that then - definitely pagefile :)

But with anything test. I have seen some get by without it.

And if you go with a pagefile, hard set it.

We are doing tests with and without.  Without we didn’t even consider until we saw posts discussing it.

It's all about your applications being predictable really, with no pagefile one user eating too much memory will seriously hinder other users or worse, crash the box.

<@ULLRRKZ8E> don't even think going pagefileless in an Enterprise environment without tons of testing. Some apps would not even work if they don't see pagefile on and most will run into serious issues when they'll require more memory. Some have a bad habit of using both by default so leave pagefile on and manually setup the size. Cheers!

seen it done and done it, honestly not worth the effort.  if your that short on storage that your trying to not have a PF your projects already in trouble

Have you seen this? <https://docs.citrix.com/en-us/tech-zone/design/reference-architectures/virtual-apps-and-desktops-azure.html>

I have and it's really good,  unfortunately Citrix Cloud is not really an option for us at this time.  The licensing is rough for High Ed.

yip that's correct, between the SNIP and VDA is 1494/2598

Thanks. Is that the SNIP that’s on the same network as the VDA or the SNIP in the DMZ?

SNIP that can see the VDA's, the outbound SNIP.

Ok. I think this helps. Will share with network admin

Just to clarify Steve, it'll be the SNIP that respects the route for the VDA. Unless the SNIP is on the same network as the VDA.

Is it not TCP 2598 between SNIP and VDA for SR - UDP 2598 would be EDT if enabled.

Yes (although not a PVS 1906.2 that I know of)

Yes, we have multiple customers on it. No issues other than the ones they now fixed. Which they fixed in the right way, public hotfix and then repackage of the installers

That's quite interesting. I used to be a team lead that was responsible for this content (TechNotes), but it was internal-only newsletters and I was not aware that someone published it on <http://support.citrix.com|support.citrix.com> :slightly_smiling_face:

If it's not required to detect it via WMI, I highly can recommend RDAnalyzer, you can use the free community edition <https://www.rdanalyzer.com/features/>

Yes we created RDAnalyzer for that. The community edition is a free "display only" version. With the subscription version you can live switch between policies

i have a script which detects falsely for last few versions of XenApp servers. Any help in identifying the HDX graphics mode being used using WMI?

There is a tech paper being prepared for Tech Zone to cover that. But yes, that method is still valid.

image.png

might be able to override it using the custom\style.css file

Thanks. Dont have something to debug at the moment

Working. Much appreciated :slightly_smiling_face:

Just tested it on our validation servers, seems to work

time will do that if they are offset alot

yes, we were running SQL on a non-standard port, when we switched back to 1433 it cleared up.

Well this is only affecting one server out of a lot.  No port changes on the SQL side either.... mmm

Sooo this is interesting.. its looking like it may have been the ica_response probe we setup for monitoring storefront from ca nimsoft..... still verifying...