Jump to content
Welcome to our new Citrix community!
  • 1

PVS LTSR 7.6 to 7.15 - Target Device Boot Issues


Tyson Glaser1709155160

Question

Curious if anyone has heard of or seen any boot issues after upgrading from 7.6 LTSR to 7.15. I’ve had scheduled reboots for various target devices for over a year and never had a problem. After upgrading to 7.15 I’m having a pretty good failure rate on devices coming back up after a reboot. The target devices finds the disk on PVS but then freezes at the black screen with the windows logo (no spinning dots). Happens with both 7.6 or 7.15 target device software and didn't start until after the PVS servers were upgraded. Also happens on 3 different vDisks running different software. I have ticket open but so far no luck. Running on VMmware esxi 6.0 and servers are 2012 R2. Before the upgrade I could select a handful of servers, typically 24 at a time at most, and could restart all of them without issues. Now I'm lucky if half come back up. I haven't updated the VMware tools in awhile because I hate doing a reverse image but it was working fine before the upgrade so I don't think that would be it. Thanks for any info! 

 

Link to comment
  • Answers 86
  • Created
  • Last Reply

Top Posters For This Question

Recommended Posts

  • 0

For those of you having similar behaviors to the issue experienced by the OP.

 

This issue is very environment specific and should not be confused with other "slow boot or hung target" issues.

The issue found in this case is specific to the Bootstrap files on PVS 7.8 and later.

 

Before matching your issue to the same issue experienced by the OP, please keep the following in mind:

  • As of right now this specific issue is confined only to VMWare environments.
    • If your Targets are running on XenServer or Hyper-V, then it is most likely not the same issue, even though the behavior might be the same.
  • If the environment was working correctly on any version of PVS 7.8 or later, and you upgraded to another version and experienced this behavior, then it is most likely not the same issue.
    • This  issue is specific to bootstraps 7.8 and later, which means that if your environment was working correctly with 7.8 or later then it's not the same issue.
  • If the slow boot or hung Targets occur randomly or intermittently, then it is most likely not the same issue,
    • This particular issue will affect 100% of the Targets 100% of the time.
  • If you are experiencing a similar issue please open a service Request with Citrix Support before attempting to use different Bootstrap versions.

Sincerely,

 

-rauld

  • Like 1
Link to comment
  • 2

Here are a collection of older bootstraps. I pretty much tried all of them one at a time. Just remove the underscore and version number and make a copy of your current one. Also, if you have multiple PVS servers it needs to be on all of them and I restarted the TFTP service to make sure it was booting to the correct one. You can see the bootstrap version when it first boots so you know for sure you are getting the correct one for testing. 

Older Bootstraps.zip

  • Like 2
Link to comment
  • 1
18 hours ago, Tyson Glaser1709155160 said:

Here are a collection of older bootstraps. I pretty much tried all of them one at a time. Just remove the underscore and version number and make a copy of your current one. Also, if you have multiple PVS servers it needs to be on all of them and I restarted the TFTP service to make sure it was booting to the correct one. You can see the bootstrap version when it first boots so you know for sure you are getting the correct one for testing. 

Older Bootstraps.zip

Perfect! Thanks! It works flawless with 7.7.

 

 

  • Like 1
Link to comment
  • 1

We have a similar problem.

We updated the PVS servers from 7.9 to 7.15 and also the PVS target device software in our XenApp 6.5 image (W2008R2, Xenserver 6.0.2) about 2 months ago.

Everything seemed to work as usual, although I recently got the impression that booting the servers takes longer than before.

Then we installed a new image with W2016 on Xenserver 7.0 and about 5 of 20 servers didn't start the Windows at all, and several others rebooted some time after they have already been up.

I suspected a problem with LACP configuration on the hypervisor and switched back to a single NIC - no change.

Then I tried Xenserver 7.1 instead, again without a change in the behaviour.

I'll try the 7.7 bootstrap next I guess, but it'll take some time due to few HR ;-)

(We currently don't have a ticket opened at Citrix support)

 

  • Like 1
Link to comment
  • 1

Hello,

 

According to my experience, this issue is solved by doing the next 2 actions:

- upgrade your PVS Server AND PVS Target software to the 7.15 CU2 version (= 7.15.3)

- upgrade the SEP Management Server AND SEP Agent to the version 14.0 RU1 MP2 (= 14.0.3929.1200)

These were until recent the latest and most recent verions.  And I have several setups running (on W2K16 VDA's) without problems.  

 

Greetings,

  Chris

  • Like 1
Link to comment
  • 0

Hi all,

i have in different environments this issue too.

PVS Server have Version 7.15 and the target tools, too.

This happens with XEN Server and vSphere ESX as Hypervisor.

I saw this issue with Windows Server 2012 R2 as Worker/XenApp/PVS-Target.

@Citrix, do you have input for us?

PVS Target Tools where updated inside a new v-Disk Version. (no Fresh install)

Best regards

 

Jürgen

Link to comment
  • 0

The issue only seems to affect our Win 10 fleet that we only recently started ramping up the roll out of. The problem has probably been there from the beginning of the Win10 project but with just a handful of VM’s and plenty of development work we probably just chalked them up as anomalies with whatever change was being made. Our Win 7 fleet is unaffected apparently.

main differences are VM hardware type (9 vs 11) and Win 7 does PXE but Win 10 uses BDM (PVS doesn’t Echo DHCP option 82 required for Cisco ACI) so I can’t really tell what version was ever working. Beyond that, same PVS servers, same network, same storage, same VM hosts.

Following a majorish, unrelated bug (dhcp renewal issue with Cisco ACI) MCS is starting to look pretty attractive at this point...

Link to comment
  • 0

Anything new? This happens for me aswell. New fresh pvs 7.15 install and server 2016 targets. 1 image. Happens randomly at reboot. I firsth thought it was a performance issue as I got the feeling it was happening when i was rebooting multiple vms on the same time. Tried with both cache in ram with overflow and cache on server.

 

Running on vmware both targets and pvs servers

Link to comment
  • 0

The ticket is still open as they are trying to figure out what the difference is between 7.7 and 7.8 bootstraps and why it would be causing a problem in my environment. We did find that even though our VMware side is on 6.0, our vSwitch in the environment is still on 5.1 which could be causing some weird issues but we haven't tested that theory out yet. Waiting on the VMware team to upgrade it. But currently as it stands, everything is working with my 7.7 bootstrap file in place. I have several different vDisks (all server 2012R2) and everything boots fine now. As soon as I change to any other bootstrap higher than 7.7 the issue comes back so that's 100% my issue. Just not exactly sure why. 

Link to comment
  • 0
16 hours ago, Tyson Glaser1709155160 said:

The ticket is still open as they are trying to figure out what the difference is between 7.7 and 7.8 bootstraps and why it would be causing a problem in my environment. We did find that even though our VMware side is on 6.0, our vSwitch in the environment is still on 5.1 which could be causing some weird issues but we haven't tested that theory out yet. Waiting on the VMware team to upgrade it. But currently as it stands, everything is working with my 7.7 bootstrap file in place. I have several different vDisks (all server 2012R2) and everything boots fine now. As soon as I change to any other bootstrap higher than 7.7 the issue comes back so that's 100% my issue. Just not exactly sure why. 

 

Can you share the boostrap file? =)

Link to comment
  • 0
On 2017-11-16 at 3:48 PM, Tyson Glaser1709155160 said:

Good deal. Do you have a ticket open currently? If not can you create one and reference mine (74179924). They would like all the data they can get on this issue. 

 

Ill see what i can do. No open ticket atm, not sure how the customer wants to handle it. Ill ref you ticket if we create one.

 

-rog

Link to comment
  • 0

Any news about this? Have the same behavior at a customer site with a new vmware environment (esxi 6.5 and tools 10.1.15), pvs and target tools 7.15. with server 2016 management systems and server 2012r2 xenapp worker. Only 50 % of the targets are booting the other 50% are freezing at the windows logo. The customer has to go produktiv at november 29 so I build up a workaround with PVS 7.6.5 and get all xenapp worker up and running...

 

Marcel

Link to comment
  • 0

I just tried with the 7.7 bootstrap.

Our XenApp 6.5 servers on Windows Server 2008R2 are up within 3 minutes now, with the 7.15 bootstrap it took more than 10 minutes.

But our newly set up XenApp 7.15 servers on Windows Server 2016 show the same behavior: about half of  them is booting, the other half is stuck at a black screen.

Furthermore most of the servers that are booting into the GUI don't recognize the Xenserver management agents and the console seems to be very slow.

I also tried the "interrupt safe mode" in the bootstrap configuration but that didn't do the trick, either.

 

Regards, Stefan

 

 

Link to comment
  • 0

We have been seeing a slow boot issue like this since 7.13 ... our interim solution has been to restart 1 of our 2 PVS streaming services and this seems to clear it up. At 7.13 it was a daily scheduled task, after an upgrade to 7.15 I got a week out of it. I'm glad to hear that CU1 might also be a fix for this.

Edited by ezdzcho952
fixed version
Link to comment
  • 0
2 minutes ago, Erric Zdzchowski1709158692 said:

We have been seeing a slow boot issue like this since 7.13 ... our interim solution has been to restart 1 of our 2 PVS streaming services and this seems to clear it up. At 7.13 it was a daily scheduled task, after 7.16 I got a week out of it. I'm glad to hear that CU1 might also be a fix for this.

Not really the case in our instance. Literally the target device just stops receiving packets from the PVS server and it's something specific to bootstrap files that are 7.8 or higher. Still have an open case with Citrix. 

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...