Delivery Group Reboot Schedule issues

[Andre Jochemsen]

Hi all,

We have four physical HP Moonshot servers in place. One NIC connected to PVS platform, and one other NIC connected to customer network.

On the Delivery Group we have a reboot schedule in place to reboot the servers every night. We notice some servers do get rebooted, but others not. Servers that are not rebooted will never reboot, so we have to manually reboot those machines. Its not that always the same servers wont get rebooted. Its randomly.

 

On the affected Citrix server i dont see anything in the event viewer related to Citrix (reboot). 

On the DDC is only see this event 3105

 

The Citrix Broker Service has terminated the processing of the 'GroupReboot' for 'Citrix Desktop'. The reboot cycle (UID = '33') ended with the final state 'Abandoned'. Summary: '0' machines successfully rebooted, '0' machines failed to reboot, '4' machines were not processed.

 

This is also not true. 2 machines did get rebooted. The other 2 didn't. 

 

there are other ways to reboot the machines. But if this is solvable, i would like to keep using the reboot schedule on the delivery group.

 

OS = Server 2012 R2.

Citrix server Studio version is 7.9.0.86

 

If somebody has some idea on what could be wrong i would love to know.

Thanks.

[Georg P.]

same Problem here, startet with XA7.12

 

-Server OS is 2008R2

-All Citrix Components are on 7.12 now

 

There´s one Delivery Group with a daily reboot Cycle (since forever), which restarts successfully everyday.

the weekly ones (or the groups i configured a daily reboot afterwards) aren´t.

 

today for example i wanted to reboot 20 Machines (distributed among 5 Groups), one of them did the trick, all the others were abandoned (whatever this means).

 

if i create a new Delivery Group, using the same Machine Catalog, everything works fine :/

[Andre Jochemsen]

Hi Georg,

 

So in ur case creating a new Delivery Group solves the issue?

I havent done the same yet. So i could also create a new Delvery Group and test the reboot schedule. 

Thanks.

[CarlStalhood]

Is it this issue? https://jhmeier.com/2017/01/23/scheduled-reboots-not-working-on-citrix-xendesktop-7-12/

[Andre Jochemsen]

Well, the issue is not with all our servers. 2 out of 4 are not rebooted.

We are on XenApp 7.9. So if that article is really for 7.12 then i dont think it will apply.

I did however saw the reactions underneath the article.One of the reactions said disabling the notification was the solution for him. We also have a notification setup, so i have disabled the notification and will look tomorrow if that had any effect.

Thanks.

[Georg P.]

seems disabling the notification works

would bet i tested this before but never mind :)

[Andre Jochemsen]

Unfortunately disabling the notification is not the solution for me.

 

Last night again 2 out of 4 servers were rebooted. It looks like a load balancing issue? There are two Delivery Controllers, but the first installed CDC is the primary right? What i also saw, the Citrix servers have a registry entry called ListofDDCs. Only the first DDC FQDN is listed in that setting. The second CDC is not listed.

I would say that doesnt matter really. In Studio -> Controllers -> i see the first DDC has all 4 Citrix servers registered. I could only imagine it goes wrong if the second CDC is involved when it comes to the reboot schedule... Could that be the case?

[Aaron OReilly]

I successfully used Remove-BrokerRebootSchedule -DesktopGroupName "devdesktopgroup1" to completely get rid of the non working reboot schedule and then recreated it in Studio, but this time with no notification.  All servers in it rebooted successfully that night. 

 

Adding back in a notification will break the reboot schedule again and will need to be deleted and recreated to get working again.  So it appears that once there is a notification set, the only way to fix the issue is to fully recreate the schedule without one.  Am in the middle of a ticket with Citrix running traces over servers with the various scenarios to find a solution.

[Brad Schoonover]

I am having this problem right now.  I am on 7.16.  Is there any other solution other than remove the Warning message?

[Jochen Steenbrugge]

Can't get it to work either.

 

It's now a V2 thing: BrokerRebootCheduleV2.

 

I have this:

Active                : False
Day                   : Monday
Description           : 
DesktopGroupName      : DGname
DesktopGroupUid       : 1
Enabled               : True
Frequency             : Weekly
MaxOvertimeStartMins  : 0
MetadataKeys          : {}
MetadataMap           : {}
Name                  : Wekelijks
RebootDuration        : 60
RestrictToTag         : 
StartTime             : 03:00:00
Uid                   : 2
WarningDuration       : 0
WarningMessage        : 
WarningRepeatInterval : 0
WarningTitle          : 

 

It says enabled, but not active. I would think active as if it was in the 1h reboot duration. I was carefull not to use a WarningMessage. 

 

I'm on 1912.

 

thx

[Louis Pomerleau]

Having the same issue. I just deleted the schedule and recreated it w/o notifications. Let's see...

When I check the schedule I get the same - Active : False as Jochen. But the schedule is enabled. What does "Active:False" mean?

 

[Michael Martis]

I think it means, if it's currently running the reboot schedule. 

I'm on 7.15 CU9  this has been happening for a while now for me. Only to 2 out of 9 delivery groups.  Guessing it's no coincidence the 2 affected DGs are ones that I had notifications set up. Removing the schedules via the GUI didn't work. Going to try  via powershell and rebuild . We'll see how that goes.