Hello,

we bought a new Server and installed it with Windows Server 2019 Datacenter Edition.

I would like to replicate a Virtual machine from an other Windows Server 2012 using Certificate Based Authentication.

Both servers are on the same Workgroup.

But When I try to configure the replication on the 2012 Server I get the error message :

Hyper-V failed to enable replication for virual machine : the connection with the server was terminated abnormally (0x00002EFE).

Hyper-V failed to establish a connection with the Replica server. Error: The connection with the server was terminated abormally (0x00002EFE).

Any Idea ? Are Windows Server 2012 and 2019 not able to work together for replica ?

Best regards,

David

Hello everyone,

I'm having trouble getting a combination of SR-IOV, NIC teaming and MAC address spoofing to work correctly.

• If I team on the host, there is no SR-IOV (and I don't get the throughput).
• If I team in the guest with SR-IOV enabled and MAC address spoofing disabled, the VM has trouble communicating with some of the other hosts on its subnet.
• If I team in the guest with both SR-IOV and MAC address spoofing enabled, the Hyper-V host doesn't grant the VM a virtual function, i.e. SR-IOV does not work.

When I try to boot the VM with both SR-IOV and MAC address spoofing enabled, Hyper-V-SynthNic logs event 12585 for each VNIC. When I boot the VM with SR-IOV enabled and MAC address spoofing disabled, Hyper-V-SynthNic logs events 12584 and 12588 for each VNIC. If, after booting with the latter configuration, I tick "MAC address spoofing" in the VM's VNIC properties, Hyper-V-SynthNic logs events 12586 and 12590; once I un-tick it again, I see events 12584 and 12588 again.

The configuration is as follows:

• two Windows Server 2012 ["R1"] hosts in a failover cluster
• Windows Server 2012 R2 guest
• Intel X520-DA2 (dual-port 10Gbit SFP+)
• in-VM NIC team is configured with switch-independent teaming mode and the load-balancing algorithm set to TransportPorts (I haven't had any more success with IPAddresses or MacAddresses, though)

The mentioned log events of Hyper-V-SynthNic follow these message formats:

• Event 12584 message: '{VMNAME}' Network Adapter ({GUID}--{GUID}) allocated a virtual function. (Virtual Function ID 0 LUID 0:{4HEX} Virtual Machine ID {GUID})
• Event 12585 message: '{VMNAME}' Network Adapter ({GUID}--{GUID}) failed to allocate a virtual function: The request is not supported. (0x80070032): IOV networking might be incompatible with other configured networking features. (Virtual Machine ID {GUID})
• Event 12586 message: '{VMNAME}' Network Adapter ({GUID}--{GUID}) freed a virtual function. (Virtual Function ID 0 Virtual Machine ID {GUID})
• Event 12588 message: '{VMNAME}' Network Adapter ({GUID}--{GUID}) assigned a virtual function. (Virtual Function ID 0 LUID 0:{4HEX} Virtual Machine ID {GUID})
• Event 12590 message: '{VMNAME}' Network Adapter ({GUID}--{GUID}) unassigned a virtual function. (Virtual Function ID 0 Virtual Machine ID {GUID})

Does anyone know if there is a supported configuration of SR-IOV, in-VM NIC teaming and MAC address spoofing? Alternatively, is there a course of action that would make MAC address spoofing unnecessary?

Thanks in advance!

Cheers,
~~ Ondra

~~ Ondra "Ravu al Hemio" Hošek

Hi,

I have been testing Hyper-V Replica and all has been going quite well in a test environment.

When applying the same knowledge to a production environment i'm seeing strange results.  I have a Windows Server 2012 Hyper-V server in a workgroup running a guest SBS 2011 VM with two drives, one is 131GB and the other is 314GB.  I have moved the pagefile onto a third vhdx which isn't being replicated over to our replica partner.   

Looking into the VM's, there is minimal file structure change but yet i see lot's of big .hrl files being queued up for replication.  I've been through the majority of the files on the SBS server to make sure file changes are a minimum, i.e. IIS / Exchange logs and having kept an eye on the server throughout the day i can't see any big increases in storage consumption on the partitions.  within 7 hours today i can see that i have 11.5GB of .hrl files waiting to ship over to the replica partner.  The SBS server is running on an ADSL line is being crippled due to the upload.

From my test environment, after moving the pagefile onto a drive that wasnt being replicated i was seeing average .hrl files of KB's rather than GB's.  Does anybody have any advice on where the huge file sizes can be coming from ?

This is already my third seed and i'm still seing the same problem.

Alex

Hi guys,

I have two clusters, PR and DR, and have deployed the PR Hyper V replica broker role and DR Hyper v replica broker role. However, I just got the error "

a virtual machine resource was not found in a clustered virtual machine . . ."

when doing my initial replication job. Could this be due to the different vswitches on both sites, or what might be the reason?

And, considering each VM is on a different CSV, so i need a central CSV and make it the replication destination, or I'll have to point each CSV for it's corresponding destination on the DR?

Hi. We have a 2 node Windows Server 2016 Datacenter Hyper-V failover cluster. We have 2 seperate SANs connected. One holds the production VM's, the other one is for backup purposes, as well as a few test VM's and one node of our Exchange 2016 DAG. This morning the backup SAN failed and crashed completely. Reason yet to be determined. Hyper-V first logged IO retries, and soon enough it logged the lost volumes were stalled:

Cluster Shared Volume 'Hyper-V Test' ('Hyper-V Test') has entered a paused state because of 'STATUS_IO_TIMEOUT(c00000b5)'. All I/O will temporarily be queued until a path to the volume is reestablished.

Then about a minute later FailoverClustering started logging event ID's 1230, with messages like:

A component on the server did not respond in a timely fashion. This caused the cluster resource '<VM ON BACKUP SAN>' (resource type 'Virtual Machine', DLL 'vmclusres.dll') to exceed its time-out threshold. As part of cluster health detection, recovery actions will be taken. The cluster will try to automatically recover by terminating and restarting the Resource Hosting Subsystem (RHS) process that is running this resource. Verify that the underlying infrastructure (such as storage, networking, or services) that are associated with the resource are functioning correctly.

That's to be expected, VM's that loose their storage go down. Right after that, as expected FailoverClustering event 1146:

The cluster Resource Hosting Subsystem (RHS) process was terminated and will be restarted. This is typically associated with cluster health detection and recovery of a resource. Refer to the System event log to determine which resource and resource DLL is causing the issue.


However, after that FailoverClustering also stopped the production VM's. That storage had never been down. We don't have any virtual disks for production VMs on the backup SAN. All production CSV's were available. Cluster quorum is on production SAN as well. For every VM I had a message like this:

Failover clustering event 1069

Cluster resource '<PRODUCTION VM>' of type 'Virtual Machine' in clustered role '<PRODUCTION VM> Resources' failed.

Based on the failure policies for the resource and role, the cluster service may try to bring the resource online on this node or move the group to another node of the cluster and then restart it.  Check the resource and group state using Failover Cluster Manager or the Get-ClusterResource Windows PowerShell cmdlet.

But it doesn't state WHAT exactly failed. Also I just don't see why they fail if another storage box, that has nothing to do with these VM's goes down, failover clustering kills all other VM's.
By the way we have ODX disabled, which seems a best practice these days because of all bad implementations of it.

As soon as the backup SAN came up again, the VM's started booting again resulting in a bootstorm of course, but at least we slowly came up and running again.

I don't have a clue really where to look next. The logs show the VM's failed, but not why. I can't find any depencies on that backup SAN for any of the production resources. All drivers / firmwares are up to date as per HP's SPP of April 2017.

Hello everyone!

I have got a crash dump of Server Manager (WS 2016 Datacenter)

I opened dump with windbg and have got a next stacktrace (1419 frames):

 # Child-SP          RetAddr           Call Site
00 00000008`e5700e00 00007ffe`7f0c962a ntdll!RtlDispatchException+0x3c
01 00000008`e5701500 00007ffe`62e7ff0c ntdll!KiUserExceptionDispatch+0x3a
02 00000008`e5701c08 00007ffe`62b8d18a clr!DontCallDirectlyForceStackOverflow+0x10
03 00000008`e5701c10 00007ffe`629b77af clr!CLRVectoredExceptionHandler+0xa8
04 00000008`e5701c70 00007ffe`7f085ef0 clr!CLRVectoredExceptionHandlerShim+0xa3
05 00000008`e5701ca0 00007ffe`7f05fa7b ntdll!RtlpCallVectoredHandlers+0x104
06 00000008`e5701d40 00007ffe`7f0c962a ntdll!RtlDispatchException+0x6b
07 00000008`e5702440 00007ffe`62e7ff0c ntdll!KiUserExceptionDispatch+0x3a
08 00000008`e5702b48 00007ffe`62b8d18a clr!DontCallDirectlyForceStackOverflow+0x10
09 00000008`e5702b50 00007ffe`629b77af clr!CLRVectoredExceptionHandler+0xa8
0a 00000008`e5702bb0 00007ffe`7f085ef0 clr!CLRVectoredExceptionHandlerShim+0xa3
0b 00000008`e5702be0 00007ffe`7f05fa7b ntdll!RtlpCallVectoredHandlers+0x104
0c 00000008`e5702c80 00007ffe`7f0c962a ntdll!RtlDispatchException+0x6b
0d 00000008`e5703380 00007ffe`7f061fd1 ntdll!KiUserExceptionDispatch+0x3a
0e 00000008`e5703a80 00007ffe`6292e761 ntdll!RtlVirtualUnwind+0x91
0f 00000008`e5703b10 00007ffe`62a6c0fd clr!RtlVirtualUnwind_Wrapper+0x81

....

....

556 00000008`e57faf90 00007ffe`62a6c0fd clr!RtlVirtualUnwind_Wrapper+0x81
557 00000008`e57fafe0 00007ffe`62a6c028 clr!FixupDispatcherContext+0xc5
558 00000008`e57fb520 00007ffe`7f0ca4bd clr!FixContextHandler+0x48
559 00000008`e57fb550 00007ffe`7f05fd83 ntdll!RtlpExecuteHandlerForException+0xd
55a 00000008`e57fb580 00007ffe`7f0c962a ntdll!RtlDispatchException+0x373
55b 00000008`e57fbc80 00007ffe`7f061fd1 ntdll!KiUserExceptionDispatch+0x3a
55c 00000008`e57fc390 00007ffe`6292e761 ntdll!RtlVirtualUnwind+0x91
55d 00000008`e57fc420 00007ffe`62a6c0fd clr!RtlVirtualUnwind_Wrapper+0x81
55e 00000008`e57fc470 00007ffe`62a6c028 clr!FixupDispatcherContext+0xc5
55f 00000008`e57fc9b0 00007ffe`7f0ca4bd clr!FixContextHandler+0x48
560 00000008`e57fc9e0 00007ffe`7f05fd83 ntdll!RtlpExecuteHandlerForException+0xd
561 00000008`e57fca10 00007ffe`7f061a99 ntdll!RtlDispatchException+0x373
562 00000008`e57fd110 00007ffe`7b544c48 ntdll!RtlRaiseException+0x2d9
563 00000008`e57fd8f0 00007ffe`628a11e9 KERNELBASE!RaiseException+0x68
564 00000008`e57fd9d0 00007ffe`628a121b clr!NakedThrowHelper2+0x9
565 00000008`e57fda00 00007ffe`53a409ff clr!NakedThrowHelper_RspAligned+0x1e
566 00000008`e57fdf28 00007ffe`53a43a5e Microsoft_Windows_ServerManager_HyperV_Plugin_ni+0x409ff
567 00000008`e57fdf30 00007ffe`628a6a93 Microsoft_Windows_ServerManager_HyperV_Plugin_ni+0x43a5e
568 00000008`e57fdf70 00007ffe`628a6948 clr!CallDescrWorkerInternal+0x83
569 00000008`e57fdfb0 00007ffe`628a720d clr!CallDescrWorkerWithHandler+0x4e
56a 00000008`e57fdff0 00007ffe`62905324 clr!MethodDescCallSite::CallTargetWorker+0xf8
56b 00000008`e57fe0f0 00007ffe`629052ae clr!TryCallMethodWorker+0x1c
56c 00000008`e57fe150 00007ffe`629d1774 clr!TryCallMethod+0x76
56d 00000008`e57fe240 00007ffe`614ea76b clr!RuntimeTypeHandle::CreateInstance+0x2e4
56e 00000008`e57fe580 00007ffe`614ea519 mscorlib_ni!System.RuntimeType.CreateInstanceSlow(Boolean, Boolean, Boolean, System.Threading.StackCrawlMark ByRef)$##6001197+0x8b
56f 00000008`e57fe600 00007ffe`6145115c mscorlib_ni!System.Activator.CreateInstance(System.Type, Boolean)$##6000749+0x69
570 00000008`e57fe650 00007ffe`583d9406 mscorlib_ni!System.Activator.CreateInstance(System.Type)$##6000746+0xc
571 00000008`e57fe680 00007ffe`6145650c Microsoft_Windows_ServerManager_Common_ni!Microsoft.Windows.ServerManager.Common.Plugin.ReflectionRolePluginFactory+<>c__DisplayClass0_0`1[System.__Canon].<CreatePlugin>b__0()$##6001B35+0x16
572 00000008`e57fe6c0 00007ffe`61456128 mscorlib_ni!System.Lazy`1[System.__Canon].CreateValue()$##6000F19+0xec
573 00000008`e57fe720 00007ffe`5830c51c mscorlib_ni!System.Lazy`1[System.__Canon].LazyInitValue()$##6000F18+0xb8
574 00000008`e57fe790 00007ffe`628a6a93 Microsoft_Windows_ServerManager_Common_ni!Microsoft.Windows.ServerManager.Common.Plugin.PluginLoader`1[System.__Canon].InitializePluginWorker(Microsoft.Windows.ServerManager.Common.Plugin.RolePluginRegistrationInfo, System.Object, System.ComponentModel.AsyncOperation)$##6000453+0x9c
575 00000008`e57fe810 00007ffe`628a6948 clr!CallDescrWorkerInternal+0x83
576 00000008`e57fe850 00007ffe`62a51efd clr!CallDescrWorkerWithHandler+0x4e
577 00000008`e57fe890 00007ffe`62a51a3f clr!CallDescrWithObjectArray+0x6b5
578 00000008`e57feb10 00007ffe`61e93574 clr!CStackBuilderSink::PrivateProcessMessage+0x26d
579 00000008`e57fefb0 00007ffe`61475ab3 mscorlib_ni!System.Runtime.Remoting.Messaging.StackBuilderSink.AsyncProcessMessage(System.Runtime.Remoting.Messaging.IMessage, System.Runtime.Remoting.Messaging.IMessageSink)$##6005B31+0x1b4
57a 00000008`e57ff050 00007ffe`61475944 mscorlib_ni!System.Threading.ExecutionContext.RunInternal(System.Threading.ExecutionContext, System.Threading.ContextCallback, System.Object, Boolean)$##6003AD4+0x163
57b 00000008`e57ff120 00007ffe`614b5b43 mscorlib_ni!System.Threading.ExecutionContext.Run(System.Threading.ExecutionContext, System.Threading.ContextCallback, System.Object, Boolean)$##6003AD3+0x14
57c 00000008`e57ff150 00007ffe`614b4e22 mscorlib_ni!System.Threading.QueueUserWorkItemCallback.System.Threading.IThreadPoolWorkItem.ExecuteWorkItem()$##6003CAC+0x73
57d 00000008`e57ff190 00007ffe`628a6a93 mscorlib_ni!System.Threading.ThreadPoolWorkQueue.Dispatch()$##6003C8B+0x152
57e 00000008`e57ff230 00007ffe`628a6948 clr!CallDescrWorkerInternal+0x83
57f 00000008`e57ff270 00007ffe`628a720d clr!CallDescrWorkerWithHandler+0x4e
580 00000008`e57ff2b0 00007ffe`628abfc9 clr!MethodDescCallSite::CallTargetWorker+0xf8
581 00000008`e57ff3b0 00007ffe`628a79d1 clr!QueueUserWorkItemManagedCallback+0x2a
582 00000008`e57ff4a0 00007ffe`628a7940 clr!ManagedThreadBase_DispatchInner+0x39
583 00000008`e57ff4e0 00007ffe`628a787d clr!ManagedThreadBase_DispatchMiddle+0x6c
584 00000008`e57ff5e0 00007ffe`628a7a0f clr!ManagedThreadBase_DispatchOuter+0x75
585 00000008`e57ff670 00007ffe`628abf30 clr!ManagedThreadBase_FullTransitionWithAD+0x2f
586 00000008`e57ff6d0 00007ffe`628a804c clr!ManagedPerAppDomainTPCount::DispatchWorkItem+0xa0
587 00000008`e57ff850 00007ffe`628a7df5 clr!ThreadpoolMgr::ExecuteWorkRequest+0x64
588 00000008`e57ff880 00007ffe`628ac1cf clr!ThreadpoolMgr::WorkerThreadStart+0xf5
589 00000008`e57ff920 00007ffe`7ccc84d4 clr!Thread::intermediateThreadProc+0x86
58a 00000008`e57ffc60 00007ffe`7f08e8b1 kernel32!BaseThreadInitThunk+0x14
58b 00000008`e57ffc90 00000000`00000000 ntdll!RtlUserThreadStart+0x21

I think issue did appeared after install new updates for .Net/C#

How I can rebuild assemblies for HyperV plugin in ServerManager ? 

Is it possible to do some kind of cleanup, and get a chance of stable execution for ServerManager ?

Sergey Pachkov

I ran the survey script and I get the following error;

BIOS kept control of PCI Express for this device. Not assignable.

Unfortunately, this machine doesn't support using them in a VM.

To use SR-IOV on this system, the system BIOS must be updated to allow Windows to control PCI Express. Contact your system manufacturer for an update.

SR-IOV cannot be used on this system as the PCI Express hardware does not support Access Control Services (ACS) at any root port. Contact your system vendor for further information.

My current setup is;

CPU:- Intel Xeon e5-2680v2 (Supports all Virtualisation elements https://ark.intel.com/content/www/us/en/ark/products/75277/intel-xeon-processor-e5-2680-v2-25m-cache-2-80-ghz.html)

Ram: 32gb Corsair Vengence

Motherboard: Asus P9X79LE

HDD: 2 x 500gb Kingston A400

GPU 1: Nvidia 1050ti 2gb gddr5

GPU 2: AMD Radeon HD 6670 1gb gddr5 - wish to passthrough this

Sound: Realtek onboard 7.1 hd audio

Lan: Realtek 1gb

Operating System: Windows Server 2019 Standard

I have enabled hyperthreading, intel virtualisation and vt-d, I know I can do passthrough as I tested with esxi 6.5 update 1 and I can passthrough graphic cards, sata drives, audio and usb connections.

Does DDA require me to enable anything else in bios?

Hi folks,

While configuring settings for Live Migrations in H-V there are 2 authentication options, one of them is CredSSP. There is pretty solid info about Kerberos, but I wonder why and when would you go forCredSsp?

Any example. I understand that no further config is needed in this case. There is very little info about this bird.

Thanks for your insights!

I have several servers that export their running VMs using Mike Galvin's script here:
https://gal.vin/2017/09/18/vm-backup-for-hyper-v

In short this script checks for running VMs, exports the running VMs to a particular location, logs it, and can email if desired.

All but one of our servers runs this script fine. On the server in question I can log personally, open powershell in admin mode and successfully run this script with zero errors. However sometime in the last year the automated way we run this script stopped working... sort of.

On this server there are 7 VMs running. When the script is kicked off via Scheduled Tasks, 6 of the 7 VMs take the following error:
Event ID 19050 'ServerX' failed to perform the operation. The virtual machine is not in a valid state to perform the operation. (Virtual machine ID 7B70D5A0-F71C-45BA-8506-AB5A983594AF)

We have a dedicated account that runs these scripts. Again all other servers running this script have zero errors. I changed the account running this script to my own since I can successfully run it manually. Still pulls that error.

I've done a little searching on it and most of the data I find doesn't help. Does anyone have any ideas?

I have just set up a Hyper-V VLAN on my office workstation for the purpose of testing network changes prior to implementation. The VLAN has a host server running Windows Server 2016 with DHCP, DNS, Windows Deployment Service and Active Directory. There are also 2 clients running Windows 10. Currently the VLAN is on a Private switch, so there is no risk of my VLAN traffic bleeding into my office network. I'm at the point where I need to connect this VLAN to the internet, but in a way that there is no risk of introducing any VLAN traffic, rogue DHCP servers, DNS explosions or any other assorted goodies onto my office's production network. My workstation hosting this VLAN is a client on my office network which has a Windows 2012 R2 server also running DHCP, DNS and AD, so there's plenty to go wrong if don't do this properly. I've set my VLAN on a different subnet, 192.168.10.# vs the office's 192.168.20.#. I'm not sure if this is sufficient, or if there's more that I'm missing. 

I know I'm not the first person to do this, but despite all my searching, I can't seem to find anything that addresses my particular configuration. I'm still new to this, though, so maybe I'm not searching the right terms. Also, if someone knows a good tutorial to point me toward, that would be great too.

Thanks.

Hi,

I've seen this problem for a while now - First in existing deployments (which made me think it's an issue with the infrastructure), but now on new Hyper-V deployments as well:

When I try to restart a VM by actually pressing the restart button in the Windows Server 2019 Guest OS, the VM appears to boot into the Windows Boot Manager from time to time:

There is absolutely no issue with the attached VHDX-Files, the problem can occur on the simplest deployments with one local disc. And if I select the "windows server"-entry, everything is booting like a charm. 

My problem right now is, that this boot manager does not have a countdown. That's a real problem for our staff, which has no access to the Hyper-V Console, but is able to restart a VM by accessing it via RDP. 

My question is now: Am I the only one out here with that problem? I did some research, but I either used the wrong search phrases or I am missing something big here.

Thanks in advance for your time and effort - and sorry for my bad english, I'm not a native speaker.

Carsten Brenner - IT-Engineer @ cloud4you AG (Germany)

Get connected:
   

Hi,

I setup two hosts with Server 2019 and enabled certificate based replication on both.

Failover between primary and secondary host works fine. However when I reverse the replication I get this error message:

Hyper-V could not replicate changes for virtual machine 'Lubuntu_test' as Hyper-V is not in a state to accept replication on the Replica server. 

I have to remove the entire VM on the primary in order to have it succeed. This causes a lot of wasted bandwidth.

Both servers are in a different datacenter. I forwarded port 443 in the firewall.

If I create a new replication from secondary to primary host, it works fine.

What gives?

Hey guys, I'm looking to solve this riddle, but I'm stuck and can find a solution, I wonder if is possible to have two subnet both servers have AD/DC/DNS/DHCP roles in the same virtual switch in Hyper-V.

Example:

HYPER-V VIRTUAL SWITCH:

VM1 server IP configuration 10.0.0.0/24

VM2 server IP configuration 10.0.1.0/24

the second one, won't resolve the configuration.

Is this configuration possible?

Thank you in advance.

Valter

Good day everyone.

I have a slew of stand-alone hyper-v hosts running around 10-15 guest VMs each.  I'm looking to start clustering the hosts to have a higher degree of confidence in availability.

I created a lab cluster with two 2016 hosts using CSV on an EMC SAN. I have 3 test VMs running on each host, and as a test i did a few live migrations to the other cluster member, no issues, 1 second to move.  So...today i wanted to replicate a real failure so with each host running (3) VMs each, i unplugged one of  the servers.  The fail-over was not immediate as i was expecting.  From cluster manager on the running host, it showed "unmonitored" for the 3 VMs on the dead host for approximately 3 or 4 minutes and then started running them natively on the up host.

Any ideas why the failover wasn't immediate?

Thanks,

Danny

how can i get out from this state? can't stop, can't kill the process.

any advice?

Hi guys, I',m doing a cluster to cluster replication, and created a DR and PR replication broker. I'm doing an initial replication, but i get the error "a virtual machine resource was not found in clustered virtual machine"

Might this be due to different VSwitch Names for both PR and DR? I cant figure out what's causing this

In preparation for setting up S2D, I want to test the RDMA (iWARP) communication.
We use Qlogic QL41164HMRJ nics (Dell R640 - Server 2019).
RDMA communication works great (diskspd.exe) if I use the pNICs directly for this.
The counters in PerfMonitor give me positive results.

When using SET switch with vNICs it only works from vNIC to pNIC direction (SET and vNICs at the sending server, no SET at receiving server) not in the opposite direction (pNIC to vNIC). So, The vNIC does send RDMA traffic but can never receive it (Firewall is disabled on both servers)

In the end, both servers must have a SET switch with vNICs of course

Any ideas?

I have enabled jumbo packets everywhere in my network.

I verified that jumbo packets are enabled on the Hyper-V host physical NICs and Hyper-V virtual switch.

I tested pinging full size frames with no fragmenting from the Hyper-V host to another physical machine.

I tested Internet downloads from the Hyper-V host at full speed.

Inside the Hyper-V guest, if I leave jumbo packets disabled, all is fine.

If I switch the guest to 4K or 9K jumbo packets I get extremely slow LAN and WAN speeds.

Has anyone run into this before?

Thanks.

Hi guys, I'm experiencing the attached error when i try to enable replication. This is even when I try exporting the initial replica to an external drive. Any ideas what might be causing this?

The recommended firewall ports have all be opened to allow replication as well.