Win 10 Guest Loses Network

[rbane]

I have a Mac OS 12.6.8 host running Virtualbox 6.1.46 with 2 Linux guests and 1 Win 10 guest. No issues with the Linux guests but I am frequently losing all network connectivity on the Windows 10 guest. All 3 guests are using Intel PRO/1000 MT Desktop adapters in bridged mode.The guest is running guest additions version 6.1.46 and is Windows 10 Pro version 22H2. The only way to get network connectivity back is to restart the guest. VBox.log attached. Any insight is appreciated.

[rbane]

bump. Anyone have any ideas? Thanks!

[mpack]

Ok, I'll take pity: with only a partial log provided I wasn't previously willing to spend any time on this. You need to make sure the VM is shut down before grabbing the log, and the log should also be zipped so that it fits within forum size limits.

It would also be useful to know what "losing all network connectivity" actually means. What did you test? Pinging Google? As background we obviously understand that VirtualBox is a hardware simulator and the hardware must be simulated pretty well if you have at least one guest OS that works on that hardware on your host... implying perhaps a config problem inside the guest OS.

I'm not aware of anyone else reporting that the hardware sim spontaneously stops working after a time.

[rbane]

Apologies as you can see this was my first forum post here. Attached are what I believe are the full logs. I shutdown the VM this time. I couldn't fit log.3 but its from months ago anyhow (I can post it in another post if you think its relevant). Let me know if I'm missing something there. I'm certainly open to it being an issue with guest OS. From the Windows guest side I'm not seeing anything in the event logs other than items like Windows Update all of sudden no longer resolve DNS. By all network connectivity, I mean my Nagios server alerts me this host is down. From the guest OS I cannot get echo replies from internal or external hosts. I cannot pull up websites with Edge from the guest or check for Windows Updates. Cannot RDP to it. The VM runs headless but when I show it there is also no link light activity at the bottom of the screen. I've also tried a disconnect/reconnect of the cable in VirtualBox. The only way to reestablish connectivity so far is to restart the VM. It may run a day without issue, a week, or only a few minutes.

[mpack]

Did you allow the VM to have a randomly assign MAC or did you try to force the MAC value for some reason?

In a similar vein, since the VM is bridged to your physical network, does the VM use an IP address or hostname that could conceivably conflict with any other device on the host network? I'm thinking that when the other device comes online it screws with the routing tables on your network and leads to this problem. One way to test would be to ping an external IP address (not a name), that should rely on LAN resources to work, provided the Internet gateway remains accessible.

[rbane]

Screenshot of network adapter settings attached. I don't believe I forced a particular MAC address. I'd have no reason to. If you think generating a new one would help I could try that. Its not impossible that there is a host/IP conflict but unlikely. The guest IP is outside of the scope for our DHCP pool that regular client machines receive. I can certainly try pinging 8.8.8.8 or something when it happens again.

[mpack]

A picture won't help: I can't identify a conflicting MAC by sight. You would presumably know if you kept the MAC the same (e.g. by cloning other VMs without changing the MAC and then running both at the same time on the same network). You would definitely know if you forced the MAC in an attempt e.g. to keep the hardware signature similar to some other PC - people try this every now and then.

The guest IP is outside of the scope for our DHCP pool that regular client machines receive.

I don't understand this statement. Your first post mentioned the VM is bridged to your physical network, hence it presumably receives an IP from your LAN's DHCP server like any other device on that network. You didn't mention making any arrangments for a different address pool. In that case how is the VM identified as belonging to a different subnet?

[rbane]

Now that I understand the question: No the VM was not cloned, so no the MAC was not forced. It was built from scratch, not converted from physical or trying to match anything existing.

The VM is configured with a static IP. No a DHCP reservation. Same subnet as other machines. I was just trying to indicate that its not part of the same pool of addresses that are doled out by our DHCP to regular clients (laptops, tablets, etc).

[scottgus1]

Let's see if we can get some hard numbers on this problem.

First, Right-click the VM in the main Virtualbox window's VM list, choose Show in Explorer/Finder/File Manager. Zip the VM's .vbox file (not the .vbox-prev file), and post the zip file, using the forum's Attachments tab. (Configure your host OS to show all extensions if the folder that opens does not show a .vbox file.)

Also please be sure you're starting the VM each time from full normal shutdown, not save-state.

Run the following sets of tests twice, please, once when everything is working, and completely again when the Windows VM loses network. Please label which is which:

1. In both host and VM OS's, run the correct command from the following in the Command Prompt or Terminal:
   ipconfig /all
   ifconfig
   ip address
   Post the outputs, label which is which.

2. In both host and VM OS's, ping 8.8.8.8. Post the outputs, label which is which.

3. Shut down the VM from within the VM's OS if possible. If not possible, close the Virtualbox window for the VM with the Power Off option set.
   
   Right-click the VM in the main Virtualbox window's VM list, choose Show in Explorer/Finder/File Manager. In the "Logs" subfolder, zip the VM's "vbox.log", and post the zip file, using the forum's Attachments tab. (Configure your host OS to show all extensions so you can find the "vbox.log", not "vbox.log.1", etc.)

[rbane]

Working data attached. Will upload data for failed state as soon as it happens again. Also want to confirm I am not using save-states and have been shutting down from within guest os. Thank you

[rbane]

Not working data attached. Thanks!

[fth0]

19:02:29.293991 Changing the VM state from 'RUNNING' to 'SUSPENDING'
19:02:29.299922 AIOMgr: Endpoint for file '/Users/administrator/VirtualBox VMs/Win2/Win2-disk001.vdi' (flags 000c0781) created successfully
19:02:29.420748 PDMR3Suspend: 126 710 761 ns run time
19:02:29.420785 Changing the VM state from 'SUSPENDING' to 'SUSPENDED'
19:02:29.420799 Console: Machine state changed to 'Paused'
19:02:30.918948 Changing the VM state from 'SUSPENDED' to 'RESUMING'
19:02:30.919410 AIOMgr: Endpoint for file '/Users/administrator/VirtualBox VMs/Win2/Win2-disk001.vdi' (flags 000c0723) created successfully
19:02:30.919799 Changing the VM state from 'RESUMING' to 'RUNNING'
19:02:30.919814 Console: Machine state changed to 'Running'
[...]
19:23:44.770435 VMMDev: Guest requests the VM to be turned off

In the VBox.log file from Working.zip, ~21 minutes before shutting down the guest OS, the VM was paused for ~1.5 seconds. Do you know if this is related to the network problem?

In any case, please provide a zip file with the contents of the VirtualBox global configuration folder (~/Library/VirtualBox).

Pinging 8.8.8.8 with 32 bytes of data:
Reply from 172.16.1.21: Destination host unreachable.

This probably means that even the gateway at 172.16.1.1 isn't reachable (doesn't respond to ARP requests). You could verify that by pinging the gateway.

[rbane]

VirtualBox global configuration folder attached. I don't think the pausing is related to the network issue. However, if it legitimately happened at 7:02PM it wouldn't have been me pausing it. I'm sure the gateway is also unresponsive but I can verify. Thank you for the feedback.

[fth0]

00:00:00.018442 Log opened 2023-09-12T17:10:35.401184000Z
[...]
19:02:29.293991 Changing the VM state from 'RUNNING' to 'SUSPENDING'
[...]
19:23:44.770435 VMMDev: Guest requests the VM to be turned off

FWIW, it was around 2023-09-13T12:13:04Z (UTC), and the guest OS was shut down around 2023-09-13T12:34:20Z (UTC).

Unfortunately, the global log files didn't give any relevant insights.

[rbane]

Thanks, yeah so that would still put it at 8:13PM EST and 8:34PM EST which is after hours for me. When I check those times on the guest VMs system log for 9/13 I don't see that it was shutdown.