Intermittent WAN Issues After wanos Install

[Spiffster]

After installing wanos on the ESXi server at a remote branch I am having some intermittent issues with some of the devices on the remote network. Right now its just the one wanos install, so nothing is being optimized, but traffic is flowing through the wanos vm. My network monitor was showing random devices on the remote network going down then up again. When I take the wanos VM out of the equation and reconnect the pfSense VM from the Bridge vSwitch to the LAN vSwitch, things start working reliably again.

Any ideas at all? I will keep testing and report back as well.

[ahenning]

Hi, I have seen this happen when promiscuous is enabled on the port group as well as the vswitch level. If vlans/port groups are used, disable promiscuous on the vswitch level. When lan0/wan0 are connected to the same network this can also happen due to a loop.

Also check in the wanos log, whether it displays any useful info to help troubleshoot.

[Spiffster]

Im not using port groups and vlans to make the bridge, just a separate vSwitch like this: http://wanos.co/forum/index.php?topic=110.0
On both the LAN vSwitch and BRIDGE vSwitch I do have promiscuous mode enabled at vSwitch level.
wan0 is connected to the BRIDGE vSwitch along with pfSense while lan0 is connected to the LAN vSwitch.

[ahenning]

Thanks, I am not sure, unless perhaps the wanos vm struggles to process all the traffic at certain times e.g. backup run. If pfSense is running as a vm and processing all the traffic then I doubt this, but guess its a possibility with the additional vswitch + promiscuous ports. Are the notifications generated frequently e.g is it possible to see icmp ping drops? Also, is this occurring on one of the 100/20 Mbps links or 400 Mbps?

[Spiffster]

This is a remote office with a 100/20 link. pfSense is processing all of the traffic, and the wonos box is setup to bypass all traffic #99 rule. The handful of times I have turned on wanos and changed pfSense's LAN connection from LAN vSwitch to the BRIDGE vSwitch the strange disconnects happen. Most of the issues seem to be related to devices connected to the physical switch, which is odd. When I run continuous pings to wanos IP and the physical switches management IP I get random dropped packets.

I feel like there has got to be some kind of loop somewhere, but when I draw out the topology, I cant see how this would be the case.

Im not sure if this matters but just having wanos on by itself doesnt cause an issue, its only when I move pfSense's LAN interface to the BRIDGE vSwitch so traffic traverses the wanos VM that things start to go sideways.

[ahenning]

That does sound like a loop. I just looked at the esxi vswitch config again and perhaps you're right. This could be possible if the vmnic0 and vmnic1 are connected to the same network segment. To eliminate this possibility would it be possible to place the interfaces on the WAN vswitch in specific WAN vlan  port group? Or alternatively check that vmnic1 is not in trunk mode. If the vmnic1 is in an access vlan on the physical switch, and this vlan is not trunked on vmnic0, there should be no loop.

[Spiffster]

Well I think I got it figured out. All I did was delete the BRIDGE vSwitch, then created another one. After that I enabled promiscuous mode on the port group within the BRIDGE and LAN vSwitches. So the vSwitches have promiscuous mode disabled, but the port groups have it enabled... Im not sure how that is different being that I did not use the VLAN method, but it seems to work just fine now. Still have some testing I need to do to make sure its all good, but I will update this thread when I determine the issue is 100% fixed.

[Spiffster]

Two days in and things are looking good. After all, I have had wanos running at my house for months without issue. Still not sure what caused the loop here, but I have made mistakes before ;-) Gonna put the other pieces in place this weekend and test some more.

[ahenning]

Awesome, hope it all goes well.