Peplink Balance 305 HW Version 2 - Major Incoming Traffic problems

Product Discussion Peplink Balance
1

We have two Peplink Balance 305 routers in High Availablity mode at our Datacenter. They have been working great for the about a year now.
They have been running firmware Version 8.0.0 build 2636
We host a business application via Web Services behind them.

This morning get calls due to downtime and when I check the servers web services are fine, but I am unable to even ping the IP’s for the services. No changes have been done to the router near the time of the issue and the last piece was to add port forwarding for a SSTP VPN on a server and removal of the local PPTP VPN.

We have a site-to-site VPN which during the issue works fine and allows me to connect to the routers.

A simple reboot of the master router resolved the issues and so I thought it best to update the router firmwares. Did the Slave first and then the Master with no issues.

Currently running:
8.0.2 build 2721

Got a call not 40 minutes later and see the same issue is happening again.
I see nothing in the event logs to guide to what is the issue

Again a reboot and it works again right away when the Slave takes over and continues after the master is back.

Need some assistance as I have nothing I can find to go on.

2

The more I work on this it appears to be 20 minutes between loss of port forwarding.

Screenshot
Screenshot1162×490 70.7 KB

Attached Screenshot of when the issue begins, data transfers cut out.

I have diagnostics from when the issue starts that I can provide.

Honestly there seems to be no way of getting critical support for what is a basic function of the router.

3

Have you had any work done upstream in the datacenter?
If you go to the support.cgi page do the WAN interface stats show anything is amiss?

I have only seen regular failure like that when there was a network loop WAN side causing a broadcast storm.

Have you logged a ticket with engineering?

4

So I managed to get some assistance from Peter from West Networks. It does seem to be an issue at the datacenter with someone statically assigning our IP’s which cause our High Availability setup to flap.

Never thought I would deal with that at a datacenter, only seen that with cable modems before.

Thanks for the help!

5

Glad to hear you worked it out. You’re in very safe hands with Peter. Thanks for the update!