Since a few months my Peplink Balance One looses internet connectivy randomly. Wifi and wired internet does not work, internal traffic eg to my NAS is working as normal. Cannot ping google.com, can ping my NAS. I have two WAN connections configured in the default Outbound Policy as priority, so if the first one fails the second one should take over. Enforced all traffic over one, then the other, to no avail. Cannot see anything from the logs, would not know where to start debugging. After a reboot all is back to normal, until it starts failing again…
A few weeks ago I switched to firmware 8.5.0 and then to 8.5.1 but somehow I feel it is not related to that as I had this same issue a few years back. It then happened for some time, then never happened anymore, until a few months back…
Weird thing is that when this happened last night (lost internet connectivity while browsing the internet, and daughter complaining about wifi being down), the YouTube music video that was playing on the TV through the Apple TV, happily kept on playing. It seems that any traffic that is routed to a fixed WAN through a dedicated outbound policy (like I configured for the Apple TV) keeps on working, but all traffic that is routed to either one of the WANs (based on the priority rule) starts failing and keeps on failing after one of the WANs was lost (even if that WAN is recovered again shortly after). And the only way to restore connectivity is to reboot or do a power cycle.
What can be the cause of this? How can I debug this one? Nothing to be found in the logs…
Thanks for the suggestion! I was not even aware that I had some QoS settings applied, but I had… so I disabled them. Will check if connectivity issues are still ongoing or not, after one of my WANs has gone offline…
I have been having the same problem since upgrading from firmware 8.4.0 which is very stable. I have seen it on both my Balance One and 20X. I filed a support ticket and through that process was then instructed via a bot to file a support ticket…. So I let it drop. I assume this will get addressed through a firmware upgrade once enough people report it. BTW - no QOS settings on my networks
Just to be clear, with B One do you mean the new model or the older Balance One? I have the older (not so much older) Balance One (with 2 WAN and 8 LAN ports)
Sorry… I was talking… in this post about the new B-ONE. My mistake…
And… also… I have the old BPL-ONE… Same at you… To stop the random reboot of this old BPL-ONE, I disable the WiFi. (I have few devices with all licenses)
Thanks for your input, will check if downgrading is possible, just to try it out for a while. The problem seems to be related to one of the two WANs dropping, and then the other not taking over as expected.
It may be worth to follow this up with support if you’re still eligible to it; I am not and many people may not be anymore since it’s an older model now. Otherwise, I am just hoping support is reading this post…
Today, while I was working, this problem occurred again. And I can confirm that it is related to one of the two WANs dropping, even if only briefly, then the router cannot switch to the other one but “thinks” there is no WAN and has no internet connectivity anymore. Why would the Balance One do this, when it is configured to fail-over to the other WAN in the outbound policies? I have configured my two WANs in one outbound policy using the “priority” algorithm, with “when no connections are available” set to “drop the traffic” and “terminate sessions on recovery” to “enabled”. Could these settings be wrong maybe?
Furthermore, the WAN connection was restored 6 minutes later but the router was not capable to restore connectivity by itself. A reboot was needed to do that. Why is this? Is it misconfiguration? Or a technical flaw?
This would be consistent with my experiences, in which I find that a VLAN with an Outbound Priority set, will sometimes stop working for no reason. A quick test is to revert to firmware 8.4.1 and see if your problems go away.
If so, you can add yet another person to the long list of us having serious troubles with firmware 8.5.0 and 8.5.1.
That is exactly what I have configured, a VLAN with outbound priority rule, prioritising my fastest WAN.
Your comment about firmware clarifies a lot for me: I experienced these problems in the past, then they were gone for more than a year. Now they are back since six or eight weeks or so. And, six or eight weeks ago, I upgraded to 8.5.0… now I am on 8.51. Hm. Hopefully 8.5.2 will fix this all and will be released soon.
I pulled the cable on one of the two WANs and it failed over on its own backup 4G line (cabled WAN was down) so internet connectivity was not lost. However, when I re-inserted the cable into the Balance One it was not capable to get internet connectivity working again, even though now both WANs were up and running again… This is very strange to me.
I am a bit fed up with it. I am going to cancel my second backup fibre connection since the Peplink won’t allow me it to use it as a backup anyway. Instead I am going to rely on one fibre connection only, with its own reliable failover mechanism to 4G (hosted in its own modem). So far my trust in Peplink. My next router will not be a Peplink anymore, I am afraid.
These are my outbound policies (that is what you meant, right, with outgoing rules?).
“No VPN IPs” is an IP range in my network which will always be routed through one WAN, my Apple TV is in that range. Now I come to think of it, this rule will not make sense anymore once the second WAN is gone. This rule is a left-over from the days that I used FusionHub, since Netflix and the likes won’t play through it.
“AirVPN IPs” is an IP range in my network that is routed through AirVPN. I switch my desktop or laptop to it only when I need a VPN.
The Default rule is set as Priority, prioritising KPN fibre then SKP fibre. Setting the Default to Auto won’t change a thing, same problem will occur.
Before the last two rules…
I don’t like to use enforced for a single wan… I prefer to use priority… When use enforced and this wan go down… everything stops works… this is the reason, in my case, to use priority.
