AC68U:Dual WAN causes primary to drop for no reason

[Nocturnal]

Hi,

I have a primary cable modem and u-verse for backup. Until I got the 68U a month ago I used a Cisco RV42 (I think) dual WAN router to front my DRL-655. Cisco being Cisco this was a bit clunky but it worked well enough in fail over mode.

With the AC68U in fail over mode however thing don't work well at all. It drops WAN connectivity every few hours. Once I disabled dual WAN, disconnected the u-verse router and only kept the cable modem connected things got much more stable.

I've registered my router with ASUS but their website is horribly broken and won't allow me to open a support ticket. I get some vague 'value can't be null' error even though all fields are filled in.

Does anyone have any tips for making Dual WAN work properly? Cable is much faster but less reliable than u-verse so I want to use fail-over mode. I'm considering switching to the Merlin firmware and give that a try. Still, first hand experience with Dual WAN would be very much appreciated.

Thanks,

Sander

 

[Jak Crow]

I've been seeing the same issue. Cable modem on the WAN port, DSL modem on port 1, and frequent disconnects of the cable connection without the DSL connection picking up, PLUS any web browser open getting prompted by the router to log in at the time of these disconnects. Running on the latest Asus release, so I sure as hell would like to know what the problem is because it's not the network connections.

 

[Nocturnal]

Unfortunately it's just as bad with the Merlin build. I would go as far as saying Dual WAN is completely broken. It really makes no sense to enable this feature or even advertize it. I suppose the fact that it's not in the manual should be an indication.

2 18:39:03 WAN(0) Connection: ISP's DHCP did not function properly.
Jul 2 18:39:03 stop_nat_rules: apply the redirect_rules!
Jul 2 18:39:10 dnsmasq[3289]: exiting on receipt of SIGTERM
Jul 2 18:39:10 dnsmasq[18147]: started, version 2.69 cachesize 1500
Jul 2 18:39:10 dnsmasq[18147]: asynchronous logging enabled, queue limit is 5 messages
Jul 2 18:39:10 dnsmasq-dhcp[18147]: DHCP, IP range 10.1.1.2 -- 10.1.1.100, lease time 1d
Jul 2 18:39:10 dnsmasq-dhcp[18147]: DHCP, sockets bound exclusively to interface br0
Jul 2 18:39:10 dnsmasq[18147]: read /etc/hosts - 5 addresses
Jul 2 18:39:10 dnsmasq[18147]: read /etc/hosts.dnsmasq - 0 addresses
Jul 2 18:39:10 dnsmasq-dhcp[18147]: read /etc/ethers - 17 addresses
Jul 2 18:39:10 dnsmasq[18147]: using nameserver 75.75.75.75#53
Jul 2 18:39:10 dnsmasq[18147]: using nameserver 75.75.76.76#53
Jul 2 18:39:10 dnsmasq[18147]: using nameserver 192.168.1.254#53
Jul 2 18:41:30 rc_service: wanduck 539:notify_rc restart_wan_if 0
Jul 2 18:41:30 stop_wan(): perform DHCP release

I have several blocks like this in my log. In single WAN mode there aren't these constant complaints about DHCP not working. And why would it need to check DHCP so often anyway? Surely the lease is measured in days, not minutes.

Really a pretty crappy product for what costs a premium. That I can't contact ASUS support also really irks me. You get a nice VIP member card but it's completely pointless. Entering a support ticket on their website is futile.

Damn that makes me mad. I don't mind spending a few extra $$ for a premium product but to spend this kind of cash for something that's average at best isn't cool.