What's new

RT-AC88U - WAN drops after ISP replaced fiber modem

  • SNBForums Code of Conduct

    SNBForums is a community for everyone, no matter what their level of experience.

    Please be tolerant and patient of others, especially newcomers. We are all here to share and learn!

    The rules are simple: Be patient, be nice, be helpful or be gone!

tranceaddict84

Occasional Visitor
On Thursday my ISP swapped my ancient Zhone ONT/modem/router/AP for a simple GPON ONT/modem. The Zhone was in bridge mode, connected to the WAN port on my AC88U. The new ONT is purely a modem so replicates this setup.

Since then everything is good except that every 13 minutes 42 seconds (yes, I've timed it - it is this exact and predictable to the second) my internet connectivity will drop completely. It always recovers within ~5 seconds.

The drop causes RDP sessions to disconnect, any ongoing downloads to fail, and Sonos playback from web services to stop about 3 minutes later (presumably the size of Sonos's buffer). The RDP thing is the most annoying as I live in RDP sessions pretty much constantly for work.

Weird thing is I've had continuous pings running to 8.8.8.8 from my PC and from the router CLI, as well as a 100ms interval, port 443 tcping to my work's external Citrix Gateway IP, and these are not interrupted at all - not a single packet dropped. I've also had a laptop on my phone hotspot pinging my fiber WAN IP and this never drops, and I've tailed the router syslog file but nothing is logged there either.

Because of how predictable it is I was even able to test copying a large file to my NAS at the time of the drop, and the file transfer continued uninterrupted.

This happens simultaneously on all clients, wired or wireless. It doesn't happen when connecting a PC directly to the modem.

I was on Asuswrt-Merlin 384.19. I have updated to 386.2_2 as well as done a full settings reset but no difference.

If anyone has any insight into this I'd love to hear it, as I'm completely baffled. Attached my WAN settings as this is where I suspect the problem lies. I spoke to my ISP and they were helpful up to the point of diagnosing that their modem is fine and it's my router's fault. I've asked them to review my WAN settings too.

Cheers!
 

Attachments

  • wanconfig.jpg
    wanconfig.jpg
    59.2 KB · Views: 139
I got 4 years out of an AC88 until wonky stuff started happening on the WAN port. The AC88 got repurposed as an AiMesh node.
One thing that may help isolate the issue is to setup Dual WAN and connect your modem to the alternate port.
Otherwise see the full nuclear reset process on the forums. Could be a firmware glitch.
 
Weird thing is I've had continuous pings running to 8.8.8.8 from my PC and from the router CLI, as well as a 100ms interval, port 443 tcping to my work's external Citrix Gateway IP, and these are not interrupted at all - not a single packet dropped. I've also had a laptop on my phone hotspot pinging my fiber WAN IP and this never drops, and I've tailed the router syslog file but nothing is logged there either.
So the WAN hasn't dropped. Are there any error messages in the router's syslog at this time?

What are you using for your custom DHCP->DNS server(s)?
 
If anyone has any insight into this I'd love to hear it, as I'm completely baffled. Attached my WAN settings as this is where I suspect the problem lies. I spoke to my ISP and they were helpful up to the point of diagnosing that their modem is fine and it's my router's fault. I've asked them to review my WAN settings too.
Is there a need to extend TTL value?
 
So the WAN hasn't dropped. Are there any error messages in the router's syslog at this time?

What are you using for your custom DHCP->DNS server(s)?

Fair point, but it's such a weird symptom I don't know how else to describe it. All open TCP connections are closed?! Nothing is logged in /tmp/syslog.log, and I've been trying to grab a netstat immediately before and after to see if it really is dropping all open connections, but inconclusive.

DHCP > DNS would have been 8.8.8.8 & 1.1.1.1 or vice versa. My main PC has a static config the same.

I did another factory reset (still on Merlin) and initially did nothing other than run through the setup wizard. It was stable for about an hour so I began setting up the very basics again. I changed from 192.168.1.0/24 to 192.168.0.0/24, set my DHCP range, disabled UPnP and recreated my port forwards. Shortly afterwards it began happening again.

So I've done another factory reset and this time changed absolutely nothing beyond the setup wizard. That was almost 2 hours ago and it's been stable so far.

So either a setting I changed caused it to start happening again, though none of the few basic things I changed jump out as a possible culprit, or another device on my LAN is doing [something]. Leaving it on 192.168.1.0/24 means that my statically configured devices (Synology NAS, Raspberry Pi, possibly a couple others) are offline. This seems like a long shot, but I'm wondering if one of those is somehow causing it. The precise 13 minute 42 second interval is most intriguing though.

I'll switch back to 192.168.0.0/24 shortly to bring the NAS & Pi back online, change nothing else and see what happens.
 
Last edited:
You raised my interest when I read Zhone GPON ONT. I am in a similar situation (and here) as you are. Let me explain.
My Fiber ISP upgraded me to gigabit service and replaced my media converter with a DASAN Zhone ZNID-GPON-2424-A1-NA Rev. F1 GPON ONT which is in bridge mode. Also, the whole network infrastructure was changed from switches to passive splitters.

Since this upgrade my WAN Internet connections has been going down within hours or after up to 5 days. It is completely intermittent. Only rebooting the router seems to bring back the WAN Internet connection.

I tracked it down so far that it is not related to the DHCP renewal. Suddenly, the router receives no incoming traffic anymore but it still retransmits TCP packets and eventually sends ARP requests "where is everybody?". No answers.

I know all of the ONT LEDs are green and the router is connected to the ONT Ethernet port which flashes when the WAN Internet connection is down.

I strongly believe that the Zhone ONT is the culprit. Some, but not all of my neighbors that were also upgraded, have similar problems. All with the same ONT.

Btw, this was on Merlin 384.19 and now on 386.2_2 with the exact same result. Before the gigabit upgrade, I have never experienced these WAN drops that the router did not recover on its own.

Can you please tell me the exact model number of your Zhone ONT?
 
Last edited:

Similar threads

Latest threads

Sign Up For SNBForums Daily Digest

Get an update of what's new every day delivered to your mailbox. Sign up here!
Top