RT-AX58U on TC4400: Router not Automatically Re-Connecting

[RickySnoots]

Hi Folks, I turn here as maybe a last resort, and hoping some experienced users or super-users might be able to share some opinions or light.

My problem is relatively simple to explain, although over multiple calls to ISPs, Asus, firmware updates, Merlin, etc, I am coming up short.

Kindly asking what might the expertise here say about this:

My cable ISP has shoddy service, with frequent short duration internet drops. My Technicolor TC4400 goes trough it's cycle, re-syncs, and shortly thereafter reestablishes the internet connection - on the modem.. When outages occur, the Asus AX58U also loses it's connection, but when reestablished on the modem, the router does not automatically reconnect. I have to either manually flip the internet connection off and on through the GUI, or power cycle the router via rear switch.

I have tried everything to resolve this.

Asus told me to clone the MAC address, to no avail, I am on the latest ASUS FW 3.0.0.4.384_10177 sadly same issue with Merlin RT-AX58U_384.19_0.. Factory resets, hard resets, I even swapped out the Technicolor earlier today - same behavior.

Even more troubling, is this issue is starting to affect my personal life; my girlfriend/neighbor shares this internet connection via WiFi, and now knows I have to physically restart the router manually to regain access - texting me out of bed, and causing all sorts of relationship problems. This is not good!

I've searched the forum, find much more complex issues, but nothing really similarly aligned - Would anyone have any suggestions as to how I might fix this, and get the router to automatically reconnect to a reestablished internet signal after drop??

Many thanks

Rick

 

[ColinTaylor]

This can sometimes help in these sorts of situations: Go to WAN > Internet Connection and set DHCP query frequency to Continuous.

It may not solve your problem but it's worth a shot.

 

[RickySnoots]

Thanks Colin, I tried this, worked the first time, but not after a second test.

One thing I did notice was it reconnected when the details in the Internet Status panel were reset - to 0's. This last test I did, the WAN IP, Subnet, DNS and Gateway addresses all remained, yet router was in disconnected status - again, manually turning off the connection in the GUI resets these values, manually turning on the connection repopulates them and re-connection happens.

It's almost as if the router doesn't refresh itself.

 

[L&LD]

When is the last time a full reset to factory defaults performed? Without using any type of saved backup config files?

See the link in my signature below for the M&M Config and ideally the Nuclear Reset guides to get your router back to a good/known state.

 

[RickySnoots]

Hey L&LD, just the other day actually, I've done this a few times, and the router is basically brand-spanking..

I appreciate these suggestions...

 

[ColinTaylor]

If you need to resort to more drastic measures there are various user scripts that have been written for use with Merlin's firmware. They typically ping the internet and then restart the WAN interface or reboot the router if the ping fails.

See here: http://www.snbforums.com/threads/need-a-script-that-auto-reboot-if-internet-is-down.43819/

and here: https://github.com/MartineauUK/Chk-WAN

There are some other simpler scripts around the forum or you could write your own if you wanted to.

 

[RickySnoots]

Thanks again Colin, I've started to look at this, careful to not getting quickly overwhelmed - reading your post, modem drop happened again! Router disconnected, but connected again by hitting 'apply' on wan settings - which I think does the same thing as chkwan.

Amazed somewhat why the router wouldn't do this out of the box, or especially with Merlin!

 

[RickySnoots]

Ha.. This time I captured the logs:

Dec 10 12:27:24 rc_service: httpd 6200:notify_rc restart_wan_if 0
Dec 10 12:27:25 WAN Connection: WAN was exceptionally disconnected.
Dec 10 12:27:25 nat: apply redirect rules
Dec 10 12:27:26 wan: mac clone: [wan0_hwaddr] == [HOME COMPUTER MAC]
Dec 10 12:27:26 nat: apply nat rules (/tmp/nat_rules_eth4_eth4)
Dec 10 12:27:27 wan: finish adding multi routes
Dec 10 12:27:27 miniupnpd[27816]: shutting down MiniUPnPd
Dec 10 12:27:27 miniupnpd[10492]: version 1.9 started
Dec 10 12:27:27 miniupnpd[10492]: HTTP listening on port 57714
Dec 10 12:27:27 miniupnpd[10492]: Listening for NAT-PMP/PCP traffic on port 5351
Dec 10 12:27:27 kernel: Archer TCP Pure ACK Enabled
Dec 10 12:27:27 dhcp client: bound xx.xx.xx.xx/xx.xx.xx.xx via xx.xx.xx.xx for 2697 seconds.
Dec 10 12:27:39 WAN Connection: WAN was restored.

 

[RickySnoots]

Check this out. I pulled the coax to simulate an outage, and received this in the logs:

Dec 10 12:39:49 rc_service: httpd 6200:notify_rc restart_logger
Dec 10 12:39:49 kernel: klogd started: BusyBox v1.24.1 (2020-11-07 01:24:16 CST)
Dec 10 12:41:04 kernel: eth4 (Int switch port: 4) (Logical Port: 4) (phyId: c) Link DOWN.
Dec 10 12:41:09 kernel: eth4 (Int switch port: 4) (Logical Port: 4) (phyId: c) Link Up at 1000 mbps full duplex
Dec 10 12:41:13 kernel: eth4 (Int switch port: 4) (Logical Port: 4) (phyId: c) Link DOWN.
Dec 10 12:41:19 kernel: eth4 (Int switch port: 4) (Logical Port: 4) (phyId: c) Link Up at 1000 mbps full duplex

**PLUGGED IN COAX and it sits like this, disconnected**

Click the WAN off then back on, and this:

Dec 10 12:44:32 rc_service: httpd 6200:notify_rc restart_wan_if 0
Dec 10 12:44:36 WAN Connection: WAN was exceptionally disconnected.
Dec 10 12:44:36 nat: apply redirect rules
Dec 10 12:44:37 rc_service: httpd 6200:notify_rc restart_wan_if 0
Dec 10 12:44:39 wan: mac clone: [wan0_hwaddr] == [HOME COMPUTER MAC]
Dec 10 12:44:39 nat: apply nat rules (/tmp/nat_rules_eth4_eth4)
Dec 10 12:44:39 wan: finish adding multi routes
Dec 10 12:44:39 miniupnpd[10492]: shutting down MiniUPnPd
Dec 10 12:44:39 miniupnpd[15482]: version 1.9 started
Dec 10 12:44:39 miniupnpd[15482]: HTTP listening on port 41484
Dec 10 12:44:39 miniupnpd[15482]: Listening for NAT-PMP/PCP traffic on port 5351
Dec 10 12:44:39 kernel: Archer TCP Pure ACK Enabled
Dec 10 12:44:39 dhcp client: bound xx.xx.xx.xx/xx.xx.xx.xxvia xx.xx.xx.xxfor 3600 seconds.
Dec 10 12:44:56 WAN Connection: WAN was restored.

ne znayu!!

So, what might this tell someone who might (humbly) be somewhat limited in this area - maybe disable uPNP and mess around with my cables - they're brand new cat 6.

 

[RickySnoots]

ok! this time the service dropped, and captured this --

Gents, I hope you can appreciate this zeal, not intending to spam this forum board, but would greatly enjoy any perspectives on this - just this one line, modem reconnected, router sitting at this (disconnected):

Dec 10 13:29:25 wlceventd: wlceventd_proc_event(464): eth6: Deauth_ind UNKNOWN MAC!, status: 0, reason: Deauthenticated because sending station is leaving (or has left) IBSS or ESS (3)

 

[ColinTaylor]

wlceventd are normal WiFi messages and unrelated to your problem.

The WAN Connection and associated messages are what you want to see when the connection drops. The problem comes when you don't see those messages. That probably means that even though your ISP connection was reset the Ethernet connection between the modem and the router was still active so the router doesn't know what's happened. Your ISP is probably expecting the router to send another DHCP request but the router never does because it doesn't realise there's been a problem.

 

[RickySnoots]

Sure seems like that Colin, intreresting though how the Router does 'know' as it's in disconnected status! ha, what a thing..