WAN connection drops

[cassandra]

Hello again guys, I am sadly back with a new issue.

Ever since upgrading to 386.1 and subsequently to 386.2 beta 1 I am experiencing consistent connection drops on my ac68u. I don't think its my ISP supplied FTTH modem fault, since I don't have any issues if I take the modem out of bridge mode and connect directly to it. I also don't think I have my PPPoE parameters set to wrong values, since as I mentioned before, I had 0 drops on exactly the same config with older versions of merlin firmware.

If anybody has any idea what is wrong and is willing to help out I will be very grateful, since I exhausted all my knowledge.

Here's the log:

Mar 11 17:06:48 syslog: wlceventd_proc_event(491): eth2: Deauth_ind 00:0A:F5:60:AF:C8, status: 0, reason: Disassociated due to inactivity (4), rssi:0
Mar 11 17:07:37 syslog: wlceventd_proc_event(491): eth1: Deauth_ind 68:94:23:4B:71:4B, status: 0, reason: Disassociated due to inactivity (4), rssi:0
Mar 11 17:07:48 syslog: wlceventd_proc_event(491): eth1: Deauth_ind F4:60:E2:F1:E8:E4, status: 0, reason: Disassociated due to inactivity (4), rssi:0
Mar 11 17:07:48 syslog: wlceventd_proc_event(491): eth1: Deauth_ind 68:E7:C2:CF:65:EF, status: 0, reason: Disassociated due to inactivity (4), rssi:0
Mar 11 17:20:00 Diversion: updated ads counter: 235,712 total, 19,985 this week, 2,929 new since last count
Mar 11 18:00:27 uiDivStats: Stats updated successfully
Mar 11 18:13:43 syslog: wlceventd_proc_event(527): eth2: Auth 00:0A:F5:60:AF:C8, status: Successful (0), rssi:0
Mar 11 18:13:43 syslog: wlceventd_proc_event(556): eth2: Assoc 00:0A:F5:60:AF:C8, status: Successful (0), rssi:0
Mar 11 18:14:32 syslog: wlceventd_proc_event(527): eth1: Auth 68:E7:C2:CF:65:EF, status: Successful (0), rssi:0
Mar 11 18:14:32 syslog: wlceventd_proc_event(556): eth1: Assoc 68:E7:C2:CF:65:EF, status: Successful (0), rssi:0
Mar 11 18:21:05 syslog: wlceventd_proc_event(527): eth1: Auth F4:60:E2:F1:E8:E4, status: Successful (0), rssi:0
Mar 11 18:21:05 syslog: wlceventd_proc_event(556): eth1: Assoc F4:60:E2:F1:E8:E4, status: Successful (0), rssi:0
Mar 11 18:22:01 syslog: wlceventd_proc_event(527): eth1: Auth 74:C6:3B:BB:47:7F, status: Successful (0), rssi:0
Mar 11 18:22:01 syslog: wlceventd_proc_event(556): eth1: Assoc 74:C6:3B:BB:47:7F, status: Successful (0), rssi:0
Mar 11 18:39:03 kernel: nvram: consolidating space!
Mar 11 18:41:04 syslog: wlceventd_proc_event(508): eth1: Disassoc 74:C6:3B:BB:47:7F, status: 0, reason: Disassociated because sending station is leaving (or has left) BSS (8), rssi:0
Mar 11 18:41:13 syslog: wlceventd_proc_event(527): eth1: Auth 74:C6:3B:BB:47:7F, status: Successful (0), rssi:0
Mar 11 18:41:13 syslog: wlceventd_proc_event(556): eth1: Assoc 74:C6:3B:BB:47:7F, status: Successful (0), rssi:0
Mar 11 19:00:28 uiDivStats: Stats updated successfully
Mar 11 19:22:19 syslog: wlceventd_proc_event(508): eth1: Disassoc 74:C6:3B:BB:47:7F, status: 0, reason: Disassociated because sending station is leaving (or has left) BSS (8), rssi:0
Mar 11 19:33:10 kernel: nvram: consolidating space!
Mar 11 19:56:46 pppd[1775]: Connect time 173.6 minutes.
Mar 11 19:56:46 pppd[1775]: Sent 3608629221 bytes, received 1239791623 bytes.
Mar 11 19:56:46 wsdd2[2563]: error: wsdd-mcast-v4: wsd_send_soap_msg: send
Mar 11 19:56:46 lldpd[184]: removal request for address of 94.127.30.1%13, but no knowledge of it
Mar 11 19:56:49 WAN_Connection: Fail to connect with some issues.
Mar 11 19:57:16 pppd[1775]: LCP: timeout sending Config-Requests
Mar 11 19:57:16 pppd[1775]: Connection terminated.
Mar 11 19:57:16 pppd[1775]: Sent PADT
Mar 11 19:57:16 pppd[1775]: Modem hangup
Mar 11 19:57:26 pppd[1775]: PPP session is 1 (0x1)
Mar 11 19:57:26 pppd[1775]: Connected to a4:7b:2c:9c:9f:44 via interface eth0
Mar 11 19:57:26 pppd[1775]: Using interface ppp0
Mar 11 19:57:26 pppd[1775]: Connect: ppp0 <--> eth0
Mar 11 19:57:26 pppd[1775]: CHAP authentication succeeded: CHAP authentication success
Mar 11 19:57:26 pppd[1775]: CHAP authentication succeeded
Mar 11 19:57:26 pppd[1775]: peer from calling number A4:7B:2C:9C:9F:44 authorized
Mar 11 19:57:26 lldpd[184]: removal request for address of 146.212.223.4%14, but no knowledge of it
Mar 11 19:57:26 wsdd2[2563]: error: wsdd-mcast-v4: wsd_send_soap_msg: send
Mar 11 19:57:26 pppd[1775]: local  IP address 146.212.223.4
Mar 11 19:57:26 pppd[1775]: remote IP address 94.127.30.1
Mar 11 19:57:26 lldpd[184]: removal request for address of 94.127.30.1%14, but no knowledge of it
Mar 11 19:57:26 wsdd2[2563]: error: wsdd-mcast-v4: wsd_send_soap_msg: send
Mar 11 19:57:26 lldpd[184]: removal request for address of 146.212.223.4%14, but no knowledge of it
Mar 11 19:57:26 lldpd[184]: removal request for address of 146.212.223.4%14, but no knowledge of it
Mar 11 19:57:27 custom_script: Running /jffs/scripts/firewall-start (args: ppp0)
Mar 11 19:57:27 wan: finish adding multi routes
Mar 11 19:57:27 rc_service: ip-up 14442:notify_rc stop_samba
Mar 11 19:57:27 custom_script: Running /jffs/scripts/service-event (args: stop samba)
Mar 11 19:57:27 rc_service: ip-up 14442:notify_rc start_samba
Mar 11 19:57:27 rc_service: waitting "stop_samba" via ip-up ...
Mar 11 19:57:27 wsdd2[2563]: Terminated received.
Mar 11 19:57:27 wsdd2[2563]: terminating.
Mar 11 19:57:28 Samba_Server: smb daemon is stopped
Mar 11 19:57:28 kernel: gro disabled
Mar 11 19:57:28 custom_script: Running /jffs/scripts/service-event (args: start samba)
Mar 11 19:57:29 WAN_Connection: WAN was restored.
Mar 11 19:57:29 custom_script: Running /jffs/scripts/dnsmasq.postconf (args: /etc/dnsmasq.conf)
Mar 11 19:57:32 Diversion: restarted Dnsmasq to apply settings
Mar 11 19:57:32 uiDivStats: dnsmasq has restarted, restarting taildns
Mar 11 19:57:32 kernel: gro enabled with interval 2
Mar 11 19:57:32 Samba_Server: daemon is started
Mar 11 19:57:33 wsdd2[15853]: starting.
Mar 11 19:57:38 GsJ4cYdYVuUFMCh: Started taildns from .

Edit: Here's the whole log: https://pastebin.com/uN7J74Bd

 

[drinkingbird]

By drops do you mean packet loss or total connection loss?

If total loss looks like you're having the same issue we're discussing the FIOS thread. WAN DHCP seems to be broken in these releases.

When it happens, go to your internet status in the router, click to turn internet off then back on. If it is the same issue you'll get a new WAN IP and connection will start working again.

 

[alex1236hk]

By drops do you mean packet loss or total connection loss?

If total loss looks like you're having the same issue we're discussing the FIOS thread. WAN DHCP seems to be broken in these releases.

When it happens, go to your internet status in the router, click to turn internet off then back on. If it is the same issue you'll get a new WAN IP and connection will start working again.

AC86U here. Having that issue also, I have FTTH installed but I am not with FIOS.
by turning internet off then back on in status page does solve that issue, but the internet will still drops out randomly in 1 or 2 days.
Sigh.... I dont really want to go back to 384.19 since I need AiMesh 2.0.

 

[cassandra]

By drops do you mean packet loss or total connection loss?

If total loss looks like you're having the same issue we're discussing the FIOS thread. WAN DHCP seems to be broken in these releases.

When it happens, go to your internet status in the router, click to turn internet off then back on. If it is the same issue you'll get a new WAN IP and connection will start working again.

Yes, I meant total connection loss, sorry about ambiguity. It might be the same issue as the FIOS problem, however my connection restores automatically.

 

[Vexira]

Try continuous mode instead of agresssive or normal in the the Wan tab.

 

[cassandra]

Try continuous mode instead of agresssive or normal in the the Wan tab.

I have actually set PPPoE as WAN connection type, since automatic IP doesn't work, it errors out with "You ISP DHCP isn't functioning properly". I am not sure if this is the same issue as in FIOS thread, since I think my ISP specifies that I need to setup connection as PPPoE

 

[cassandra]

I found somebody who had the exactly the same issue on openwrt system. On openwrt as far as I understood, the issue was fixed with kernel patch. Is there anyway to adopt this patch to merlin?
Here's the thread on openwrt forums: https://forum.openwrt.org/t/pppoe-disconnects-every-few-hours/61239/70

 

[Vexira]

I have actually set PPPoE as WAN connection type, since automatic IP doesn't work, it errors out with "You ISP DHCP isn't functioning properly". I am not sure if this is the same issue as in FIOS thread, since I think my ISP specifies that I need to setup connection as PPPoE

Try automatic IP, then set it to continuous mode that's what I'm using with my ISP the errors are gone and have been gone for ages.

 

[Vexira]

Also try a Mac clone of an isp router to see if that helps, I know form experience because my ISP is painful like that.

 

[cassandra]

Sadly none of these workarounds work, automatic IP even in continuous mode still errors out with "your isp dhcp is not functioning properly" and no internet connectivity. Same goes for mac cloning, if I clone my FTTH modem mac, I loose my internet connection.

Its really strange, since as I mentioned before something must have changed from 384.19, I am guessing by ASUS. I would report the issue to them, but I doubt it will get fixed anytime soon, since as it seem it is not very widespread.

 

[Vexira]

Sadly none of these workarounds work, automatic IP even in continuous mode still errors out with "your isp dhcp is not functioning properly" and no internet connectivity. Same goes for mac cloning, if I clone my FTTH modem mac, I loose my internet connection.

Its really strange, since as I mentioned before something must have changed from 384.19, I am guessing by ASUS. I would report the issue to them, but I doubt it will get fixed anytime soon, since as it seem it is not very widespread.

Did you reboot after the Mac clone and the continuous mode?

 

[cassandra]

Yes, mac clone errors out with pppd[25693]: Unable to complete PPPoE Discovery

 

[Vexira]

Yes, mac clone errors out with pppd[25693]: Unable to complete PPPoE Discovery

So did your ISP give you a username and password to sign in or just a modem?

 

[cassandra]

So did your ISP give you a username and password to sign in or just a modem?

Both, I got a modem that's supposedly in bridge and username/password.

 

[Vexira]

Both, I got a modem that's supposedly in bridge and username/password.

If you were supplied with login credentials then you should be set to pppoe rather than automatic IP which I have which is dhcp form my ISP (ipoe).

 

[cassandra]

If you were supplied with login credentials then you should be set to pppoe rather than automatic IP which I have which is dhcp form my ISP (ipoe).

Yes, I have PPPoE however its still dropping connection occasionally.

 

[Vexira]

I would disable rebind protection and DNS sec see if that helps and see if removing the DNS over TLS helps.

Also when was the last time you preformed a factory reset?

 

[cassandra]

Hmm, I haven't tried removing dnssec, rebind protection and DoT, will try it now.

I reset the router yesterday, its been a bit more stable since then, however its still bothered by something as I mentioned above. I noticed that whenever I get "lldpd[184]: removal request for address of (WAN IP)%, but no knowledge of it" in the syslog that guarantees that the connection will drop eventually. The strange thing is that I don't always get this lldpd message in the syslog and I have no idea what is affecting it. And whenever that lldpd is missing from the syslog the connection is more or less stable.

 

[Vexira]

Hmm, I haven't tried removing dnssec, rebind protection and DoT, will try it now.

I reset the router yesterday, its been a bit more stable since then, however its still bothered by something as I mentioned above. I noticed that whenever I get "lldpd[184]: removal request for address of (WAN IP)%, but no knowledge of it" in the syslog that guarantees that the connection will drop eventually. The strange thing is that I don't always get this lldpd message in the syslog and I have no idea what is affecting it. And whenever that lldpd is missing from the syslog the connection is more or less stable.

Are you running QoS by any chance or so protection, if not then it seem like your ISP has some sort of short lease time, i would Google what is my ip to verify if the router is receiving the public IP address which confirms that your in bridge mode with the mode.

Is it possible to get a replacement modem or is it on your wall, here in Australia they have what is called a network termination device which is basically the modem for fiber-optic connections that is mounted on the way generally.

I would ask your ISP about Thier lease times.

 

[cassandra]

I am not running QoS or any protection as far as I am aware. Modem replacement would theoretically be possible, as it is a standalone unit, but I doubt it would happen. ISP would most likely just tell me to put modem out of bridge and connect to it directly.

Anyway I disabled all DNS protection/privacy settings, I will have to wait and see if the router drops connection again.

Thanks for all your help.