Frequent Disconnects

[gusz]

I have an Asus RT-66U running Mernin 3.0.0.4.374.33, behind a Thompson TG585V8 DSL router, configured as a bridge, to Telmex DSL. Thus the authentication is by the Asus/Merlin thru a PPPoE connection, thru the Thompson bridge.

I am getting frequent disconnects. My Asus log for a typical disconnect is below. I have replaced the Telmex DSL router/bridge twice, and Telmex has rerun the line from their pole into my house, and says the DSL signal is good. On the Telmex DSL router/bridge I record SN Margin 22/16.5 (dn/up) and Attenuation 6.4/1.7 dn/up, which Telmex says is within spec

Every time I see a disconnect/restore I see a different IP address in the Asus/Merlin log, as in pppd[330]: local IP address 189.177.242.180 remote IP address 200.38.193.226. These are Telmex IP's - I assume remote IP is the address of their DSL server and local is the WAN IP they assign to my router - but I wonder if there's an IP change on their end that's causing the disconnect???

Can someone provide ideas how to fix this?

Thx, Gus

Dec 31 18:00:33 WAN Connection: WAN was restored.
Dec 31 18:00:37 dnsmasq-dhcp[369]: DHCPREQUEST(br0) 192.168.1.142 04:54:53:2a:38:b3
Dec 31 18:00:37 dnsmasq-dhcp[369]: DHCPACK(br0) 192.168.1.142 04:54:53:2a:38:b3 users-iPad
Dec 15 17:07:03 rc_service: ntp 405:notify_rc restart_upnp
Dec 15 17:07:03 rc_service: ntp 405:notify_rc restart_diskmon
Dec 15 17:07:03 rc_service: restart_diskmon is waitting restart_upnp...
Dec 15 17:07:03 miniupnpd[414]: HTTP listening on port 43883
Dec 15 17:07:03 miniupnpd[414]: Listening for NAT-PMP traffic on port 5351
Dec 15 17:07:04 disk monitor: be idle
Dec 15 17:07:04 crond[303]: time disparity of 1555146 minutes detected
Dec 15 17:07:07 pppd[330]: System time change detected.
Dec 15 17:08:31 dnsmasq-dhcp[369]: DHCPREQUEST(br0) 192.168.1.212 f0:d1:a9:f2:29:0d
Dec 15 17:08:31 dnsmasq-dhcp[369]: DHCPACK(br0) 192.168.1.212 f0:d1:a9:f2:29:0d Randys-iPad-2
Dec 15 17:13:44 dnsmasq-dhcp[369]: DHCPREQUEST(br0) 192.168.1.212 f0:d1:a9:f2:29:0d
Dec 15 17:13:44 dnsmasq-dhcp[369]: DHCPACK(br0) 192.168.1.212 f0:d1:a9:f2:29:0d Randys-iPad-2
Dec 15 17:21:33 pppd[330]: No response to 10 echo-requests
Dec 15 17:21:33 pppd[330]: Serial link appears to be disconnected.
Dec 15 17:21:33 pppd[330]: Connect time 15.1 minutes.
Dec 15 17:21:33 pppd[330]: Sent 233768 bytes, received 767333 bytes.
Dec 15 17:21:33 miniupnpd[414]: Failed to get IP for interface ppp0
Dec 15 17:21:33 miniupnpd[414]: SendNATPMPPublicAddressChangeNotification: cannot get public IP address, stopping
Dec 15 17:21:33 dnsmasq[369]: read /etc/hosts - 5 addresses
Dec 15 17:21:33 dnsmasq[369]: using nameserver 208.67.220.220#53
Dec 15 17:21:33 dnsmasq[369]: using nameserver 8.8.8.8#53
Dec 15 17:21:37 WAN Connection: Fail to connect with some issues.
Dec 15 17:21:37 stop_nat_rules: apply the redirect_rules!
Dec 15 17:21:37 dnsmasq[369]: exiting on receipt of SIGTERM
Dec 15 17:21:37 dnsmasq[458]: started, version 2.65 cachesize 1500
Dec 15 17:21:37 dnsmasq[458]: compile time options: IPv6 GNU-getopt no-RTC no-DBus no-i18n no-IDN DHCP DHCPv6 no-Lua TFTP no-conntrack
Dec 15 17:21:37 dnsmasq[458]: asynchronous logging enabled, queue limit is 5 messages
Dec 15 17:21:37 dnsmasq-dhcp[458]: DHCP, IP range 192.168.1.100 -- 192.168.1.252, lease time 1d
Dec 15 17:21:37 dnsmasq[458]: read /etc/hosts - 5 addresses
Dec 15 17:21:37 dnsmasq[458]: using nameserver 208.67.220.220#53
Dec 15 17:21:37 dnsmasq[458]: using nameserver 8.8.8.8#53
Dec 15 17:21:39 pppd[330]: Connection terminated.
Dec 15 17:21:39 pppd[330]: Sent PADT
Dec 15 17:21:39 pppd[330]: Modem hangup
Dec 15 17:21:49 pppd[330]: PPP session is 35686 (0x8b66)
Dec 15 17:21:49 pppd[330]: Connected to 00:1a:a1:d6:77:78 via interface eth0
Dec 15 17:21:49 pppd[330]: Using interface ppp0
Dec 15 17:21:49 pppd[330]: Connect: ppp0 <--> eth0
Dec 15 17:21:52 pppd[330]: PAP authentication succeeded
Dec 15 17:21:52 pppd[330]: peer from calling number 00:1A:A16:77:78 authorized
Dec 15 17:21:52 miniupnpd[414]: Failed to get IP for interface ppp0
Dec 15 17:21:52 miniupnpd[414]: SendNATPMPPublicAddressChangeNotification: cannot get public IP address, stopping
Dec 15 17:21:52 miniupnpd[414]: Failed to get IP for interface ppp0
Dec 15 17:21:52 miniupnpd[414]: SendNATPMPPublicAddressChangeNotification: cannot get public IP address, stopping
Dec 15 17:21:52 miniupnpd[414]: Failed to get IP for interface ppp0
Dec 15 17:21:52 miniupnpd[414]: SendNATPMPPublicAddressChangeNotification: cannot get public IP address, stopping
Dec 15 17:21:52 pppd[330]: local IP address 187.144.167.198
Dec 15 17:21:52 pppd[330]: remote IP address 200.38.193.226
Dec 15 17:21:52 dnsmasq[458]: read /etc/hosts - 5 addresses
Dec 15 17:21:52 dnsmasq[458]: using nameserver 208.67.220.220#53
Dec 15 17:21:52 dnsmasq[458]: using nameserver 8.8.8.8#53
Dec 15 17:21:52 start_nat_rules: apply the nat_rules(/tmp/nat_rules_ppp0_eth0)!
Dec 15 17:21:52 dnsmasq[458]: exiting on receipt of SIGTERM
Dec 15 17:21:52 dnsmasq[486]: started, version 2.65 cachesize 1500
Dec 15 17:21:52 dnsmasq[486]: compile time options: IPv6 GNU-getopt no-RTC no-DBus no-i18n no-IDN DHCP DHCPv6 no-Lua TFTP no-conntrack
Dec 15 17:21:52 dnsmasq[486]: asynchronous logging enabled, queue limit is 5 messages
Dec 15 17:21:52 dnsmasq-dhcp[486]: DHCP, IP range 192.168.1.100 -- 192.168.1.252, lease time 1d
Dec 15 17:21:52 dnsmasq[486]: read /etc/hosts - 5 addresses
Dec 15 17:21:52 dnsmasq[486]: using nameserver 208.67.220.220#53
Dec 15 17:21:52 dnsmasq[486]: using nameserver 8.8.8.8#53
Dec 15 17:21:53 rc_service: ip-up 468:notify_rc stop_upnp
Dec 15 17:21:53 rc_service: ip-up 468:notify_rc start_upnp
Dec 15 17:21:53 miniupnpd[501]: HTTP listening on port 60299
Dec 15 17:21:53 miniupnpd[501]: Listening for NAT-PMP traffic on port 5351
Dec 15 17:21:56 WAN Connection: WAN was restored.

 

[KGB7]

Ive been having some type of disconnects or internet drops for the past 4 hours.

I checked Router logs and Modem logs and its all clear. Rebooted router several times, and connection goes back up and i have internet, but after 15min i lose internet again.
Basic internet surfing there seems to be no issue, but when there is constant data flow, no matter how low or how high, the connection gets disrupted.

I dont know who to blame anymore; my isp, this firmware, asus them selfs...??


My connection was a lot more stable with OEM .371 and .454.


Something weird is going on. I set up an RT-N66U router for a buddy of mine two weeks ago using OEM firmware, and its been running solid. Dont recall what firmware its running, i did an update through routers GUI. All his devices are connected through wifi 2.4Ghz. He has about 10 devices connected to it and hasnt complained once.




gusz

Install factory firmware and see if you still having issues. Because tomorrow morning im going back to factory/oem firmware.

 

[gusz]

Went back to (latest) Asus firmware. No joy, still frequent disconnects. Telmex has been out here three times - they can't figure it out either. Gus

 

[RMerlin]

Your line stats are indeed pretty good, so I don't think it's a line issue, unless it's intermittent.

You do have line filters installed on all your other phone devices in the house (assuming you are using your main phone line and not a separate dry loop)?

Make sure that on the WAN page, you have Idle Disconnect set to "0".

Beside that I don't have any other idea, sorry.

 

[gusz]

I was able to borrow and try a different router, Ubiquiti RSpro running OpenWRT 12.09, still behind the same Thompson TG585V8 configured as a bridge. Seems to work perfectly, no disconnects in 24 hours.

Regards comments above:

> have line filters installed on all your other phone devices
Yes
> have Idle Disconnect set to "0"
Yes

I'd like to "save" the RT-AC66U, which cost me $150 bucks and has great specs. I'm guessing the problem is somewhere in the WAN/ethernet/PPPoE connection, but I'm hardly an expert. Is there a way to do further diagnostics, determine what's causing the disconnects, whether it's hardware or software, etc.? Thx, Gus

 

[RMerlin]

I'd like to "save" the RT-AC66U, which cost me $150 bucks and has great specs. I'm guessing the problem is somewhere in the WAN/ethernet/PPPoE connection, but I'm hardly an expert. Is there a way to do further diagnostics, determine what's causing the disconnects, whether it's hardware or software, etc.? Thx, Gus

I've heard that some ISPs are having issues with the more recent versions of rp-pppoe. Asus upgraded to 3.11 with FW 354. As a test, you could try making a backup of your settings (just in case), then downgrading to 270.26b, and see if after a day or two it's more stable. If it is, I could look at the possibility of downgrading to 3.10 (I haven't downgraded yet because, so far nobody could confirm to me that 3.11 was really less stable than 3.10). I had been skeptical because 3.11 was released a few years ago, and its author hadn't released any update, so I was led to believe that the reports of its instability might have been anecdotal at beast.

 

[gusz]

Thanks, and I will try it and report back.

Do you think there's a possibility that this is a hardware problem - say, causing lost packets and therefore disconnects?

As a note, my Asus -> Thompson WAN/PPPoE cable is Cat6, and I've tried different cables, problem persists.

Gus

 

[RMerlin]

Thanks, and I will try it and report back.

Do you think there's a possibility that this is a hardware problem - say, causing lost packets and therefore disconnects?

As a note, my Asus -> Thompson WAN/PPPoE cable is Cat6, and I've tried different cables, problem persists.

Gus

It's possible, tho the first thing I would have suspected with your symptoms would be the modem - which you already replaced, and the phone line - which shows good sync stats.

I don't think it's a HW issue with the router, because you would have seen log entries about the Ethernet port going down if that were the case. Plus, your router doesn't crash - it merely loses its PPPoE session with the ISP.

 

[junkone]

just my 2 cents. i had a similar issue last week and finally figured out that when i swithced from dsl to cable and vice versa, i had to reset factory settings. otherwise something wierd was happening.

once i reset factory settings. my connection via dsl or via cable was stable. but every time i switched, i had to reset my factory settings

 

[sinshiva]

is hw acceleration considered ruled out as a source of instability?

 

[RMerlin]

is hw acceleration considered ruled out as a source of instability?

HW acceleration-related issues tend to cause an actual router reboot. In this case, what seems to happen is that the PPPoE session simply times out.

Might still be worth testing with HW acceleration disabled just to be sure.

 

[gusz]

> try downgrading to 270.26b, and see if after a day or two it's more stable.

I did, and amazingly, things seem better. About 20 hours on downgrade now, and only one log AsusWRT-Merlin log message since reconnect using new version:

Dec 19 23:38:26 miniupnpd[613]: recv (state0): Connection reset by peer

I think we should wait a day or two before concluding that's the fix, and I'll post back on progress, but it seems promising. BTW my ISP is Telmex in Mexico.

On the RT-AC66U 270.26b downgrade I see Samba messages. Does Merlin/Samba only work when there's USB storage attached to the router?

Thanks, RMerlin, for sticking with this!

-------------------------------
Update: well, with 270.26b, after about 48 hours, the frequent disconnects started happening again. Now I'm really stumped. A typical log for a disconnect is

Dec 21 15:19:08 WAN Connection: Ethernet link down.
Dec 21 15:19:08 notify_rc : stop_nat_rules
Dec 21 15:19:08 stop_nat_rules: apply the redirect_rules!
Dec 21 15:19:41 pppd[484]: Serial link appears to be disconnected.
Dec 21 15:19:41 miniupnpd[981]: Failed to get IP for interface ppp0
Dec 21 15:19:41 miniupnpd[981]: SendNATPMPPublicAddressChangeNotification: cannot get public IP address, stopping
Dec 21 15:19:41 dnsmasq[548]: failed to load names from /etc/hosts.dnsmasq: No such file or directory
Dec 21 15:19:47 pppd[484]: Connection terminated.
Dec 21 15:19:48 pppd[484]: Modem hangup
Dec 21 15:20:33 pppd[484]: Timeout waiting for PADO packets
Dec 21 15:20:33 pppd[484]: Unable to complete PPPoE Discovery
Dec 21 15:20:38 WAN Connection: Fail to connect with some issues.
Dec 21 15:21:18 pppd[484]: Timeout waiting for PADO packets
Dec 21 15:21:18 pppd[484]: Unable to complete PPPoE Discovery
Dec 21 15:21:28 pppd[484]: Connected to 00:1a:a1:d6:77:78 via interface eth0
Dec 21 15:21:28 pppd[484]: Connect: ppp0 <--> eth0
Dec 21 15:21:31 pppd[484]: PAP authentication succeeded
Dec 21 15:21:31 pppd[484]: peer from calling number 00:1A:A16:77:78 authorized
Dec 21 15:21:31 miniupnpd[981]: Failed to get IP for interface ppp0
Dec 21 15:21:31 miniupnpd[981]: SendNATPMPPublicAddressChangeNotification: cannot get public IP address, stopping
Dec 21 15:21:31 miniupnpd[981]: Failed to get IP for interface ppp0
Dec 21 15:21:31 miniupnpd[981]: SendNATPMPPublicAddressChangeNotification: cannot get public IP address, stopping
Dec 21 15:21:31 miniupnpd[981]: Failed to get IP for interface ppp0
Dec 21 15:21:31 miniupnpd[981]: SendNATPMPPublicAddressChangeNotification: cannot get public IP address, stopping
Dec 21 15:21:31 pppd[484]: local IP address 187.144.165.32
Dec 21 15:21:31 pppd[484]: remote IP address 200.38.193.226
Dec 21 15:21:31 pppd[484]: primary DNS address 200.23.242.227
Dec 21 15:21:31 pppd[484]: secondary DNS address 200.23.242.211
Dec 21 15:21:31 miniupnpd[981]: ioctl(s, SIOCGIFADDR, ...): Cannot assign requested address
Dec 21 15:21:31 miniupnpd[981]: Failed to get IP for interface ppp0
Dec 21 15:21:31 miniupnpd[981]: SendNATPMPPublicAddressChangeNotification: cannot get public IP address, stopping
Dec 21 15:21:31 dnsmasq[548]: failed to load names from /etc/hosts.dnsmasq: No such file or directory
Dec 21 15:21:31 notify_rc : start_nat_rules
Dec 21 15:21:31 start_nat_rules: apply the nat_rules!
Dec 21 15:21:31 notify_rc : stop_upnp
Dec 21 15:21:31 notify_rc : start_upnp
Dec 21 15:21:31 miniupnpd[2243]: HTTP listening on port 49195
Dec 21 15:21:31 miniupnpd[2243]: Listening for NAT-PMP traffic on port 5351
Dec 21 15:21:33 WAN Connection: WAN was restored.