Is this a bug on Asus firmware? WAN disconnections

[ryball]

Hey, all. I joined just so I could add my voice to this issue. I am having identical issues as BlowfishBoy.

http://www.smallnetbuilder.com/forums/showpost.php?p=119371&postcount=127

I am on fiber, so no modem. I am on CAT6 cable. My ancient Belkin pre-n router has no issues dropping.

I have gotten some great ideas from this thread and am going to try some when I get back in front of the router.

changing DHCP settings to normal
cloning the MAC

I also read somewhere (don't think it was this thread) about dropping to WPA instead of WPA2.

 

[ryball]

Reporting back, the MAC cloning worked for me. It's been stable for three days. SMH.

 

[Draghmar]

Hi
I have this problem for two days on AC66U with 376.49_4 firmware. I don't have modem between - just router connected directly do ISP. It's all stated yesterday when ISP had some problems. I had many drops but they told me that they fixed everything. They've also mentioned that they did something with their DHCP server but without any details.

Now my router drops WAN IP, Subnet mask and Gateway to 0.0.0.0 between 5-35min <- there's no specific time frame. Easiest way to get internet back is to click Apply on WAN page, even without changing anything.

I'm trying to find some tips of what can be wrong. Especially I'd like to confirm that it isn't problem on my side.
I've done this 10min disconnection just to be sure. I've checked cable: cat 5e. And I've tried MAC cloning - this didn't work completely as I need it to be like it was set by ISP.
All I did, that was somehow unusual from normal day to day, was enabling Dual WAN when ISP had its issues to use my phone as a secondary WAN. It's disabled right now.

I tried many different settings but nothing seems to help. These are logs from time when router is losing assigned IP:
1. Automatic IP

Jan  5 21:55:14 WAN(0) Connection: ISP's DHCP did not function properly.
Jan  5 21:55:14 stop_nat_rules: apply the redirect_rules!
Jan  5 21:55:20 dnsmasq[432]: exiting on receipt of SIGTERM
Jan  5 21:55:20 dnsmasq[755]: started, version 2.72test3 cachesize 1500
Jan  5 21:55:20 dnsmasq[755]: warning: interface ppp1* does not currently exist
Jan  5 21:55:20 dnsmasq[755]: asynchronous logging enabled, queue limit is 5 messages
Jan  5 21:55:20 dnsmasq-dhcp[755]: DHCP, IP range 192.168.2.99 -- 192.168.2.125, lease time 1d
Jan  5 21:55:20 dnsmasq[755]: read etc-hosts - 5 addresses
Jan  5 21:55:20 dnsmasq[755]: using nameserver 91.90.160.5#53 for domain connecta.pl
Jan  5 21:55:20 dnsmasq[755]: using nameserver 91.90.160.4#53 for domain connecta.pl
Jan  5 21:55:20 dnsmasq[755]: using nameserver 91.90.160.4#53
Jan  5 21:55:20 dnsmasq[755]: using nameserver 91.90.160.5#53

2. Static IP - this is weird because I don't use PPPoE - don't have any login/password etc

Jan  5 22:08:01 WAN(0) Connection: Detected that the WAN Connection Type was PPPoE. But the PPPoE Setting was not complete.
Jan  5 22:08:01 stop_nat_rules: apply the redirect_rules!
Jan  5 22:08:07 dnsmasq[840]: exiting on receipt of SIGTERM
Jan  5 22:08:07 dnsmasq[1386]: started, version 2.72test3 cachesize 1500
Jan  5 22:08:07 dnsmasq[1386]: warning: interface ppp1* does not currently exist
Jan  5 22:08:07 dnsmasq[1386]: asynchronous logging enabled, queue limit is 5 messages
Jan  5 22:08:07 dnsmasq-dhcp[1386]: DHCP, IP range 192.168.2.99 -- 192.168.2.125, lease time 1d
Jan  5 22:08:07 dnsmasq[1386]: read etc-hosts - 5 addresses
Jan  5 22:08:07 dnsmasq[1386]: using nameserver 91.90.160.4#53
Jan  5 22:08:07 dnsmasq[1386]: using nameserver 91.90.160.5#53

I've manage to catch this thing also:

Jan  5 22:11:23 dhcp client: bound 91.90.164.11 via 91.90.164.9 during 86400 seconds.
Jan  5 22:11:36 WAN(0) Connection: WAN was restored.

And I don't know if this has anything to do with this issue but on Tools->Sysinfo there is a table with ports status and WAN has vlan set to 2. Maybe that was like that all the time - unfortunately I don't remember this.

WAN	2	100 Full Duplex	00:1A:A0:5B:99:E3
LAN 1	1	Down	<none>
LAN 2	1	1000 Full Duplex	
LAN 3	1	1000 Full Duplex	
LAN 4	1	100 Full Duplex

Is there anything I could do to test something? We have holiday right now so no one's going to help me from ISPs support.

 

[Draghmar]

Update - the problem suddenly disappeared...I mean, it was there in the morning but after failed attempt to configure DualWAN I turn it off, made reboot and it works from that time (over 40h).
I have no idea what was that...and I really hope it will not come back.

 

[mavack]

Im not sure if this is the same or not, it seam similar, but since updating to 376.49 i have been getting this during the night. The ISP modem going down is one problem and i have caught it occurring when i'm at the computer. However the WAN DHCP process gets the IP but then falls over later.


pasted here since cloudflare is blocking pasting of code....
http://pastebin.com/w8y1vGSB


This has been occurring during the night and when i get up i just click the disconnected button on the asus and it auto detects and dhcp works fine then. I've tried aggressive/normal dhcp mode no difference




Why is this process working and then failing?
All of the AAA.BBB.CCC.DDD instances are a valid WAN IP just changed. Additionally they are the same WAN ip, it does change but rarely because lease is long so you can usually get the same IP back.

 

[MilkManX]

Well I installed the latest MERLIN on my RT-N66U and it still dropped off.

Now I had to restart both the SB 6120 Modem and the 66U to get my internet back at about 7:50pm last night. I have an auto timer that reboots the router every morning at 0500.

Here is the log.

*cant post it*

Dropbox link

https://www.dropbox.com/s/glh7l9c5jdlsn1q/syslog.txt?dl=0

It craps out at 17:42 with DNSMASQ to various devices in the house and the log just drops.


My 6120 is over 3 years old at this point. I am thinking of pulling the trigger on the newer SB 6141 Modem unless you guys think maybe the n66U is crapping out? I have only had it since Sep 2014.

 

[mavack]

It craps out at 17:42 with DNSMASQ to various devices in the house and the log just drops.

Thats not a log of a device crapping out, DNSMASQ is functioning fine, it's doing what it's supposed to do dnsmasq is client side anyway not WAN side.


If your old SB is crapping out and needs rebooting to restore service you should log into the webpage on the modem and check the log (should be 192.168.100.1) you might be able to see if the modem is going down and maybe you should log it with your ISP if thats the case.

If i get a few more during the night i will also be going down that path, my timezone was wrong so i was unsure of the times in the log, but now i have it going to a syslog server to catch it properly next time.

I have no problems with the WAN dropping out from time to time, however the router should restore DHCP correctly, or at least retry at some point?

 

[MilkManX]

Well you know that guy Murphy. I stopped and picked up a new SB6141 and tonight the SB6120 is going great. It actually has a faster speediest today than it has had in days. I am starting to think my cable provider is messing around with its network....

 

[vaszago]

How can I reset the dhcp client.

dhclient -r does not work on putty

 

[mstombs]

The router will be using udhcpc for wan dhcp, I assume the following still current:-

http://www.snbforums.com/threads/is...-and-renew-wan-ip-addresses.10187/#post-62330

 

[Rankdropper84]

Not really sure what all has been said and what hasn't so i will just say what i would do. If using a cable modem go to cable modems webpage (usually 192.168.100.1) and check out the signal levels. You want downstream to be between -8 and +8 with 0 being the best. Upstream you want it to typically be anywhere from 35-50 although i have seen some work even higher than 55. Most importantly the signal to noise ratio. 40 means that your connection is noise free and the lower the number the more noise there is. 35 and up should be considered good although I try to get this as good as possible.

Check your modem. If your modem is disconnecting or having a problem then it will list it in its log which from there you can google the code or post back here.

You could always do a simple ping test with a big packet size also to test for bad cables since the bad cables would report a higher than normal latency or even drop packets. Latency should be in most cases always 1ms with small spikes ok. Recommend you also test the cables with a large packet size. Also you could check the latency from your computer to router and to the modem etc. Which if your wired the latency should be 1ms to both. Anything other and you have to wonder why.

Also have seem multiple people having problems with this router but have never reset it (cleared the nvram) so please try that also as menial as it may sound. Also want to add my AC68U holds it connection steady even with what seems like a never ending amount of botherdering going on. I also only have cat 5e cables and almost all of which are at least 4 years old or more

 

[mstombs]

I found out this morning that my ISP changed modem config (free upgrade!) and the firmware wand_duck? and my own wan_check script failed - the SIGUSR1 syntax in the above link didn't seem to work on current firmware, what did work is

killall -USR1 udhcpc

or maybe it was just trying a second time....

 

[toqchista]

Hi guys .

your are can tell me what is the solution to this problem?

one person have similar problem asus router and what should do i that this problem resolved ?

if cause is asus firemware bug /?



ofter only 4 second don't turn off often turn off 5 minute and net don;t access any device.

 

[toqchista]

i and asus support talked and told me that this problem is asus firmware bug ,

i right now uses rt-n10e and gave me beta firmware and i upload beta firmware else now i'm wait what will happens

this is rt-n10e beta firmware who want ?
if is asus other model then send message asus support and will give beta firmware,

download >

https://www.dropbox.com/home?preview=RT-N10E_7.0.0.72.trx
or
http://www75.zippyshare.com/v/zycLqw88/file.html

 

[Svnn72]

I've been lurking for a while (read the entire topic) and have struggling with the same problems as most people above.
After about 9 months i tried just about everything, i was ready to give up on my cable ISP and switch back to DSL.
The last possible option was my own AC68u, everything else had been checked and replaced at least once.
Furthermore i have good experiences with asus and i am the last to suspect my own expensive hardware.

Long story short; in the end, about 5 days ago, I switched from Merlin to the Tomato firmware, and i am safe to say my problems are over
No more random disconnects, just a steady 200/20 mbit connection on the same DHCP lease since then.

I'm not here to bash Merlin in any way, and switching to Tomato definitely wasn't my first choice, but for those out there
with problems and a compatible router, you might wanna give this a go before losing your hair completely.
Tomato is not as user friendly and easy to set up as Merlin, but the documentation is pretty good.

 

[mstombs]

I still have problems with WAN dhcp on either Tomato or Merlin so have my own ping monitor script, It may be significant that Asus/Merlin always put a message in log about ISP dhcp not behaving correctly, but windows or Linux PCs bypassing the router do not have a problem, I am sure it is the router udhcpc that is the problem. The symptom I have is that the auto simple renew at half lease time does not work, whereas a full renew after release does (but with interruption to internet). I did experiment with an iptables command to translate unicast to broadcast, that used to work with Tomato, not sure about Merlin...

 

[shdwkeeper]

Instead of posting my experiences twice, I'm having these issues with an RT-AC68P that I just bought two days ago at Best Buy. See thread here: http://www.snbforums.com/threads/does-ac68u-and-ac68p-hang-for-anyone-else.25701/#post-193927

 

[shdwkeeper]

This router obviously has a major flaw in that it keeps losing the WAN IP address which results in the internet going away until you re-apply the Wan settings or basically restart the router. This has locked up on me about 10 times in the last few days and this is BRAND NEW. I have re-flashed with 3 different firmwares as well. My Netgear never did this.

 

[sloach]

toquista, it's interesting that ASUS admitted it's a firmware bug. Did they give you any other information or some identifier for the bug?

 

[Goofie Newfie]

I'm have the same issue with a 87u. Every few days I get disconnected,ISP DHCP error. I need to reboot. Merlins latest firmware. Tried normal and aggressive modes. I thought normal fixed it but sure enough it did it again tonight.

Sep 13 19:44:15 WAN Connection: ISP's DHCP did not function properly.
Sep 13 19:44:15 stop_nat_rules: apply the redirect_rules!

bell fiberop