OpenWrt Forum Archive

Topic: Problem with WNR1000 v3

The content of this topic has been archived on 10 Apr 2018. There are no obvious gaps in this topic, but there may still be some posts missing at the end.

So I set up a dust-collecting Netgear WNR1000 v3 router with the trunk build firmware and it ran beautifully. I could ssh into the router and was attempting to get an internet connection on it. I then (blindly) followed some instructions and messed with the switch settings in /etc/config/network and after the reboot, I can no longer ssh into the router. Ping, telnet, and ssh all claim "No route to host." So, I searched the problems other people were having but it doesn't seem like many people run with this router and I have come up with nothing. Reading the Peacock thread was helpful in getting some more information regarding the problem, but has left me even more stumped.

Running a ping on the router's IP before bootup to monitor it will be unresponsive (naturally) until all LAN leds flash once (I assume when the firmware has loaded) at which point the router responds temporarily. The TTL is 64 (consistently; I also never see TTL 100 or 128) but after an average 4-6 responses it goes back to "Destination Host Unreachable." According to the Peacock thread, TTL=64 means that the firmware is what is responding to the ping. My best guess is the switch settings are being set a little bit after the firmware loads and when they do get set, the firmware doesn't know where to route packets. Every thread I've read, they either see TTL 64 while pinging (not just before total bootup) or other problems that I don't think are related.

I have tried the 30/30/30 reset, just holding the reset button for 30 seconds without restarting, attempts to see if I could tftp (I wasn't sure because I didn't understand if TTL 64 would allow that or not).

If any more information is needed I will respond ASAP.

After holding the reset button for 5 seconds after the router is fully booted the configuration should revert to default.

While it's booting during the time you can ping it it's checking for a tftp server to flash new firmware.  There are many free ones available out there you can download.  Set a static ip on your pc in the 192.168.1.x range other than 192.168.1.1 start the server telling it what to load then boot the router to flash it.

I tested the reset button again with variations of 5 to 30 seconds with no change. However, you mention tftp server and I admit I didn't even think about that, I was attempting to do it through a tftp client. I'm going to test that out and will respond with the details. Thanks for your response.

Sometimes it is the WPS button, if the router has one.  Plug in and wait for it to boot fully, then press and hold the button for more than 5 seconds, then release.  If it did cause a reset all the LEDs will blink and it will reboot to stock settings.

There is also failsafe mode.  If this is supported you will see one of the LEDs (usually power) blink rapidly for 2 seconds during bootup, then blink slowly, and finally stay on.  During the time it is blinking rapidly, press the reset or WPS button several times.  If this was successful the LED will start to blink even more rapidly and continue to do so.  This is recovery mode where the settings are temporarily cleared.  You can log in by SSH and issue the firstboot command to reset and reboot.

Alright, so I set up a tftp server and watched with tcpdump. Something I didn't see before was the router address (192.168.1.1) sent a broadcast (192.168.1.255) UDP packet. I seen that tftp requires good timing (at least on the client) so I am unsure if this is an error on my part or not.

Some note-worthy information (I hope):
a request gets sent asking who has 192.168.1.1 to tell my computer
this request is then met with an ARP reply saying 192.168.1.1 is at [router mac address] (I verified)
then it's an ICMP echo request from my computer and is met with an ICMP echo reply (during the time before ping stops responding)
also during the tcpdump, I see an ARP request to tell 192.168.1.1 who has 192.168.1.1

The router is definitely getting set to 192.168.1.1 but then it just loses it which by following the information thread, I think that's the firmware setting itself to the default IP address?

As for @mk24, holding the WPS button turns the power LED completely off (it's originally solid amber) so I'm not sure what that's doing but holding it has made no change. The power LED during bootup does not flash, it turns and stays solid amber.

(Last edited by disp032 on 8 Nov 2017, 20:01)

Yes that is why I advised you to start the tftp server first so it's ready for the very short time the router asks for firmware.  The address is always 192.168.1.1 during that short lived process.

It may take a few tries to get the timing of everything just right.

after a while of testing and probing for stuff, something I've noticed is output from a netstat -rn:

for my router connected to the internet:
Destination         Gateway         Genmask
0.0.0.0                  10.0.0.1            0.0.0.0
10.0.0.0                 0.0.0.0            255.255.255.0

but for my Netgear router:
Destination         Gateway         Genmask
192.168.1.0            0.0.0.0            255.255.255.0

is no gateway for that network a cause for concern? I think I may be skimping on getting the tftp connection but I can't get it for the life of me. I did manage to receive 1 ping reply from the router that had a TTL=100 if that matters but nothing had happened.

some more notes:
I have a static IP set to 192.168.1.2 (and as stated before, the router does respond like it seems it should during bootup)
when I set the static IP, I also define the gateway to be 192.168.1.1 but it doesn't show up in the netstat command
I tried a tftp server, verified the tftp server was running, and had it started before plugging the router back in like you said
I've tried with some different tftp clients, but still no luck

I feel like I'm missing something obvious.

[edit]
holding the WiFi on/off button while powering the router on keeps all LEDs off and the power LED in a very low brightness state. Nothing seems to be happening, not sure what that is.

(Last edited by disp032 on 9 Nov 2017, 03:53)

sorry, I haven't worded that the best. I'm running netstat from my computer to view the network that I'm hooked up to. I am unable to get into the router at all, every time I try I am met with "No route to host."

the line with destination 0.0.0.0 is called a default route, since your netgear router is not working propertly you don't get that when connected to it.

Still no luck, guide was a fail. Though I have found a consistency for getting TTL=100 responses. A 30/30/30 reset gets the ttl=100 then a short downtime which leads into ttl=64 and then it goes down again. The one thing I'm confused about is if the router is running a tftp client, how is it supposed to know which file to grab? I have the original netgear firmware in /srv/tftp with the original name.

Also, using route to set a default gateway actually changes the communication with the router (while it's able to respond). When the gateway isn't set, the router will respond to a who-has ARP request but won't send one for my computer. When I set the gateway, then the router sends a who-has for my computer. If you have knowledge of the switch settings with openwrt, is any of this related or am I chasing dead ends? Thanks for the time by the way, this is maddening lol.

If you followed the guide you should have specified the ip, password and filename when you set up tftp2 on your pc


as far as the default gateway goes it should only come into play if you try to access an address that is not in the same network as your pc is.  The default gateway simply says send all traffic here when I don't know where it is.

hmm. the only thing is I have no indication of when I'm going to be able to start the tftp connection as the power LED stays a constant solid amber on the 1000 v3. I will test by ping responses/tcpdump and get back with the results.

disp032 wrote:

hmm. the only thing is I have no indication of when I'm going to be able to start the tftp connection as the power LED stays a constant solid amber on the 1000 v3. I will test by ping responses/tcpdump and get back with the results.

You have to hold the reset button and do not release it  while you apply power it MUST turn flashing green or it will not work.

Unplug all port connections from router except the PC you are using to upload.
Turn router OFF for 10 seconds.
Hold down the reset button on the back of router with a paper clip.
Power ON the router while holding down the reset button.  THIS MAY TAKE 30 SECONDS!  BE PATIENT!!!!
Watch the Power LED. It starts with orange color, and then will start flashing orange.
Power LED will then turn to flashing green. Count at least 10 flashing green LED and release the reset button.
Click Upgrade button on the tftp2 utility to do firmware upgrade.

If that is not working it could be your reset button is bad.

(Last edited by WWTK on 9 Nov 2017, 06:53)

I never paid attention to the functions when the router was in use but right now, if I hold the reset button in before powering the router on (30/30/30, or anything as long as I hold it while powering on), once I let go of it after the power LED is on, the 4 LAN LEDs will go solid green, then go off and whatever port is plugged in will come back on.

[edit]
Bypassing the reset button powers LAN 1 and 2 LEDs and then all lights flash once and it's functioning "normally." When I let go of the reset button or remove the wire connecting the reset button pins, it seems like the power LED flashes to green very quickly but it only does this once. This very well could be a hardware problem as this router survived a thunderstorm that a PC and modem hooked up to it did not.

(Last edited by disp032 on 10 Nov 2017, 02:18)

hmmmm strange that it was working when you originally flashed it though.  Since you see a difference with the button pressed or not it seems the button works.  No matter how much you messed up (or not) lede the hold the reset button should still work as described on the netgear site since this is controlled by the boot rom and the firmware doesn't even come into play yet so that part is strange.

If you want to proceed further with the device consider getting a ttl serial cable to see what is going on via the console.  You can also control tftp mode via that interface.

Eh, LAN port 4 has never worked (I never noticed until I had the LAN ports maxed out and I'm pretty sure it was a while after that storm) and there's a slight component burn smell so at this point I might as well just invest in a new one and if I get a serial cable I can check this router out if it's still around to see what it says. Thanks for the help though man, and hopefully if somebody else with the same router finds this, they can use some of the information in here to figure it out (and hopefully it's not the same problem as this one lol).

The discussion might have continued from here.