NETGEAR is aware of a growing number of phone and online scams. To learn how to stay safe click here.
Forum Discussion
Solved
WNR2000v4 keeps changing internal ip address
Hi there, Since a while I have a strange problem with my WNR2000v4. Every now and again the router changes the internal ip address of itself. The standard/default ip address is 192.168.1.1 but fo...
Hi Andy,
I have some more information on this subject from my ISP community. It is a rough translation with google translate, I hope you understand it.
This seemingly "strange phenomenon" can be explained. What I can say at this stage is that both parties have no guilt here. This phenomenon stems from two requirements that a manufacturer must meet when NAT is used with the product. See https://tools.ietf.org/html/rfc4787#section-4.4 for details.
Why you can see this at a modem reset is because the RF interface of the modem is not connected yet to the CMTS. As a result, the DHCP server of the modem itself is active until there is a connection to the CMTS. The WAN interface of your router detects a link down / up event and then try again to obtain an IP address. The WAN interface from your router will then receive a private IP from the modem of the same sequence which is then used your router at that time. The modem does not know that your router uses internally the same private IP range. This clearly presents a problem and configure your router responds correctly by itself with another private IP range. Why is the DHCP server of the modem active when there is no connection to the CMTS? Because you still can logon to the management interface of the modem.
Manufacturers must implement one of the following two things to make it work.
All traffic between internal / exterene clients correct translation and forwarding. This is difficult because you need a solution such as Session Traversal Utilities for NAT server. If the two clients are the same ISP who does double NAT, you are also dependent on the correct NAT hairpinning implementation of the ISP router.
The second solution is automatically ensure that the internal and external IP ranges do not overlap / are identical.
Manufacturers logically opt for the second easy solution. How can you solve this problem permanently? If your local network eg. 192.168.0.0 Class C network can be used eg. Set your router to use IP network 192.168.1.0.
Well thanks very much for that.
If you do find out any more information about the modem I would be interested, but I think your idea of changing the router LAN subnet to avoid the issue un future is sound.
Regards,
Andy
Hi Andy,
I have some more information on this subject from my ISP community. It is a rough translation with google translate, I hope you understand it.
This seemingly "strange phenomenon" can be explained. What I can say at this stage is that both parties have no guilt here. This phenomenon stems from two requirements that a manufacturer must meet when NAT is used with the product. See https://tools.ietf.org/html/rfc4787#section-4.4 for details.
Why you can see this at a modem reset is because the RF interface of the modem is not connected yet to the CMTS. As a result, the DHCP server of the modem itself is active until there is a connection to the CMTS. The WAN interface of your router detects a link down / up event and then try again to obtain an IP address. The WAN interface from your router will then receive a private IP from the modem of the same sequence which is then used your router at that time. The modem does not know that your router uses internally the same private IP range. This clearly presents a problem and configure your router responds correctly by itself with another private IP range. Why is the DHCP server of the modem active when there is no connection to the CMTS? Because you still can logon to the management interface of the modem.
Manufacturers must implement one of the following two things to make it work.
All traffic between internal / exterene clients correct translation and forwarding. This is difficult because you need a solution such as Session Traversal Utilities for NAT server. If the two clients are the same ISP who does double NAT, you are also dependent on the correct NAT hairpinning implementation of the ISP router.
The second solution is automatically ensure that the internal and external IP ranges do not overlap / are identical.
Manufacturers logically opt for the second easy solution. How can you solve this problem permanently? If your local network eg. 192.168.0.0 Class C network can be used eg. Set your router to use IP network 192.168.1.0.
- Yes, that’s pretty much how I understand it to work. For me it’s very rare for the ISP connection to fail, but on those rare occasions in the past before I had an intervening router I would find that the attached PC would have a 192.168.100.x IP address which allowed me to access the modem and see what the issue was. Since getting a router I have always operated my LAN with a subnet of 192.168.12.x and this has never caused any issues even with several changes of modem model.
Babylon5 wrote:
Yes, that’s pretty much how I understand it to work. For me it’s very rare for the ISP connection to fail, but on those rare occasions in the past before I had an intervening router I would find that the attached PC would have a 192.168.100.x IP address which allowed me to access the modem and see what the issue was. Since getting a router I have always operated my LAN with a subnet of 192.168.12.x and this has never caused any issues even with several changes of modem model.I didn't know this. For me it is a first. Although I have to say that I had never problems with a modem before. A router is another story though.
Anyway, thank you! The last thing to do for me is to mark an answer to be the solution. But which one? I assume the best one is the one which help others too. Any ideas on that?
- I would suggest your post 41.