Internet service keeps dropping

Here is a strange one. We have a dedicated Fiber service with our ISP running 10mb/Full. Their demark router is a Cisco 1800 with a single fiber transceiver. Our side is using a RAS server to connect to the 1800. The Server is Win 2003 configured for RAS, DHCP and NAT. DNS is handled by the ISP's server via DNS forwarding. The ISP has replaced the Fiber module and the 1800. It was replaced with another 1800 and it is running the same config.

The problem is that the internet access will drop sometimes 3x a day. There are no events created on either the server or the router and the consistent fix has been to reset the interface on the router by either cycling the router, disconnecting and reconnecting the ethernet cable or by cycling the server. Once the interface is reset service is restored. When the service goes down I am still able to ping the public side of the router but no farther.

Any suggestions would be appreciated. Also we have tried just running a laptop directly to the router and had the same result with it going down.

Who is Participating?
rfc1180Connect With a Mentor Commented:
>When the service goes down I am still able to ping the public side of the router but no farther.
What does the arp table look like when the service is down? I am assuming you do not have access to the 1800.

Have you executed a traceroute to see where the last hop that packet traverses; by IP and not by DNS?

Sound like you really need to involve the ISP during the outage if you do not have access to the 1800 as you will very limited in the troubleshooting that you can accomplish.

however, the fact that when you recycle the router, the connection restores indicates the issue is more than likely related to the 1800.

1. Memory exhaustion:
If you point a static route to a broadcast interface, the route is inserted into the routing table only when the broadcast interface is up. This configuration is not recommended because when the next hop of a static route points to an interface, the router considers each of the hosts within the range of the route to be directly connected through that interface. For example, ip route Ethernet0.


2. pps performance:
the 1800 can handle about 70K of packets per second (35.84Mbps) at 64 byte packets

what does the CPU look like: 'show proc cpu hist'

3. Network Utilization:
Another expert posted a comment on saturation; great point, this would more than likely result it loss of packets when the network is 100 percent utilized.

4. Inconsistencies between the RIB and FIB if you have IP CEF enabled.

5. Routing protocol issues

6. IOS bugs

ultimately, involve the ISP, they should be able to assist; however, as they have already replace the router, the issue persists. Maybe you can escalate the issue to a high tier of engineers.

OriNetworksConnect With a Mentor Commented:
Is the network link being saturated? I have seen this when my organization was reaching the 5mb  bandwidth limit of our internet connection. This would happen at random times of the day sometimes for a few minutes, sometimes up to an hour. Maybe your ISP can help you report on your bandwith utilization during times of outage.
I would look at changing the IOS image version on the router. Either an upgrade or downgrade. Sometimes different versions have different issues. We had one 1800 that caused similar issues and downgrading the IOS resolved the issue.
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.