Reduce the complexity of the number of subnets routed, ACL's, etc. Any rules that have to be processed or encryption will increase utilization. If this is the default gateway for a bunch of LAN devices, then consider putting another routing device in front of it to take some of the load and only push through the traffic that needs to go across the WAN links on the edge router, or redesign the LAN addressing scheme so traffic on the local LAN does not have to go to the router to be routed.
In the IT trenches? So am I - read my <a href="http://itknowledgeexchange.techtarget.com/it-trenches">IT-Trenches blog</a>
Last Wiki Answer Submitted: June 9, 2009 8:41 pm by Labnuke9932,645 pts.
All Answer Wiki Contributors: Labnuke9932,645 pts.
If you live outside the United States, by submitting your email address you consent to having your personal data transferred to and processed in the United States.
There are a number of things that can cause this. The common issues are it is trying to do a lot of packet or route, manipulation, through complex route maps, there is a lot of buffering, where a lot of traffic from a fast interface, is trying to pass over a much slower one (LAN to slow WAN), or you have a routing protocol that is having to process a lot of updates.
Have a look at the ‘show process’ command output, which should help identify what is using a lot of resources. The ‘show memory, and ‘show buffers’ can also give more information. If you have a Cisco login, do a ‘show tech’ and paste it into the output inturpreter and see what that thinks is the problem. Check also the routing protocol, not just on this router, but on the others in the network, I have seen this where there is a very large OSPF area, and the tables were huge, so one link going down caused a lot of processing (the area was way over size, but the customer refused to deal with the cause, and faulted the routers every time they fell over).
Try to avoid using debug too much, as this will only make the problem worse, as most debugs are processor hungry. If you are, only output to the console port, do not try it over the network, and only try one debug at a time.
There is always our old friend the software bug. You could do worse than check the version against the bug tracker and see if this is a known issue. Or you could just try a different IOS version.
There are a number of things that can cause this. The common issues are it is trying to do a lot of packet or route, manipulation, through complex route maps, there is a lot of buffering, where a lot of traffic from a fast interface, is trying to pass over a much slower one (LAN to slow WAN), or you have a routing protocol that is having to process a lot of updates.
Have a look at the ‘show process’ command output, which should help identify what is using a lot of resources. The ‘show memory, and ‘show buffers’ can also give more information. If you have a Cisco login, do a ‘show tech’ and paste it into the output inturpreter and see what that thinks is the problem. Check also the routing protocol, not just on this router, but on the others in the network, I have seen this where there is a very large OSPF area, and the tables were huge, so one link going down caused a lot of processing (the area was way over size, but the customer refused to deal with the cause, and faulted the routers every time they fell over).
Try to avoid using debug too much, as this will only make the problem worse, as most debugs are processor hungry. If you are, only output to the console port, do not try it over the network, and only try one debug at a time.
There is always our old friend the software bug. You could do worse than check the version against the bug tracker and see if this is a known issue. Or you could just try a different IOS version.
Good Luck.