Loss of connectivity

Hi there people.

I'm managing a couple of 1700 routers. Im using HP Openview as a monitoring tool. I just cant figure out why i cant connect remotely. When it does come back up, it shows the router has been always up, but there are interface resets. no errors, crc or anything. However, at the time of loss of connectivity, it seems that the router is heavily over-utilizing? Could that be the problem? Any suggestions?

Assuming all configs and links are 100%

thanx in advance

Find more posts tagged with

Free for TechExams community: Cybersecurity salary guide

Compare cert salaries and plan your next career move

Button

Comments

EdTheLad

cisco_trouble wrote:

Hi there people.

I'm managing a couple of 1700 routers. Im using HP Openview as a monitoring tool. I just cant figure out why i cant connect remotely. When it does come back up, it shows the router has been always up, but there are interface resets. no errors, crc or anything. However, at the time of loss of connectivity, it seems that the router is heavily over-utilizing? Could that be the problem? Any suggestions?

Assuming all configs and links are 100%

thanx in advance

I havent really used Openview with cisco, the only things i can think of is
make sure your IOS and MIB releases are compatable.
If yes i would still try a different MIB, are you using a generic MIB or the actuall 1700 MIB?

cisco_trouble

The MIBs are fine. Could it be the cpu? When the router CPU starts running high, packets destined for the router itself (such as management traffic from HPOV), will queue up waiting for the CPU to free up. Could over-utilization do this? and if so, how will one go about solving this problem, other than replacing it with 10 000 series router :P

EdTheLad

cisco_trouble wrote:

The MIBs are fine. Could it be the cpu? When the router CPU starts running high, packets destined for the router itself (such as management traffic from HPOV), will queue up waiting for the CPU to free up. Could over-utilization do this? and if so, how will one go about solving this problem, other than replacing it with 10 000 series router :P

For sure the CPU can be creating this problem.But as i dont have experience
with this i cant advice what to do,sorry.Someone else ???

keenon

most monitoring programs use ping for to verify that device is active but use a bulk_snmp_get to pull stats. which i have seen cause unstablity on some devices. this sounds like the case for your 1700. my first thought is try to pull less information via snmp.

wildfire

Theres a bolt on for HPOV that shows the CPU usage, question though are u pulling the full mib table for your updates ? or just the ones you need?

How much work is the router actually doing the 1700 series was really designed as a remote site and with modern applications etc its more of a soho device than anything esle. Is involved in routing? if so and it only has one connection why not replace the routing protocol with static or default routes?

Is it running cpu intensive stuff such as NAT, is it logging to the buffers? all factors to consider and tweak.

darkuser

http://www.cisco.com/warp/public/63/highcpu.html