input queue drop

[Danilo Dy]

Hi,

What could cause input queue drop in a FastEthernet interface connected to MetroE beside those mentioned in this link http://www.cisco.com/en/US/products/hw/routers/ps133/products_tech_note09186a0080094791.shtml

I'm experiencing slow data transfer in this link (off-peak and peak) and ICMP between two Routers directly connected to MUX has packet drops.

Routers and MUXs are set to 100Mbps/Full-duplex. MUX to MUX Loopback Test (BIT Error) return nothing.

Thanks,

Dandy

[JORGE RODRIGUEZ]

Dandy, did this just developed ? I would assume this is WAN ethernet handoff right, I say if you have ruled out the link you provided I would ask provider again to do non-intrusive monitoring on the link, had similar scenario and provider returned to me with problems in their equipment, just to trully rule it out as for at least a 48 hours non-intrusive test.

Rgds

Jorge

[Danilo Dy]

Hi Jorge,

Thank you for your reply. Yes, it's WAN Ethernet handoff. MetroE Router|UTP|MUX|Fibre|Exchange|Fibre|MUX|UTP|Router. I suspect the problem has been there quite sometime before I discovers it as the input queue drops is already very large.

A month ago, this provider did a loopback test (MUX-to-MUX) found nothing. When we plug a laptop to MUX (my end) and provider is normalize (all hardcoded 100Mbps/Full-duplex), perform data transfer and also PING test, we still see PING packet drops and the data transfer speed is degrading. Common sense dictate that the problem is somewhere in provider side (cable from MUX, Router Interface, Router itself, or Human). To my surprise, this provider send 4 people to my site to replace the MUX without even testing it. We did the same laptop test and still have PING packet drops and data transfer is degrading.

After replacing the MUX, the problem is still there. Last night, they send people again to perform loopback test (MUX-to-MUX) this time the test is longer but still found nothing - which I already expected. Again, when we perform the laptop test, we still have packet drops when PING and data transfer is degrading.

Now, I'm stuck with this problem as this provider is troubleshooting the problem in the wrong end.

I also ask my router vendor to log a case with Cisco TAC. I already send them my router "show tech-support", my provider doesn't want to give their full "show tech-support" output. I doubt that the problem is in my side, because I run through the Input Queue and Output Queue Troubleshooting and also analyze my "show tech-support" to Output Interpreter", all found nothing.

Regards,

Dandy

[JORGE RODRIGUEZ]

Yes, typical .. from what you have discribed this sounds it needs to escalate, in my scenario a while back the 100 Megs termnated at 6509 switches and at first month would see errors on the port at one end after several tests came clean and provider ended up replacing one of their MUX box they managed.

I would suggest to post this issue also in METRO forum someone may provide some additional input..but I think this seems to gear more towards the provider side.

Rgds

Jorge

[Danilo Dy]

Hi Jorge,

Thanks again to your reply. I'm sure will post it to METRO forum.

I'm so vexed with this problem as I supposed to upgrade this link as we have an upcoming big event that we need to double the size of this link. Anyway, I have a drastic workaround not to use this link.

Regards,

Dandy