No traffic on inside Router Interface

[jennyjohn]

Interface Fa 0/0 on my Cisco ISR 2801 router is connected to a Linksys Switch. Every few days the traffic on Fa 0/0 interface stops and all the devices in the LAN cannot communicate to the internet. i.e., the input and output rate is zero on interface Fa 0/0. Even the ARP table on the router goes empty except for the router interfaces. A router reload will rectify this issue. In this state the show interface Fa 0/0 is as follows.

FastEthernet0/0 is up, line protocol is up
  Hardware is Gt96k FE, address is 0025.451f.c8a0 (bia 0025.451f.c8a0)
  MTU 1500 bytes, BW 100000 Kbit/sec, DLY 100 usec,
     reliability 255/255, txload 1/255, rxload 1/255
  Encapsulation 802.1Q Virtual LAN, Vlan ID  1., loopback not set
  Keepalive not set
  Full-duplex, 100Mb/s, 100BaseTX/FX
  ARP type: ARPA, ARP Timeout 04:00:00
  Last input 14:08:58, output 00:00:21, output hang never
  Last clearing of "show interface" counters never
Input queue: 76/75/0/0 (size/max/drops/flushes); Total output drops: 0
  Queueing strategy: fifo
  Output queue: 0/40 (size/max)
  5 minute input rate 0 bits/sec, 0 packets/sec
  5 minute output rate 0 bits/sec, 0 packets/sec
    1818599 packets input, 326976612 bytes
     Received 75581 broadcasts, 0 runts, 0 giants, 76054 throttles
     1 input errors, 0 CRC, 0 frame, 0 overrun, 1 ignored
     0 watchdog
     0 input packets with dribble condition detected
     2050850 packets output, 1125237264 bytes, 0 underruns
     0 output errors, 0 collisions, 5 interface resets
     0 unknown protocol drops
     0 babbles, 0 late collision, 0 deferred
     0 lost carrier, 0 no carrier
     0 output buffer failures, 0 output buffers swapped out
FastEthernet0/0.10 is up, line protocol is up
  Hardware is Gt96k FE, address is 0025.451f.c8a0 (bia 0025.451f.c8a0)
  Internet address is 172.16.1.1/24
  MTU 1500 bytes, BW 100000 Kbit/sec, DLY 100 usec,
     reliability 255/255, txload 1/255, rxload 1/255
  Encapsulation 802.1Q Virtual LAN, Vlan ID  10.
  ARP type: ARPA, ARP Timeout 04:00:00
  Last clearing of "show interface" counters never
FastEthernet0/0.20 is up, line protocol is up
  Hardware is Gt96k FE, address is 0025.451f.c8a0 (bia 0025.451f.c8a0)
  Internet address is 192.168.1.1/24
  MTU 1500 bytes, BW 100000 Kbit/sec, DLY 100 usec,
     reliability 255/255, txload 1/255, rxload 1/255
  Encapsulation 802.1Q Virtual LAN, Vlan ID  20.
  ARP type: ARPA, ARP Timeout 04:00:00
  Last clearing of "show interface" counters never

How can I rectify this issue? Will increasing the hold-queue help? There is a duplex mismatch on the same interface, can this cause this state?

%CDP-4-DUPLEX_MISMATCH: duplex mismatch discovered on FastEthernet0/0.20 (not half duplex), with SEP0025841672B8 Port 1 (half duplex).

duplex mismatch is caused by a ATA device on the LAN, which is a half duplex device.

Thanks in Advance.

[Peter Paluch · Accepted solution]

Hi,

Regarding the CDP message about the duplex mismatch, it is not directly related to your problems. Your FastEthernet interface operates at the full duplex while the ATA, as you have commented, is obviously operating at half duplex, and it is indicating it in its own CDP packet. As the CDP packet is transparently relayed through the Linksys switch, your router receives it and it discovers that the ATA is not operating in the same duplex mode so it emits a warning. However, you should try to discover why the ATA operates in half duplex mode because in a fully switched network, there is no reason for a device to operate in half duplex.

However, in your output, you have yourself highlighted the following line:

Input queue: 76/75/0/0 (size/max/drops/flushes); Total output drops: 0

This state indicates that the interface is wedged - the input queue of the interface is full. This can happen because of two most common reasons: the first is that there is a possible loop in the network connected to your interface Fa0/0 and the circling frames are filling the input queue. The second reason is that the IOS contains a bug - while it allocates new buffers for incoming packets to temporarily store them into, it forgets to deallocate them afterwards. After a certain point, no new buffers can be allocated, and the interface becomes inoperational. Does your router also display messages similar to %SYS-2-MALLOCFAIL or similar indicating a problem with memory or buffer allocation?

In any case, if possible, try to upgrade your IOS to the recent stable version - that might resolve your issue.

Best regards,

Peter

[jennyjohn]

Hi Peter,

       Thanks for the reply.

As you said this looks like a Wedged interface buffer leak.

http://www.cisco.com/en/US/products/hw/iad/ps397/products_tech_note09186a00800a7b85.shtml#topic1

I will change the IOS right away.

Thanks a lot.

[jimmysands73_2]

Plus the line :

Encapsulation 802.1Q Virtual LAN, Vlan ID  1., loopback not set

Wouldnt that limit the traffic out to only Vlan 1?

[Peter Paluch]

Jimmy,

I am sorry but I do not understand your question. Can you perhaps rephrase it differently, or give an example?

Best regards,

Peter

[jimmysands73_2]

By that line, I believe you put that interface in VLAN 1, and thus limit the traffic in/out.  I could be wrong though....

[Peter Paluch]

Jimmy,

That line about VLAN1 is merely related to the fact that there are 802.1Q subinterfaces created under the FastEthernet0/0 interface and none of them is configured for the native VLAN so the main physical interface itself (or better said, its configuration) will process the untagged frames should they arrive. You will see such line in all configurations of 802.1Q subinterfaces where no subinterface has been designed to handle the native untagged VLAN. However, the problem of the original poster was positively identified as an IOS bug - I suggest reading the document that was mentioned a few posts earlier about buffer leaks. I just hope that upgrading the IOS will resolve the issue.

Best regards,

Peter

[jimmysands73_2]

Thanks for clearing up my confusion!   I hope he gets the issue resolved also.

Cheers,

Jimmy

[jennyjohn]

Hi Peter,

    The IOS upgrade to 12.4.24T2(ED) seems to have solved the wedged interface buffer leak on my router 2801.

But now I am planning to add a new HWIC-3G-HSPA card on to my router, but this card seems to be only supported with 12.4YB IOS family. How can I verify if the wedged interface IOS bug has been fixed on the latest 12.4YB IOS. (Note: the wedged interface buffer leak was noticed with IOS 12.4.22YB2 )

Also, if you know the Bug ID of the Wedged Interface Buffer Leak, please do tell me.

Thanks

Jenny

[juancarlosorellana]

I have a Card HWIC-3G-HSPA router 881 and now in my country in HSDPA coverage is (not in HSPA and HSUPA), my question is whether even so using the Card HWIC-3G-HSPA I connect to HSDPA network without any problem?