[c-nsp] ME3600 switch interface showing Packet drops on Trunk Port.

[Ivan]

Hi Muthu,

As per that thread the default buffer size for the Gi interfaces is very 
small and can be increased with a QoS policy as Warris described.

I was seeing "Total output drops" increasing in the output of "show int 
gi0/x"  AT a quick glance I can't see this information in the output you 
provided below, so you may have a different issue.

Ivan

On 21/Sep/2012 8:32 p.m., Muthukumar Rajagopalan wrote:
> Thanks Ivan for sharing the Thread, I just glanced the thread quickly, in
> our setup, there is no QOS applied as of now.
>
> The interface is a 1 Gig interface only and here is the show controller
> output.
>
> show controllers ethernet-controller gigabitEthernet 0/1
>
>       Transmit GigabitEthernet0/1 Receive
>     3746715079 Bytes 684001607 Bytes
>     1601240611 Unicast frames 1423901907 Unicast frames
>        3956660 Multicast frames 57515 Multicast frames
>          33950 Broadcast frames 111825 Broadcast frames
>     1603960779 Dot1Q frames 1424071247 Dot1Q frames
>              0 Too old frames 668768496 Unicast bytes
>     *3690121444 Deferred frames 5900758 Multicast bytes*
>              0 MTU exceeded frames 9332353 Broadcast bytes
>              0 FCS errors 0 FCS errors
>              0 1 collision frames 0 Alignment errors
>              0 2 collision frames
>              0 3 collision frames 0 Oversize frames
>              0 4 collision frames 0 Undersize frames
>              0 5 collision frames 0 Collision fragments
>              0 6 collision frames
>             * 0 7 collision frames 2134412 Minimum size frames
>              0 8 collision frames 33319889 65 to 127 byte frames
>              0 9 collision frames 13344585 128 to 255 byte frames
>              0 10 collision frames 4930119 256 to 511 byte frames
>              0 11 collision frames 2813946 512 to 1023 byte frames
>              0 12 collision frames 803808161 1024 to 1518 byte frames*
>              0 13 collision frames 0 Overrun frames
>              0 14 collision frames
>              0 15 collision frames
>              0 Pause frames 0 Pause frames
>              0 Excessive collisions 0 Symbol error frames
>              0 Late collisions 0 Invalid frames, too large
>              0 VLAN discard frames 563720136 Valid frames, too large
>              0 Excess defer frames 0 Invalid frames, too small
>        2310138 64 byte frames 0 Valid frames, too small
>       39205786 127 byte frames
>       14671469 255 byte frames 0 Too old frames
>        6687328 511 byte frames 0 Valid oversize frames
>       69655856 1023 byte frames 0 System FCS error frames
>      951464626 1518 byte frames
>      521236018 Too large frames
>              0 Good (1 coll) frames
>              0 Good (>1 coll) frames
>
> Thanks,
> Muthu.
> On Fri, Sep 21, 2012 at 12:07 PM, Muthukumar Rajagopalan
> <rmkindu at gmail.com>wrote:
>
>> Hi..
>>
>> We have implemented ME3600 switch recently and observing random packet
>> drops on the trunk link. On the access side, we have multi-services
>> running, Normal Server Data traffic, Video + Voice. Below is the config on
>> the trunk port and Q-in-Q enabled on the access. Different Pattern of
>> traffic has been sent and observing outbound drops max up to 3%.
>>
>> *Present Version:* me360x-universalk9-mz.122-52.EY4.bin
>>
>> port-type nni
>>   switchport mode trunk
>>   mtu 1512
>>   load-interval 30
>>   speed nonegotiate
>>   spanning-tree bpdufilter enable
>>   spanning-tree bpduguard enable
>>   spanning-tree guard root
>>   ethernet cfm mip level 7 vlan 1-4094
>>   ethernet oam remote-loopback supported
>>   ethernet oam
>>   hold-queue 4096 in
>>   hold-queue 4096 out
>>
>> Please provide your view if you folks came across this kind of issue
>> before.
>>
>> Thanks,
>> Muthu..
>>
>>
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
>