[c-nsp] Help with output drops

Tony td_miles at yahoo.com
Mon Jul 13 02:21:47 EDT 2009


Hi Randy,

Is QoS enabled ? What does "show mls qos" tell you ?

Do you need QOS at all ? If not, disable it globally (no mls qos) and your problem might just go away if it's being caused by queue threshold defaults..

If it's production switch, do it during a scheduled maintenance period as it might disrupt traffic for a second.



regards,
Tony.


--- On Mon, 13/7/09, Randy McAnally <rsm at fast-serv.com> wrote:

> From: Randy McAnally <rsm at fast-serv.com>
> Subject: [c-nsp] Help with output drops
> To: cisco-nsp at puck.nether.net
> Date: Monday, 13 July, 2009, 1:51 PM
> Hi all,
> 
> I just finished installing and configuring a new 6509 with
> dual sup7203bxl
> (12.2(18)SXF15a) and a 6724 linecards.  It serves a
> simple purpose of
> maintaining a single BGP session, and managing layer3
> (vlans) for various
> access switches.  No end devices are connected.
> 
> The problem is that we are getting constant output drops
> when our gig-E uplink
> goes above ~400 mbps.  Nowhere near the interface
> speed!  See below, take note
> of massive 'Total output drops' with no other errors (on
> either end):
> 
> rtr1.ash#sh int g1/1
> GigabitEthernet1/1 is up, line protocol is up (connected)
>   Hardware is C6k 1000Mb 802.3, address is
> 00d0.01ff.5800 (bia 00d0.01ff.5800)
>   Description: PTP-UPLINK
>   Internet address is 209.9.224.68/29
>   MTU 1500 bytes, BW 1000000 Kbit, DLY 10 usec,
>      reliability 255/255, txload
> 118/255, rxload 12/255
>   Encapsulation ARPA, loopback not set
>   Keepalive set (10 sec)
>   Full-duplex, 1000Mb/s, media type is T
>   input flow-control is off, output flow-control is
> off
>   Clock mode is auto
>   ARP type: ARPA, ARP Timeout 04:00:00
>   Last input 00:00:00, output 00:00:01, output hang
> never
>   Last clearing of "show interface" counters 05:01:25
>   Input queue: 0/1000/0/0 (size/max/drops/flushes);
> Total output drops: 718023
>   Queueing strategy: fifo
>   Output queue: 0/100 (size/max)
>   30 second input rate 47789000 bits/sec, 30797
> packets/sec
>   30 second output rate 465362000 bits/sec, 48729
> packets/sec
>   L2 Switched: ucast: 27775 pkt, 2136621 bytes -
> mcast: 24590 pkt, 1574763 bytes
>   L3 in Switched: ucast: 592150327 pkt, 95608889548
> bytes - mcast: 0 pkt, 0
> bytes mcast
>   L3 out Switched: ucast: 991372425 pkt, 1214882993007
> bytes mcast: 0 pkt, 0 bytes
>      592554441 packets input,
> 95674494492 bytes, 0 no buffer
>      Received 33643 broadcasts (17872
> IP multicasts)
>      0 runts, 0 giants, 0 throttles
>      0 input errors, 0 CRC, 0 frame, 0
> overrun, 0 ignored
>      0 watchdog, 0 multicast, 0 pause
> input
>      0 input packets with dribble
> condition detected
>      991006394 packets output,
> 1214377864373 bytes, 0 underruns
>      0 output errors, 0 collisions, 0
> interface resets
>      0 babbles, 0 late collision, 0
> deferred
>      0 lost carrier, 0 no carrier, 0
> PAUSE output
>      0 output buffer failures, 0 output
> buffers swapped out
> 
> The CPU usage is nil:
> 
> rtr1.ash#sh proc cpu sort
> 
> CPU utilization for five seconds: 1%/0%; one minute: 0%;
> five minutes: 0%
>  PID Runtime(ms)   Invoked   
>  
> uSecs   5Sec   1Min   5Min
> TTY Process
>    6     3036624 
>   252272      12037  0.47% 
> 0.19%  0.18%   0 Check heaps
>  316      195004 
>    99543   
>    1958  0.15%  0.01% 
> 0.00%   0 BGP Scanner
>  119     
> 267568   2962884     
>    90  0.15%  0.03% 
> 0.02%   0 IP Input
>  172     
> 413528   2134933       
> 193  0.07%  0.03%  0.02%   0
> CEF process
>    4         
> 16     48214       
>   0  0.00%  0.00% 
> 0.00%   0 cpf_process_ipcQ
>    3       
>    0     
>    2         
> 0  0.00%  0.00%  0.00%   0
> cpf_process_msg_
>    5       
>    0     
>    1         
> 0  0.00%  0.00%  0.00%   0 PF
> Redun ICC Req
>    2     
>    772    298376   
>       2  0.00%  0.00% 
> 0.00%   0 Load Meter
>    9   
>    23964    157684   
>     151  0.00%  0.01% 
> 0.00%   0 ARP Input
>    7       
>    0     
>    1         
> 0  0.00%  0.00%  0.00%   0
> Pool Manager
>    8       
>    0     
>    2         
> 0  0.00%  0.00%  0.00%   0
> Timers
> <<<snip>>>
> 
> I THINK I have determined the drops are caused by buffer
> congestion on the port:
> 
> rtr1.ash#sh queueing interface gigabitEthernet 1/1 
> 
> rtr1.ash#sh queueing interface gigabitEthernet 1/1
> Interface GigabitEthernet1/1 queueing strategy: 
> Weighted Round-Robin
>   Port QoS is enabled
>   Port is untrusted
>   Extend trust state: not trusted [COS = 0]
>   Default COS is 0
>     Queueing Mode In Tx direction: mode-cos
>     Transmit queues [type = 1p3q8t]:
>     Queue Id    Scheduling  Num of
> thresholds
>     -----------------------------------------
>        01     
>    WRR         
>        08
>        02     
>    WRR         
>        08
>        03     
>    WRR         
>        08
>        04     
>    Priority         
>   01
> 
>     WRR bandwidth ratios:  100[queue 1]
> 150[queue 2] 200[queue 3]
>     queue-limit ratios: 
>    50[queue 1]  20[queue 2] 
> 15[queue 3]  15[Pri Queue]
> 
> <<<snip>>>
> 
>   Packets dropped on Transmit:
> 
>     queue     dropped 
> [cos-map]
>    
> ---------------------------------------------
>     1           
>        719527  [0 1 ]
>     2           
>             0  [2 3 4 ]
>     3           
>             0  [6 7 ]
>     4           
>             0  [5 ]
> 
> So it would appear all of my traffic goes into queue
> 1.  It would also seem
> that 50% buffers for queue 1 isn't enough?  These are
> the default settings by
> the way.
> 
> I'm pretty sure that wrr-queue queue-limit and wrr-queue
> bandwidth should help
> us mitigate this frustrating packet loss, but I've no
> experience and would
> like some insight and suggestions before I start making
> changes.  I am totally
> unfamiliar with these features (I come from Foundry/Brocade
> background) and
> would like any suggestions or advise you might have before
> I try anything that
> could risk downtime or further issues in a production
> environment.
> 
> And lastly, would changing the queue settings cause BGP to
> drop or anything
> else unexpected (like changing flow control would reset the
> interface, ect)?
> 
> Thank you!
> 
> --
> Randy
> www.FastServ.com
> 
> _______________________________________________
> 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/
> 


      



More information about the cisco-nsp mailing list