[c-nsp] NPE-G1 Interface errors / packetloss
Michael K. Smith - Adhost
mksmith at adhost.com
Thu Sep 29 12:55:13 EDT 2011
Can you do a "sho int x/x count error"? Also, have you tried setting the MTU to 1500 on both sides of the link? What is the MTU on the VLAN itself and for the chassis?
Mike
--
Michael K. Smith - CISSP, GSEC, GISP
Chief Technical Officer - Adhost Internet LLC mksmith at adhost.com
w: +1 (206) 404-9500 f: +1 (206) 404-9050
PGP: B49A DDF5 8611 27F3 08B9 84BB E61E 38C0 (Key ID: 0x9A96777D)
> -----Original Message-----
> From: Alexander Fossa [mailto:alex.fossa at xifos.net]
> Sent: Thursday, September 29, 2011 9:20 AM
> To: Michael K. Smith - Adhost; cisco-nsp [cisco-nsp at puck.nether.net]
> Subject: RE: NPE-G1 Interface errors / packetloss
>
> Hi,
>
> It seems you cannot disable flow-control on a NPE-G1.... I'm running 12.3.6f
>
> "My understanding is software based flow-control has been disabled on the
> 7200 because it was not scalable. But there is a cosmetic bug on the sh
> interface which shown input flow control as XON instead of unsupported.
>
> It has been fixed via CSCsr53390"
>
> There are two interconnects between the two chassis...
>
> GI0/1 (npe) > Gi1/7 (sup32) - copper native on npe, to brand new cisco 1000-
> T SFP
> GI0/3 (npe) > Gi1/4 (sup32) - fibre GBIC on npe, to brand new cisco SFP SX
>
> The wierd thing is the CRC / Frame errors followed the VLAN when it was
> moved from the fibre to the copper. Both are direct patch cables nothing
> else... distance is about 3m on each.
>
> Thanks,
>
> Alex
>
>
> ________________________________________
> From: Michael K. Smith - Adhost [mksmith at adhost.com]
> Sent: 29 September 2011 17:12
> To: Alexander Fossa; cisco-nsp [cisco-nsp at puck.nether.net]
> Subject: RE: NPE-G1 Interface errors / packetloss
>
> > -----Original Message-----
> > From: cisco-nsp-bounces at puck.nether.net [mailto:cisco-nsp-
> > bounces at puck.nether.net] On Behalf Of Alexander Fossa
> > Sent: Thursday, September 29, 2011 8:10 AM
> > To: cisco-nsp [cisco-nsp at puck.nether.net]
> > Subject: [c-nsp] NPE-G1 Interface errors / packetloss
> >
> > Hi,
> >
> > We've been trying to diagnose a packetloss issue though one of our LNS's.
> >
> > The NPE-G1 is connected directly to a SUP32 with traffic in the region of
> > 50mbit.
> >
> > At first we thought it was a dirty fibre / failed SFP, both have been replaced
> > but the problem continues.
> >
> > We then moved the main infrastructure VLAN off Gi0/3 to Gi0/1, hence
> > ruling out the SUP32 port and the physical port on the NPE-G1. The errors
> > followed....
> >
> > I've been doing quite a bit of reading and think this might be something to
> do
> > with microbursts.... but we've not had an increase in traffic, if anything a
> > decrease.
> >
> > Show interface stats from NPE-G1
> >
> > GigabitEthernet0/1 is up, line protocol is up
> > Hardware is BCM1250 Internal MAC, address is 0008.203e.441b (bia
> > 0008.203e.441
> > b)
> > Description: Link to BEDGE1.THE GI1/7
> > MTU 9216 bytes, BW 1000000 Kbit, DLY 10 usec,
> > reliability 255/255, txload 5/255, rxload 4/255
> > Encapsulation 802.1Q Virtual LAN, Vlan ID 1., loopback not set
> > Keepalive set (10 sec)
> > Full-duplex, 1000Mb/s, media type is RJ45
> > output flow-control is XON, input flow-control is XON
> > ARP type: ARPA, ARP Timeout 04:00:00
> > Last input 00:00:00, output 00:00:00, output hang never
> > Last clearing of "show interface" counters 16:39:10
> > Input queue: 0/1500/97/4 (size/max/drops/flushes); Total output drops: 0
> > Queueing strategy: fifo
> > Output queue: 0/40 (size/max)
> > 5 minute input rate 18801000 bits/sec, 4990 packets/sec
> > 5 minute output rate 21289000 bits/sec, 5288 packets/sec
> > 152579806 packets input, 2828997929 bytes, 0 no buffer
> > Received 0 broadcasts, 0 runts, 0 giants, 0 throttles
> > 0 input errors, 0 CRC, 0 frame, 142 overrun, 0 ignored
> > 0 watchdog, 265811 multicast, 0 pause input
> > 0 input packets with dribble condition detected
> > 161805464 packets output, 4109563833 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
> > DSL1.THE#
> >
> > Show interface stats from SUP32
> >
> > edge1.the#sh int gi1/7
> > GigabitEthernet1/7 is up, line protocol is up (connected)
> > Hardware is C6k 1000Mb 802.3, address is 0016.465e.3732 (bia
> > 0016.465e.3732)
> > Description: DSL1.THE_GI0/1
> > MTU 9216 bytes, BW 1000000 Kbit, DLY 10 usec,
> > reliability 255/255, txload 4/255, rxload 5/255
> > 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 00:00:21, output never, output hang never
> > Last clearing of "show interface" counters 16:25:18
> > Input queue: 0/2000/0/0 (size/max/drops/flushes); Total output drops: 0
> > Queueing strategy: fifo
> > Output queue: 0/40 (size/max)
> > 5 minute input rate 21224000 bits/sec, 5233 packets/sec
> > 5 minute output rate 19475000 bits/sec, 4943 packets/sec
> > 160219198 packets input, 76742158873 bytes, 0 no buffer
> > Received 133662 broadcasts (133662 multicasts)
> > 47 runts, 0 giants, 0 throttles
> > 25065 input errors, 24986 CRC, 24939 frame, 0 overrun, 0 ignored
> > 0 watchdog, 0 multicast, 0 pause input
> > 0 input packets with dribble condition detected
> > 151327463 packets output, 65370785219 bytes, 0 underruns
> > 0 output errors, 0 collisions, 1 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
> > edge1.the#
> >
> > As you can see lots of CRC & Frame errors the SUP32 side, and a few
> > overruns on the NPE-G1 side.
> >
> > Any ideas / thoughts?
> >
> > Thanks in advance,
> >
> > Alex
> >
>
> Hi Alex:
>
> Couple of things.
>
> 1) Flow Control - it looks like your NPE-G1 is set to XON/XON and your SUP32
> has flow control off. I would disable flow control on the NPE-G1 to, if nothing
> else, make the sides match for testing.
> 2) Copper/Fiber? - you said you replaced fiber/SFP but the interface on both
> sides say this is a copper connection. With that, I would check for distance
> issues, bad cable, etc.
> 3) If anything, it looks like the issue is coming from the NPE-G1 side, given the
> huge amount of errors on the SUP32.
>
> Regards,
>
> Mike
>
> --
> This message has been scanned for viruses and dangerous content by the
> Xifos Limited e-mail system.
> Further information can be found at www.xifos.net
>
>
> --
> This message has been scanned for viruses and dangerous content by the
> Xifos Limited e-mail system.
> Further information can be found at www.xifos.net
More information about the cisco-nsp
mailing list