[nsp] another strange 2950 behavior

Valeriu VRACIU vvraciu at iasi.roedu.net
Sun Jul 18 13:17:40 EDT 2004


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello,

We have some strange phenomena happening on a 2950. One of its FastEthernet 
ports is connecting, via optical fiber and media converters, a distant 
location. All worked fine for more than 1 year, until a couple of days ago 
when, suddenly, the connection begun to work chaotically, very slow or even 
not at all. Reloading local or remote switch helps sometimes.

The strangeness is the respective port showing more MAC addresses than there 
are at the other end, created by replacing some of real addresses figures 
usually with 5. Here is an output with 19 addresses (yesterday I've seen a 
table with more than 200 false MAC addresses on the same interface !!), at 
the distant location we have just 4 hosts and one 2950 switch.

sw-lan#sh mac-address-table interface fastEthernet 0/15
          Mac Address Table
- -------------------------------------------
Vlan    Mac Address       Type        Ports
- ----    -----------       --------    -----
  98    0002.3fd1.1408    DYNAMIC     Fa0/15
  98    0002.3fd1.1450    DYNAMIC     Fa0/15
  98    0002.3fd1.14f0    DYNAMIC     Fa0/15   < host
  98    0002.3fd1.45f6    DYNAMIC     Fa0/15
  98    0002.3fd1.55f0    DYNAMIC     Fa0/15
  98    0002.3ff1.14f8    DYNAMIC     Fa0/15
  98    0002.5fe3.5e56    DYNAMIC     Fa0/15
  98    0008.219f.62c0    DYNAMIC     Fa0/15   < remote switch
  98    0008.219f.62c1    DYNAMIC     Fa0/15   < probably also remote switch
  98    0025.3fd1.c5f0    DYNAMIC     Fa0/15
  98    0050.048a.ea71    DYNAMIC     Fa0/15   < host
  98    0050.048a.ea74    DYNAMIC     Fa0/15
  98    0069.279f.62c1    DYNAMIC     Fa0/15
  98    0076.ce60.4000    DYNAMIC     Fa0/15
  98    0080.adb7.80fd    DYNAMIC     Fa0/15   < host
  98    00b0.6409.ebc1    DYNAMIC     Fa0/15   < host
  98    505f.555d.5b04    DYNAMIC     Fa0/15
  98    525f.7555.5a55    DYNAMIC     Fa0/15
  98    8063.3d54.5d75    DYNAMIC     Fa0/15
Total Mac Addresses for this criterion: 19

Also, there are many CRC errors on the local and remote switches' interfaces, 
errors that I haven's seen before:

local switch:
sw-local#sh int fast 0/15
FastEthernet0/15 is up, line protocol is up (connected)
  Hardware is Fast Ethernet, address is 000a.b7b0.f50f (bia 000a.b7b0.f50f)
  MTU 1500 bytes, BW 100000 Kbit, DLY 1000 usec,
     reliability 236/255, txload 1/255, rxload 1/255
  Encapsulation ARPA, loopback not set
  Keepalive set (10 sec)
  Full-duplex, 100Mb/s
  input flow-control is off, output flow-control is off
  ARP type: ARPA, ARP Timeout 04:00:00
  Last input 00:00:01, output 00:00:02, output hang never
  Last clearing of "show interface" counters 01:42:42
  Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 0
  Queueing strategy: fifo
  Output queue: 0/40 (size/max)
  5 minute input rate 7000 bits/sec, 1 packets/sec
  5 minute output rate 8000 bits/sec, 1 packets/sec
     38553 packets input, 5990235 bytes, 0 no buffer
     Received 2993 broadcasts (0 multicast)
     0 runts, 0 giants, 0 throttles
     3250 input errors, 2794 CRC, 436 frame, 0 overrun, 0 ignored
     0 watchdog, 2918 multicast, 0 pause input
     0 input packets with dribble condition detected
     39362 packets output, 17656364 bytes, 0 underruns
     73 output errors, 3 collisions, 2 interface resets
     0 babbles, 73 late collision, 5 deferred
     0 lost carrier, 0 no carrier, 0 PAUSE output
     0 output buffer failures, 0 output buffers swapped out

remote switch:
sw-remote>sh int fast 0/1
FastEthernet0/1 is up, line protocol is up (connected)
  Hardware is Fast Ethernet, address is 0008.219f.62c1 (bia 0008.219f.62c1)
  MTU 1500 bytes, BW 100000 Kbit, DLY 1000 usec,
     reliability 255/255, txload 1/255, rxload 1/255
  Encapsulation ARPA, loopback not set
  Keepalive set (10 sec)
  Full-duplex, 100Mb/s
  input flow-control is off, output flow-control is off
  ARP type: ARPA, ARP Timeout 04:00:00
  Last input 00:00:15, output 00:00:01, output hang never
  Last clearing of "show interface" counters 03:41:53
  Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 0
  Queueing strategy: fifo
  Output queue: 0/40 (size/max)
  5 minute input rate 8000 bits/sec, 2 packets/sec
  5 minute output rate 6000 bits/sec, 4 packets/sec
     177712 packets input, 45037887 bytes, 0 no buffer
     Received 4771 broadcasts (0 multicast)
     0 runts, 0 giants, 0 throttles
     543 input errors, 463 CRC, 51 frame, 0 overrun, 222 ignored
     0 watchdog, 2924 multicast, 0 pause input
     0 input packets with dribble condition detected
     184373 packets output, 20061876 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


Local switch is WS-C2950-24, remote is a WS-C2950-12, both are running the 
same standard image c2950-i6q4l2-mz.121-19.EA1.bin. I have not attached the 
configurations mainly because the message is allready too long. We tried 
already to change the switches (tried a SMC switch at remote location), 
media-converters, also patchcords (UTP and fiber), the behavior is same. If 
someone has seen this before, please share some ideas, it's becoming 
nightmare. Is it possible the problem is not related to the 2950s and we just 
have problems with the physical layer (faulty optical cable or splicing 
between the two locations) ?

Thank you for patience, sorry if I bothered the list members with this long 
message, please be indulgent as this is my first post to this list -:)

Best wishes.

=================================
Valeriu VRACIU
Network Engineer at RoEduNet Iasi
tel: +40 (232) 201003
GSM: +40 (744) 615251
e-mail: vvraciu at iasi.roedu.net
=================================
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQFA+rDKncI+CatY948RAl+kAJwMOS9jAPicCLOKcPhZCyGbJQXNmgCfYpBZ
7wMDCZz+t74upCI4VodVvrY=
=EKOC
-----END PGP SIGNATURE-----



More information about the cisco-nsp mailing list