Negative Interface Control counters on fa interface

ajaju · ‎03-17-2011

Hi all,

I have a query regarding an HWIC-2FE module on a Cisco 3845 router.

The show controllers output shows negative counter for InUnicasts and InGoodOctets and I would like to know what it means.

Here is the output of show controllers fa0/0/1:

wc2437-r2-OLD#sh controllers fastEthernet 0/0/1

Interface FastEthernet0/0/1

 HWIC Registers

 Status = 0x2700            Control = 0x335

Intr Event= 0x0     SMI Spd = 0x6

Marvell switch registers=0x43800800

Number of Packets received with errors:

Bad Marvell Trailer: 0   Packets received

Marvell PHY (1) Register:

Reg 29 = 0x4380087A         Reg 30 = 0x4380087C

Control    : 0x100          Status      : 0x784D

Identifier : 0x141  0xC89

Auto Neg   : 0x1E1          Auto Neg Exp: 0x4

LP Ability : 0x0            LP NxtPg    : 0x0

PHY ctrl   : 0x4130         PHY status  : 0x2C40

Intr enable: 0x400          Intr status : 0x0

Port (1) Registers:

Status     : 0x3A00         Control     : 0x77

Identifier : 0x831          Vlan Map    : 0x2200

VID and PRI: 0x2            Rate Control: 0x30

Port Asc Vt: 0x2

TX counter : 0xEE           RX counter  : 0x15F2

Port (CPU PORT=9) Registers:

Status     : 0x3A00         Control     : 0x77

Identifier : 0x831          Vlan Map    : 0x2002

VID and PRI: 0x2            Rate Control: 0x30

Port Asc Vt: 0x200

TX counter : 0x15F2         RX counter  : 0xF0

 Software MAC address filter(hash:length/addr/mask/hits):

  0x00:  0  ffff.ffff.ffff  0000.0000.0000       115

  0x2D:  0  001f.ca58.fbe7  0000.0000.0000 -298152099

  0x58:  0  0100.5e00.0006  0000.0000.0000         0

  0x5B:  0  0100.5e00.0005  0000.0000.0000         0

  0xC0:  0  0180.c200.0002  0000.0000.0000         0

  0xC0:  1  0100.0ccc.cccc  0000.0000.0000         0

  0xC5:  0  0180.c200.0007  0000.0000.0000         0

ATU table:

   Mac Address     |   Ports   |  Entry State  | dbnum

---------------------------------------------------------

00-1F-CA-58-FB-E6  |    0x100   |     0x0F   |  0x01

1F-CA-58-FB-E6-00  |    0x200   |     0x0F   |  0x02

FE unit = 0

-------------

Counters Names Port=1, CPU_PORT=9

-----------------------------------

        InUnicasts      = -293659729, 263600355

        InBroadcasts    = 119, 63

        InPause         = 0, 0

        InMulticasts    = 0, 52707

        InFCSErr        = 0, 0

        AlignErr        = 0, 0

        InGoodOctets    = -1835494382, 1339715564

        InBadOctets     = 0, 0

        Undersize       = 0, 0

        Fragments       = 0, 0

        In64Octets      = 1724907743, 932581161

        In127Octets     = 548797536, -1 662073437

        In255Octets     = 551520051, -959709782

        In511Octets     = 209367092, 461161551

        In1023Octets    = 219105611, 413034618

        InMaxOctets     = 747609653, 1078659014

        Jabber          = 0, 0

        Oversize        = 0, 0

        InDiscards      = 0, 0

        Filtered        = 0, 0

        OutUnicasts     = 263600354, -293659728

        OutBroadcasts   = 63, 119

        OutPause        = 0, 0

        OutMulticasts   = 52707, 0

        OutFCSErr       = 0, 0

        OutGoodOctets   = 1339715492, -1835493253

        Out64Octets     = 932581161, 1724907743

        Out127Octets    = -1 662073438, 548797536

        Out255Octets    = -959709782, 551520051

        Out511Octets    = 461161551, 209367092

        Out1023Octets   = 413034618, 219105611

        OutMaxOctets    = 1078659014, 747609654

        Collisions      = 0, 0

        Late            = 0, 0

        Excessive       = 0, 0

        Multiple        = 0, 0

        Single          = 0, 0

        Deferred        = 0, 0

        OutDiscards     = 0, 0

Please advise on the same.

Thanks in advance.

Zizhen Gao · ‎03-20-2011

This is certainly a bug.I believe that CSCtn95356 was opened to address this?

zz

ajaju · ‎03-21-2011

Hi Zizhen,

thanks for your reply.

I had actually opened that bug myself when this issue occured.

Was hoping to get a quick solution from the CSC (from past expereinces of all users) coz the bug may take a lot of time to resolve.

Guess I will have to wait for it to be resolved then.

I have actually recreated this issue in the Cisco Lab and I found that when the counter value goes above 2147483647 (this is the highest 31-bit number), the counter becomes negative ( -2147483647;as soon as it becomes a 32 bit number). The counter value then keeps incrementing towards the positive side, and once it reaches 2147483647 again, it becomes negative again and this process keeps continuing.

Although this does not have any affect on the functioning of the router or of any interface as such, the counter going negative does not make any sense whatsoever. So this is more of an enhancement bug, which needs some looking into.

Regards,

Aniruddha