Switch reload by pinging Catalyst 2950 switch with option ping -r

[mvanleijden]

I have a Catalyst 2950 (24 port) switch with C2950 Software (C2950-I6Q4L2-M), Version 12.1(11)EA1. I have the following problem:

When I ping my Management VLAN IP address on the switch I get a normal response. If I use option -r (Record route for count hops-option) I give the switch a unexplainable reload which is not normal i guess.

Does anyone have a suggestion for the right IOS to use, to solve this IOS bug?

[Frederic Vanderbecq]

Could it be possible to collect a console dump from the switch when it crashes. We can try to decode it. It might be interesting to open a case with the TAC as well.

[mvanleijden]

Logging information:

00:01:19: %PLATFORM_CATALYST2950-1-CRASHED: System previously crashed with the following message:

00:01:19: %PLATFORM_CATALYST2950-1-CRASHED: Crash info file is flash:/crashinfo/crashinfo_6

Crashinfo_6:

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

C2950 Software (C2950-I6Q4L2-M), Version 12.1(11)EA1, RELEASE SOFTWARE (fc1)

Compiled Wed 28-Aug-02 10:25 by antonino

Signal = 10, Code = 0x8, Uptime 1d03h

r0: 00000000, AT: 00000000, v0: 80800000, v1: 00000007

a0: 80670000, a1: 0000E700, a2: 80DC4DE8, a3: 00000000

t0: A08F56FC, t1: 1000E701, t2: 10000000, t3: FFFF00FF

t4: 802F3CF0, t5: 04000000, t6: 00000000, t7: 00000355

s0: 807C0000, s1: 00000000, s2: 00000000, s3: 000002BC

s4: A08F56EE, s5: 00000008, s6: 80CCD814, s7: 00000001

t8: 00000000, t9: 00000000, k0: 00000000, k1: 00000000

gp: 8073C620, sp: 80CDC3E8, s8: 00000000, ra: 8043C34C

EPC: 8043B7DC, ErrEPC: 43EE17AB, BadVA: 0x00000005, SREG: 1000E703

Cause: 0x00000008 (code 2): TLB (load or ifetch) exception

Signal 10, Exception code (0x0008)!

Frame 00: SP = 0x80CDC3E8 PC = 0x8043B7DC

Frame 01: SP = 0x80CDC440 PC = 0x8043C34C

Frame 02: SP = 0x80CDC4C0 PC = 0x801BEBE4

Frame 03: SP = 0x80CDC4D8 PC = 0x801BEBD0

-Traceback= 8043B7DC 8043C34C 801BEBE4 801BEBD0

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

[KENT EITZMANN]

For what's it worth I have a WS-C2950T-24 that is running the same image, but I can't reproduce the same problem (thank goodness). I tried pings from Solaris & Linux. In case it matters I have the management VLAN and all the ports in vlan 1.

If you have a chance could you post the exact model of the 2950 you have, also are you using a separate vlan for management ?

[mvanleijden]

I use another vlan ID (VLAN 700) for Switch management in stead of the default VLAN 1, It could be possible that this cause the problem.

I use exactly the same model 2950 switch w/ 24 port fast ethernet and two port Gigabit ethernet ports.