[c-nsp] 3750 stack sync problems

[Adrian Chadd]

G'day,

I built a new 3750 stack today and I couldn't manage to bring all
7 members in. The members are a 48-port gige, 2x24 port gige, 3 x 48 port 10/100 and
1 x 24 port 10/100. This failed switch is a 24-port gige 3750.

00:29:57: %STACKMGR-4-STACK_LINK_CHANGE: Stack Port 2 Switch 5 has changed to state UP
00:30:07: %STACKMGR-4-SWITCH_ADDED: Switch 6 has been ADDED to the stack
00:30:02: %STACKMGR-4-SWITCH_ADDED: Switch 6 has been ADDED to the stack (NewCoreStack-4)
00:17:33: %STACKMGR-4-SWITCH_ADDED: Switch 6 has been ADDED to the stack (NewCoreStack-5)
00:30:05: %STACKMGR-4-SWITCH_ADDED: Switch 6 has been ADDED to the stack (NewCoreStack-2)
00:30:02: %STACKMGR-4-SWITCH_ADDED: Switch 6 has been ADDED to the stack (NewCoreStack-3)
00:30:04: %STACKMGR-4-SWITCH_ADDED: Switch 6 has been ADDED to the stack (NewCoreStack-7)
00:30:47: %STACKMGR-4-STACK_LINK_CHANGE: Stack Port 2 Switch 5 has changed to state DOWN
00:30:50: %STACKMGR-4-STACK_LINK_CHANGE: Stack Port 2 Switch 5 has changed to state UP
00:30:51: %STACKMGR-4-STACK_LINK_CHANGE: Stack Port 2 Switch 5 has changed to state DOWN
00:30:52: %STACKMGR-4-STACK_LINK_CHANGE: Stack Port 2 Switch 5 has changed to state UP
00:30:54: %STACKMGR-4-STACK_LINK_CHANGE: Stack Port 2 Switch 5 has changed to state DOWN
00:31:08: %STACKMGR-4-SWITCH_REMOVED: Switch 6 has been REMOVED from the stack
00:31:08: %CFGMGR-1-UNABLE_TO_SEND_RUN_CFG: unable to send running-cfg, bits: 0x007E, retrying...
00:31:04: %STACKMGR-4-SWITCH_REMOVED: Switch 6 has been REMOVED from the stack (NewCoreStack-4)

On switch-6:

Waiting for Stack Master Election (around 30 seconds)...
POST: PortASIC CAM Subsystem Tests : Begin
POST: PortASIC CAM Subsystem Tests : End, Status Passed

POST: PortASIC Stack Port Loopback Tests : Begin
POST: Found Stack port 1 Down
POST: Found Stack port 2 Down
POST: Skipping Stack port External loopback
POST: PortASIC Stack Port Loopback Tests : End, Status Passed

POST: PortASIC Port Loopback Tests : Begin
POST: PortASIC Port Loopback Tests : End, Status Passed

Election Complete
Switch 6 booting as Member, Switch 1 elected Master
Waiting for Port download... <hangs here>

I've replaced and reseated the stack cables in switch-6; all the switches were upgraded
to SEE3. I ran out of time before I could try inserting this switch elsewhere.

%CFGMGR-1-UNABLE_TO_SEND_RUN_CFG in documentation says "copy the entire line and
copy it to Cisco TAC". I thought I'd try here before I try my hand at the TAC.

Has anyone seen this before and have any tips for figuring out what the hell is
going on? I'd love to enable some more debugging on the stack slave to get more of
an idea of why its hanging.




Adrian