Catalyst 1900 switch issue - internal speed

birving2008 · ‎06-03-2008

Hello,

I have a catalyst 1900 switch which has been working for years without any issues. It is connected to a single subnet on one side (10 machines all on same class C) and to a service provider switch on the other.

Recently the ftp speeds from machine to machine on the inside became very slow (20K) on a 10M port, slightly faster on the 100M port.

The external (internet side) to internal machines however has not changed and is still fast as ever.

Has anyone ever come across this sort of problem and if so where could I start to troubleshoot it ?

Thank You

Burt

michael.leblanc · ‎06-03-2008

Not familiar with the 1900 user interface (CLI or menu driven, not sure).

You might want to take a look at interface counters to see if you have one or more with an exceptional number of errors.

Best to clear the counters, and then monitor them over time.

If you identify a specific port that concerns you, try swapping the patch cable connected to the port with a known good one, to eliminate that as a potential cause.

Take a look at the speed/duplex settings of the port and the host that connects to it to ensure they are compatible.

Consider whether an issue has developed with a specific switch port (i.e.: failing), and potentially retire the port (shut it down if you can), and relocate the host to a different port.

Perform and evaluate one step at a time of course.

Pravin Phadte · ‎06-03-2008

Its a bit tuff for me to understand since i am not fully aware of the setup.

1900 Switch is been configured with Vlans ?

Is there any STP running in the network

Whats the swithes up time ?

does show logging shows any errors ?

what about show int outputs any errors ?

Did the switch reload ?

show flash does is show any crashinfo ?

It would be more easy for us if these questions can be answered to isolate or point out to the problems ?

Show tech-support output would be most helpful .

Reagrds,

pravin

birving2008 · ‎06-03-2008

Hi,

Thank you for your response.

The switch (LAN side) is connected directly to 10 machines (these are webservers with multiple IPs configured)

xx.xx.xx.6-250

Switch itself is xx.xx.xx.254

*** Uptime is 19days (after a reset to try to find this problem)

*** STP (port fast mode) is enabled on all ports

*** Monitoring the receive statstics for each port shows FCS errors and Alignment Errors about 2% (each) of the total good frames. I am not sure if this is normal however there were always some FCS and Alignment errors.

*** Monitoring transmit statistics shows no errors.

*** the switch reloads when reset

*** not sure how to get crashinfo on this switch.

You can contact me offlist at cisco@harpinc.net if you need report info as this appears to be limited to 4000 chars.

THank You

Burt

birving2008 · ‎06-03-2008

Hi,

The interface is menu driven.

I am getting about the same %ge (2%) of receive errors to good frames on all ports.

The problem started at the same time on all ports. THat is why I am thinking switch configuration rather than cable/computer issue.

All devices are Full Duplex compatible and that is how the ports are set.

THank you

Burt

Pravin Phadte · ‎06-03-2008

Hi birving,

I feel totaly diffrent way then you do.

Was there any config change done on the switch or in the network ?

If not then its not the config.

The errors are very less and may be ignored.