Cat. 3500xl troubleshooting

[david-flores]

Hi, I recently installed a catalyst 3548 via fiber to my 6509. I setup dot1q trunking on both interfaces. I recently had an issue where users connected to the 3548 were experiencing very slow response times from a server on the same switch, and same vlan. I also was experiencing the same slowness in trying to telnet to the switch, and many timed out replies when doing an extented ping from my pc on a different vlan. Can anybody recommend any troubleshooting steps? Syslog did not show any unusual error messages, neither did the output of show log, I eventually replaced the entire switch with another, and so far, the issue has gone away. I want to test the switch somehow to see if it is an actual hardware problem, or something else. Thanks.

[sundar.palaniappan]

Things you might want to do on the switch that you replaced:

1. Hardcode the speed/duplex

2. Enabled STP portfast on ports connected to end devices, like PCs or servers

3. Check interface for high number of broadcast. If you do see, high broadcast traffic, identify what's causing it

4. Check CPU for high utilization

If none of the above seem to be an issue and you see no errors in the switch, there's a good chance something is wrong with the hardware as swapping it appeared to have cleared the problem.

HTH,

Sundar

[david-flores]

Hi, thanks. How many broadcasts would you consider high? I have seen as many 3000 on certain interfaces, since replacing the switch.

[sundar.palaniappan]

That depends on what applications and protocols you are using. For eg, RIP, IPX, video, AT all will cause high number of broadcasts. Some applications also rely heavily on broadcasts. STP loops can also cause high broadcasts but then it's very hard to miss any STP problems.

HTH,

Sundar