Procurve networking issues - RSTP?

[tchristopher]

Here's my topology: an HP Switch 5412zl(J8698A) and an HP ProCurve switch 5308xl.. these are out bacbones. We have about 30 Procurve 2510G-48's connecting to these two backbones via fiber.

At random times, only on non-business hours, there are 5 switches that simply fall off the network. I look in the logs and it shows that they're are quite a few RSTP root changes. Eg:

I 06/08/13 09:35:35 stp: RSTP Root changed from 0:f06281-758e00 to 32768:001185-388c80

I 06/08/13 09:35:35 stp: RSTP Root changed from 32768:001185-388c80 to 0:f06281-758e00

I 06/08/13 09:35:41 stp: RSTP Root changed from 0:f06281-758e00 to 32768:001185-388c80

I 06/08/13 09:35:55 stp: RSTP Root changed from 32768:001185-388c80 to 0:f06281-758e00

I check the logs on the backbone and see the following:

W 06/08/13 05:20:02 FFI: port F1-High collision or drop rate. See help.

W 06/08/13 05:20:02 FFI: port F8-High collision or drop rate. See help.

W 06/08/13 05:20:02 FFI: port F10-High collision or drop rate. See help.

W 06/08/13 05:20:02 FFI: port F13-High collision or drop rate. See help.

W 06/08/13 05:22:11 FFI: port D13-High collision or drop rate. See help.

W 06/08/13 05:41:47 FFI: port D13-High collision or drop rate. See help.

W 06/08/13 05:55:47 FFI: port D13-High collision or drop rate. See help.

W 06/08/13 06:12:31 FFI: port A5-High collision or drop rate. See help.

W 06/08/13 08:12:55 FFI: port A5-High collision or drop rate. See help.

If you look at the time stamps, they're not the same time, but when I log in to the backbones at the time of the issues, I see 100% utilization on all ports.

I'm baffeled why it's affecting some switches but not others, and it happens all at the same time.

ANy ideas??

P.S. This thread has been moved from Switches, Hubs, Modems (Legacy ITRC forum) to ProCurve / ProVision-Based - HP Forum Moderator

[Vince_Whirlwind]

On the 5412 aned 5308, do this:

 sh run | in spanning-tree
and copy what it says here.