Our customer have a problem in his network which is based on six T1600G-52TS switches.
Some days the switches cpu goes up to 90% and the newtork crashes resulting in a force reboot of all the switches, and the situation returns to normal state.
We have already applied loopback detection, broadcast storm control, removed the STP redundant links but the situation remains the same.
We did some internal tests with three types of switches: T1600G-52TS, T1700G-28TQ, T1700X-16TS.
All three had factory settings.
We first connected the switches to the network and to a PC. Then only a PC, leaving the network disconnected.
In both cases we launched an angry ip scanner on the default range.
We tried to launch the scan both from the PCs connected directly to the switches and from those connected to the network but the result does not change.
In both tests the T1600G-52TS reached 75% of CPU while the T1700G-28TQ and the T1700X-16TS remained on 8-10% of cpu.
Is there a problem with T1600G series?