Strange net problems with a Netgear FS608 switch

This was weird and now that the switch is replaced I haven’t been able to duplicate it, but let me explain. There was a netgear fs608 (8 port unmanaged) switch plugged into a linksys router (model number not noted.) The cable was straight (although the fs608 has support for link through straight or crossover cables.) This setup worked well for quite some time. 4 computers and a printer hooked up. 3 pcs with fixed address and 1 with DHCP for their IP address. Well, I had a call that two pcs were unable to connect to the network and when I got there and looked… sure enough 169.**** ip addresses from Microsoft’s “auto configure” pool.

OK – so maybe the ports have gone bad. The server (after all) has a fixed IP is plugged into the same switch and can talk to the internet just fine. OK so I tried another port and tried to get the dhcp to work. No luck, so I moved the two problem machines over to the 4 port switch built into the linksys router. DHCP worked, they got addresses and were on the network. (the server was still on the netgear.) So, I noticed one machine was trying to synchronize files and complained that the server was unavailable. Huh? So, I went to start… run and typed in the name of the server and I could browse the folder that was “unavailable”.

So, I went to My Computer/Network places/ entire network/windows network and the domain (workgroup) name and saw the “other” problem machine, but nothing else appeared. So, essentially, the netgear router was not allowing broadcast requests I suppose? (I’m thinking DHCP is done by a broadcast approach as is the “network browsing” of the network places menu.) Direct connections (mapped drives, type in the machine name, etc.) all seemed to connect just fine, but broadcast “stuff” seemed to be what failed.

So, we swapped out the switch and everything is now working fine, so I started to test the netgear switch to see if I could reproduce the failure. I plugged it into one of the switches here and plugged the testbox in. It picked up a dhcp address right away and then broadcast pings across (2) other switches worked – so…. that was weird. It’s now going into my suspect hardware pool. (Which includes any hardware that has proved itself faulty once, but the failure hasn’t been replicated (or hasn’t repeated with enough severity to make it COMPLETELY useless.) (Yes, I’m a packrat….)

   Send article as PDF   

Similar Posts