Replacing a HP L3 Switch with a Dell PC6248 and trying to get the configuration right. The HP switch had a lot less options since it was pretty darn cheap and sufficed for what we needed at that time.
The uplink is through a cisco switch managed by our ISP.
Testing routing between the VLANs works in testing - got a desktop set up with .231.100 and a laptop with .211.100 and they can run speed tests against eachother with max speed. The same goes for the .1.164 when i switch the laptop to it.
BUT when I set the laptop to .211.100 and then connect our uplink to 1/g1 I am no longer able to ping .211.1 or .1.164 or .231.1. But I am able to ping hosts on the internet and and our ISPs switch in the building .161.1.
As soon as I pull out the cable, Im then able to ping locally on the switch again.
The switch CPU may be getting hit with traffic from the ISP. This could prevent ping response.
Also, do you have any Cisco switches in your network connected directly or indirectly to this switch? Having Cisco switches connected via a standards based switch can cause strange problems.
I just tried your config and did not have any problem.
Instead of pinging the switch interface when you are seeing the problem, try pinging a PC plugged into the switch. If this works, this tells us the switch CPU is being hit by something.
This switch connects to our ISPs Cisco switch in the basement of our building. Is there a standards issue between Dell and Cisco switches?
The problem ONLY surfaces when I connect our ISP to the switch. I have had 3 different PCs connected to the switch - one for each VLAN and have been able to ping and use max bandwidth between those 3 PCs.
When I remove one of the PCs and connect the ISP to the switch it starts to act weirdly. I have no problem connecting to the "outside" from my PC but when I try and ping my gateway it simply gives me about 80% packet loss or worse. A trace to an outside source shows that it fails getting the first node (gateway) but continues on without a problem.
When I disconnect the ISP I am able to ping the gateway again.
I realized I hadn't turned off OSPF and RIP which I did - all default traffic should go straight to our ISP unless it is connected through our switch. This helped with with the packet loss to the ISP - I was now able to ping the gateway at our ISP without any loss.
When I try connecting to the web UI for the powerconect with the uplink active im getting huge packet losses - up to 100%. So i try pinging a PC on the switch, same thing there and same goes for pinging the switch.
It feels like a routing issue but I just can't wrap my head around it.
Any help you guys could give me before we hire someone to look into it?
I've tried switching uplink ports just to make sure there wasnt any error with that specific port.
Tried with disabled SPT, forced 1000F negotiation (thats what our ISPs switch is set to).
I also set the forward net directed broadcasts to enabled on all VLANs just to see if it made any difference.
bh1633
909 Posts
489
0
Posted April 18th, 2007 18:00