Horrible Design
-
We've just finished a script to add VLANS to our first host machine... this just seems wrong. I gotta believe this is not what intel was intending. http://beatdownbox.com/2950.jpg[^] and trust me you don't want to see the routing table either :)
-
We've just finished a script to add VLANS to our first host machine... this just seems wrong. I gotta believe this is not what intel was intending. http://beatdownbox.com/2950.jpg[^] and trust me you don't want to see the routing table either :)
Now, think about a network virus. :doh:
Nobody can give you wiser advice than yourself. - Cicero .·´¯`·->ßRÅhmmÃ<-·´¯`·.
-
Now, think about a network virus. :doh:
Nobody can give you wiser advice than yourself. - Cicero .·´¯`·->ßRÅhmmÃ<-·´¯`·.
Maybe it would get confused and infect itself?
-
Maybe it would get confused and infect itself?
Now, think about a firewall. :rolleyes: A few hundred popup windows at the taskbar alerting you that some particular application is trying to access a network resource.
Nobody can give you wiser advice than yourself. - Cicero .·´¯`·->ßRÅhmmÃ<-·´¯`·.
-
We've just finished a script to add VLANS to our first host machine... this just seems wrong. I gotta believe this is not what intel was intending. http://beatdownbox.com/2950.jpg[^] and trust me you don't want to see the routing table either :)
-
We've just finished a script to add VLANS to our first host machine... this just seems wrong. I gotta believe this is not what intel was intending. http://beatdownbox.com/2950.jpg[^] and trust me you don't want to see the routing table either :)
Err... evein if it seems worst (and in fact it is :) ) it is correct: if you place an host on 100 VLAN you are giving it 100 network virtual connection with one IP address each and one subnet each. The "real" problem is a wong design of the network. If you have so many VLANS you shild have your host on one only VLAN and the routing between VLAN to be managed by networking devices. I wonder if you ever heard about "Routers". :^)
2 bugs found. > recompile ... 65534 bugs found. :doh:
-
Err... evein if it seems worst (and in fact it is :) ) it is correct: if you place an host on 100 VLAN you are giving it 100 network virtual connection with one IP address each and one subnet each. The "real" problem is a wong design of the network. If you have so many VLANS you shild have your host on one only VLAN and the routing between VLAN to be managed by networking devices. I wonder if you ever heard about "Routers". :^)
2 bugs found. > recompile ... 65534 bugs found. :doh:
What we're doing is simulating 100 stores, so each store is its own subnet with its own router... thats why we need all the VLANS...
-
What we're doing is simulating 100 stores, so each store is its own subnet with its own router... thats why we need all the VLANS...
But you also need 100 virtual machines. One machine with 100 network interfaces (physical or virtual doesn't bother) attached to 100 routers (belonging to a same routed noetwork) dosn't demonstrate anything but how messy is the routing configuration. I suspect you're just another victim of the "one default gateway per network connection issue", that is due to the fact the Os GUI (note: only the GUI) propose the default gateway as a connection paramenter. It's not. It is an entry of the routing table that is unique for asystem. Is a GUI conceptual bug that exist from Win3.11 and had never been properly corrected. Just read the RFC791 and see why. Consider the idea not to interface 100 networks, bu to be on the 101th network (and let the routers to do the routing), ot to be with 100 host on once network each. What you did is a "networking diamond of death". Something networking engeneers continuosly battle to avoid.
2 bugs found. > recompile ... 65534 bugs found. :doh:
-
But you also need 100 virtual machines. One machine with 100 network interfaces (physical or virtual doesn't bother) attached to 100 routers (belonging to a same routed noetwork) dosn't demonstrate anything but how messy is the routing configuration. I suspect you're just another victim of the "one default gateway per network connection issue", that is due to the fact the Os GUI (note: only the GUI) propose the default gateway as a connection paramenter. It's not. It is an entry of the routing table that is unique for asystem. Is a GUI conceptual bug that exist from Win3.11 and had never been properly corrected. Just read the RFC791 and see why. Consider the idea not to interface 100 networks, bu to be on the 101th network (and let the routers to do the routing), ot to be with 100 host on once network each. What you did is a "networking diamond of death". Something networking engeneers continuosly battle to avoid.
2 bugs found. > recompile ... 65534 bugs found. :doh:
Guess I should have also mentioned that this is a virtual host. There can be up to 400 different virtual guests on here (only 4 at a time) for a lab environment.