1

any chance that wireless access point, wrt54g type with ddwrt, could cause a cisco router to go buggy and drop dhcp requests and other network issues?

looking for some advice as the units are within a foot of each other and this is turning into a blame game between two companies that installed the equipment. thanks gd

Shane Madden
  • 112,982
  • 12
  • 174
  • 248
dasko
  • 1,244
  • 1
  • 22
  • 29

4 Answers4

2

Are these hooked into the same network in anyway (wrt54g goes into the same switch as the router, or the router itself)? If that is the case, maybe you have some IP conflicts. If they are both in the same broadcast domain, and are both running dhcp, then when clients request IP addresses you don't know which will answer.

Since the dhcp servers won't know who is given what, you might end up with duplicate IPs.

Physically, the would not interfere with each other.

Kyle Brandt
  • 82,107
  • 71
  • 302
  • 444
  • if ever there was a site that had a router and a wifi box i would have turned off the dhcp scope on the wifi unit and made it a switch or just ap. there is no chance of ip conflicts, as well the layout is that one company has a cisco router and the other company has a wifi router box, they are seperate networks and isp setups. the complaint from the company that has the cisco box is that the wifi router, about 2.5 feet away, is causing their dhcp scope to not work on the mitel ip phones. i thought it was just an excuse for another issue they can't fix. thanks. –  Nov 19 '09 at 18:28
0

Logically, if configs are conflicting, sure. Physically located next to each other, I doubt it, unless the Linksys has some hardware malfunctioning in it. Is it possible to just shut off the Linksys temporarily?

DanBig
  • 11,393
  • 1
  • 28
  • 53
  • that is my next step when i am back on site, i just wanted some feedback from others since i have these types of devices at over 100 locations without ever having an issue.gd –  Nov 19 '09 at 18:29
0

I once had a device with a flakey powersupply. I don't understand exactly why, but it caused issues with other devices plugged into the same power distribution unit. So, if they are plugged into the same circuit, it is possible you might see something there, but I don't see where the physical location of a Wireless AP would affect a non-wireless device.

THe other issue I have seen, is if a client has a wired and wireless connection, and the connections are set to Bridge for some reason, you can get all kinds of wierd IP duplications with some hardware based DHCP servers. My case was a SonicWall Firewall that I was using as a DHCP, and every time this one consultant was is in the office, I'd start getting IP conflict errors all over the network. The issue was traced back to the consultant's laptop creating a second path in the network that messed with the processing of DHCP requests and acknowledgements.

BillN
  • 1,503
  • 1
  • 13
  • 30
  • just an update for everyone, there is no ip conflict, it is just dhcp scopes that don't work on one network, i really think it is something on their cisco box and they are just using the wifi unit as an excuse claiming rf interference.gd –  Nov 19 '09 at 18:34
  • You say in another comment that they are two separate networks, for two companies. So I'm guessing that they share a phone closet in an office building or something. Is it possible that someone plugs a patch cable between the two? – BillN Nov 19 '09 at 21:33
  • yes they do share the same location. they are not physically connected. the phone IT company says that the wifi box, being in close proximity rf-wise, is causing the cisco router to flub up. they are basically blaming wifi rf interference on the reason why the ip phones act up. this is why i was saying that it is more than likely their cisco hardware or linksys poe switch that is buggy. they are just looking for something to blame. i hope this makes sense now.gd –  Nov 20 '09 at 00:54
0

If these devices are 1 foot apart, how long is the patch cable connecting the wireless router to the network? If it is running FastEthernet, there is a minimum 3 feet requirement for patch cables. Anything shorter can cause network connectivity problems. If it is connected to Gigabit, this "rule" doesn't apply, but should still be followed. I would make sure there is a 10 foot patch cable connecting the wireless router to the network.

Kevin Garber
  • 311
  • 2
  • 3
  • i thought fastethernet solved the "carrier sense multiple access with collision" issue.? gd –  Nov 19 '09 at 20:06
  • No, 100Base-TX is a problem as well. Here is the excerpt from Wikipedia. All copper Ethernet segments that run the Collision Detect (CD) portion of CSMA/CD have a minimum cable length to function properly because of reflections. This applies only to 10BASE-T and 100BASE-TX standards; The 1000BASE-TX standard is covered at the end of this section. http://en.wikipedia.org/wiki/Ethernet_physical_layer#Ethernet_minimum_cable_lengths – Kevin Garber Nov 19 '09 at 20:15