One specific WAN subnet not reachable via WiFi, but is reachable via Wired
Got a good one here...
I have a home based client on an Embarq DSL line. They have a ZyXel EQ-660R DSL modem connected to a Linksys WRT54GX2 router. THe dsl modem is acting as a router as well. It has a public IP on the WAN side, and uses 192.168.2.1 as its LAN IP.
The Linksys WRT54GX2 gets it's WAN ip from the ZyXel via dhcp. Currently 192.168.2.2. I know, not really a wan ip, it's private, but as far as the linksys is concerned, it's the route to the rest of the world.
WiFi is enabled, with WPA2 encryption.
All has been fine for a while.
Today, I got a call from them that they couldn't check their email. They had internet access. I remoted in, and found that they were unable to ping any IP on the subnet that their email servers were on (which we host). Our network was fine, not an issue there.
A traceroute from the laptop in question showed successful pings to:
192.168.55.1 (the linksys ip)
192.168.2.1 (the zyxel ip)
76.2.x.x (the embarq network)
138.210.x.x (another embarq ip)
then timetimouts for a bit
sometimes it would continue to an alter.net ip, then a level3 ip, but ultimately timeout before our public IP subnet.
I was able to ping my client's public embarq address from one of our servers on that unreachable subnet.
I was able to ping 1 hop away from our subnet from the client's machine, and the traceroute there was fine all the way including right after that 4th hop, where the first timeout was consistently happening when pinging the router for our subnet.
To eliminate this being some crazy problem with the laptop being used, I had them try a second laptop over wifi. Same results.
The linksys has a diagnostic option in the admin interface. It's able to ping and do a traceroute to the subnet in question without issue.
Grasping at straws, I tried having the client connect one of the laptops to the linksys router using an ethernet cable and turn off wifi. POOF! It started working. Full access to that subnet.
I've been doing this a long time, but have never seen anything like this.
If it's a problem with WiFi, how in the world would it get out to the embarq network, but stop after the 4th hop (same spot every time), but not get stuck using a wired connection via the same box? It just doesn't make any sense to me.
Thoughts? Ideas?
I have a home based client on an Embarq DSL line. They have a ZyXel EQ-660R DSL modem connected to a Linksys WRT54GX2 router. THe dsl modem is acting as a router as well. It has a public IP on the WAN side, and uses 192.168.2.1 as its LAN IP.
The Linksys WRT54GX2 gets it's WAN ip from the ZyXel via dhcp. Currently 192.168.2.2. I know, not really a wan ip, it's private, but as far as the linksys is concerned, it's the route to the rest of the world.
WiFi is enabled, with WPA2 encryption.
All has been fine for a while.
Today, I got a call from them that they couldn't check their email. They had internet access. I remoted in, and found that they were unable to ping any IP on the subnet that their email servers were on (which we host). Our network was fine, not an issue there.
A traceroute from the laptop in question showed successful pings to:
192.168.55.1 (the linksys ip)
192.168.2.1 (the zyxel ip)
76.2.x.x (the embarq network)
138.210.x.x (another embarq ip)
then timetimouts for a bit
sometimes it would continue to an alter.net ip, then a level3 ip, but ultimately timeout before our public IP subnet.
I was able to ping my client's public embarq address from one of our servers on that unreachable subnet.
I was able to ping 1 hop away from our subnet from the client's machine, and the traceroute there was fine all the way including right after that 4th hop, where the first timeout was consistently happening when pinging the router for our subnet.
To eliminate this being some crazy problem with the laptop being used, I had them try a second laptop over wifi. Same results.
The linksys has a diagnostic option in the admin interface. It's able to ping and do a traceroute to the subnet in question without issue.
Grasping at straws, I tried having the client connect one of the laptops to the linksys router using an ethernet cable and turn off wifi. POOF! It started working. Full access to that subnet.
I've been doing this a long time, but have never seen anything like this.
If it's a problem with WiFi, how in the world would it get out to the embarq network, but stop after the 4th hop (same spot every time), but not get stuck using a wired connection via the same box? It just doesn't make any sense to me.
Thoughts? Ideas?
akahan
You're right. It doesn't make any sense. The ONLY thing that occurs to me (and it's kind of a longshot) is that the Linksys box is sending malformed packets out, and those packets are being dropped by some router upstream from the client that's doing deep packet inspection and is perhaps overly fussy. But why would packets be malformed when they're wireless, but not wired? The ONLY thing I can think of is maybe because the LInksys has "concatenation" turned on in the advanced wireless settings? (This merges contents of several packets into one, under some circumstances, to save bandwidth...but if the result is a nonstandard packet, that packet might get rejected elsewhere.) I'd check to see if concatenation is enabled and, if so, try disabling it, just for kicks.
ASKER
Thanks for the reply akahan. Good thoughts.
I've factory reset the Linksys router just to be safe. no change.
As you suggested, I've looked at Advanced Wireless Settings, but I don't see concatenation. Here's what i've got:
Basic Rate Set: Basic Rate Set #2
Transmittion Rate: Auto
CTS Protection Mode: Disable
Beakon Interval: 100
DTIM Interval: 1
Fragmentation Threshold: 2346
RTS Threshold: 2347
Preamble Type: Long
Network Density: Low
I'm going to suggest that they pick up a new wifi router when they can.
This just doesn't make sense. it worked fine the last time they were at this location, this time around, no luck and for apparently no reason.
Some more info:
The 4th and 5th hops are both level3.net routers in seattle. When wireless, both of these time out when doing a traceroute. Hops 7-11 are all alter.net (verizon) routers (all with sub 100ms ping times), then we get to our ISP's gateway, which replies but after that, no matter if we're going to one of our IP's or other clients who are at that ISP (also our clients) on different subnets, we get nothing when WiFi. Turn on the wired connection and it works.
Nuts.
I've factory reset the Linksys router just to be safe. no change.
As you suggested, I've looked at Advanced Wireless Settings, but I don't see concatenation. Here's what i've got:
Basic Rate Set: Basic Rate Set #2
Transmittion Rate: Auto
CTS Protection Mode: Disable
Beakon Interval: 100
DTIM Interval: 1
Fragmentation Threshold: 2346
RTS Threshold: 2347
Preamble Type: Long
Network Density: Low
I'm going to suggest that they pick up a new wifi router when they can.
This just doesn't make sense. it worked fine the last time they were at this location, this time around, no luck and for apparently no reason.
Some more info:
The 4th and 5th hops are both level3.net routers in seattle. When wireless, both of these time out when doing a traceroute. Hops 7-11 are all alter.net (verizon) routers (all with sub 100ms ping times), then we get to our ISP's gateway, which replies but after that, no matter if we're going to one of our IP's or other clients who are at that ISP (also our clients) on different subnets, we get nothing when WiFi. Turn on the wired connection and it works.
Nuts.
ASKER
EDIT
It's the 5th and 6th hops that are the level3.net routers.
The 4th hop is hostXX.embarqservices.net,
It's the 5th and 6th hops that are the level3.net routers.
The 4th hop is hostXX.embarqservices.net,
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
still waiting on them to get a replacement router...
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.