Sick DSL modem causes WRT54GS to hang.

I have a mystery. My home WRT54GS running DD-WRT v23 SP3 has been working just fine with a Westel something DSL modem. Yesterday, I needed a DSL modem immediately for a customers 6Mbit/sec DSL. Instead of driving to my office and grabbing a known good modem, I gave him my Westel DSL modem, and replaced mine temporarily with whatever I could find in the junk pile. That was an Efficient 5360 and later an Efficient 5260.

Both DSL modems are rather old, but should be sufficient for my

1.5Mbit/sec DSL line. They work but now the router is hanging. After about 30 minutes of traffic, the router can't be pinged from several wired and wireless PC's. If ping still works, DNS forwarding in the router fails to resolve addresses. I have to reboot the router every 30 minutes or it doesn't stay up. The modem does NOT hang and does not need to be rebooted.

Why would swapping DSL modems cause the router to hang?

The solution is obvious. I'll grab a working modem from my office and get rid of the Efficient 5260/5360 modems (tomorrow). However, I don't understand how a marginal DSL modem can cause a router to go insane.

Reply to
Jeff Liebermann
Loading thread data ...

Maybe the modem has a lease time and doesn't refresh on its own? (might be worth looking at the modem set-up)

Reply to
Cal Vanize

Cal Vanize hath wroth:

Which lease? LAN or WAN? In my case, both are static IP's.

The DSL modems are both setup as simple ethernet to ATM bridges. It's not like the later Speedstream 5100 and 4100 series, which can have the PPPoE login in the modem. The PPPoE is in the WRT54G router. In addition, I have a static IP address from my ISP, so it's certainly not the WAN side DHCP lease.

The LAN side is also static (I'm doing lots of port forwarding) IP: Ethernet adapter Local Area Connection:

Connection-specific DNS Suffix . : Description . . . . . . . . . . . : Intel(R) PRO/100+ Server Adapter (PILA8470B) Physical Address. . . . . . . . . : 00-02-B3-1E-39-ED DHCP Enabled. . . . . . . . . . . : No IP Address. . . . . . . . . . . . : 192.168.1.11 Subnet Mask . . . . . . . . . . . : 255.255.255.0 Default Gateway . . . . . . . . . : 192.168.1.1 DNS Servers . . . . . . . . . . . : 192.168.1.1 74.220.64.45 That IP is not going to change either.

You can see the status page (when it's up) at:

I would normally plug my PC directly into the DSL modem and see if I can keep it up and running. However, the system is running our neighborhood WLAN and I don't want to take it down. So, I've set DD-WRT to reboot every 30 minutes, which should be sufficient to keep it up for anything except large downloads.

I've sat monitoring the router statistics all afternoon looking for a suitable culprit. Unfortunately, I lose the connection when it goes insane. All I can find that's weird is the output of ifconfig:

eth0 Link encap:Ethernet HWaddr 00:14:BF:B9:10:13 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:27260 errors:0 dropped:0 overruns:0 frame:0 TX packets:28205 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:6294028 (6.0 MiB) TX bytes:6388671 (6.0 MiB) Interrupt:4

eth1 Link encap:Ethernet HWaddr 00:14:BF:B9:10:15 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 ==> RX packets:12428 errors:0 dropped:0 overruns:0 frame:28087 TX packets:15168 errors:204 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:1100450 (1.0 MiB) TX bytes:3093631 (2.9 MiB) Interrupt:2 Base address:0x5000

eth0 = WAN ethernet eth1 = LAN ethernet

Framing errors (corrupted packets) on eth1 (LAN side) seems ominous. Hmmmm.... (smoke coming from ears). I wonder if I have a half/full duplex mismatch on the LAN ethernet port. That has nothing to do with the WAN side DSL modem, but might be a bad LAN ethernet cable or something similar. Time to generate some traffic and see what breaks.

Reply to
Jeff Liebermann

Jeff Liebermann hath wroth:

Ooops. Backwards. eth0 = LAN. eth1 = WAN. That makes more sense as the errors are now at the DSL modem to router ethernet connection. NWAY problem I guess.

Reply to
Jeff Liebermann

Jeff Liebermann hath wroth:

Problem and mystery mostly solved. It was an NWAY negotiation problem between the WRT54GS and the Efficient DSL modems. The router crashes went away when I locked down the interface speed and protocol on the WRT54GS to 10Mbits/sec half-duplex. Finding where to do that was a bit of a challenge in DD-WRT. It's under: Setup -> VLAN's at the bottom of the page. Uncheck the boxes under the "W" column for "Auto-Negotiate", "100Mbit", and "Full Duplex". Apply and Save.

However, there are still some mysterious problems remaining (probably for another day).

  1. The large number of framing errors are stil there. The results of ifconfig after running about 12 hours is:

eth0 Link encap:Ethernet HWaddr 00:14:BF:B9:10:13 UP BROADCAST RUNNING PROMISC MULTICAST MTU:1500 Metric:1 RX packets:37911 errors:0 dropped:0 overruns:0 frame:0 TX packets:63331 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:25643053 (24.4 MiB) TX bytes:11167400 (10.6 MiB) Interrupt:4

eth1 Link encap:Ethernet HWaddr 00:14:BF:B9:10:15 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 =====> RX packets:12620 errors:0 dropped:0 overruns:0 frame:181201 TX packets:43403 errors:241 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:1632528 (1.5 MiB) TX bytes:21759628 (20.7 MiB) Interrupt:2 Base address:0x5000

It's working, so I shouldn't complain, but it would be nice to know from where the errors are coming from.

  1. The mystery of why a protocol negotiation failure at the WAN interface would trash the router, probably overscribbling working memory, remains unsolved. Similarly, why it would have an effect on DNS forwarding and LAN port operation, is also unknown.
Reply to
Jeff Liebermann

Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.