I'm experiencing a strange problem that I think my Debian server may be responsible for, but I've no idea how to troubleshoot it!
For the last month or so, I've been experiencing stalls - all access to the internet stops for a minute or two. It happens frequently...but intermittently. I assumed it was due to my ISP being rubbish...but then I noticed that I couldn't access my DSL router's admin during these stalls either. Well, I decided to solve the mystery once and for all today, and I'm even more confused now! My setup is the rather conventional: LAN<=>Eth0 - Server - Eth1<=>Router<=>Internet The server is designated as the DMZ host on the router, and the server NATs everything going out. I found that when connected to the wireless router via wifi, internet access remained constant throughout the stalls (pinging a remote host). Pinging the same host from my server during the stalls was also fine. Pinging the server's eth1, the router, or the remote host from my workstation during a stall fails, but works otherwise. This leads me to suspect that something's amiss with the NAT process somehow. FWIW, both network cards on the server are gigabit, though the router only supports 100Mb. Driver-wise, eth0 is using e1000, and eth1 is using r8169. An alternative is that the problem lies with the routing between eth0 and eth1. On the off-chance, I've tried turning off auto-negotiation, I've tried updating the firmware on the router, and I've tried replacing the cables, all to no avail. I don't even know where to begin looking for info on what the problem is. Any ideas? - Jamie -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org