My suspicion is that UPnP is causing this problem.
My tests showed that the httpd -S bug occurs even if the UPnP service is disabled (I don't use it).
Maybe UPnP is causing more troubles but the problem is still the bug in the HTTPS webservice with standard port 8080...
I bought a DIR-600 dlink router, and used dd-wrt for 3 weeks with no trouble at all. When I configured remote access and enabled https (8080), cpu utilization went to 100%. After killing httpd -S, it came back to normal (10%). Restarting httpd -S, with standard port seems to help. I'm using DD-WRT v24-sp2 (10/10/09) std (13064)
follow up:
due to the fact that this issue is still unresolved I recently switched to OpenWRT.
I used DD-WRT over years but this issue is driving me nuts...
I'll come back if there is any progress because I like the web interface
Posted: Wed Jan 20, 2010 22:53 Post subject: Changed from port 8080
Changed from port 8080 to 443 last night, so far no problems at all. Normally, it would have pegged @ 100% by now. So, the next question is, why in the heck does the port number cause a problem?
Posted: Thu Jan 21, 2010 5:19 Post subject: Re: Changed from port 8080
spadefoot wrote:
Changed from port 8080 to 443
I don't want to disappoint you but this helps only to make things a little bit better. I experienced the bug although I switched to 443. But the occurrence decreases after switching the port.
I'm not sure that this is the same problem as you wrote "https disabled".
The problem I'm facing is high CPU load due to the fact that https is enabled!
Try to telnet to your router when the web interface is dead and then run "top". You will get a list of all running processes and the CPU load of each process.
Posted: Sun Jan 24, 2010 1:24 Post subject: Re: Changed from port 8080
spadefoot wrote:
Changed from port 8080 to 443 last night, so far no problems at all. Normally, it would have pegged @ 100% by now. So, the next question is, why in the heck does the port number cause a problem?
Up over 4 days without a reboot. Don't know if it's solved, or just vastly improved, but one of the two is true.
[Edit] Just checked top, looks like httpd is using about 5.4% of CPU right now, which is a little elevated. I'll keep an eye on it over the next few days and see what happens.