well, who really knows. No one knew how the DNS issue would cause problems, until someone can take a deeper look into it. Perhaps, someone can tell us how to through the dlink in debug mode, or add a cpu chart in the status page.
Could be a rogue thread causng high cpu usage when it's logging too many requests than it can really handle, which is what I had assumed with my problem. My logs have never reached 2000 entries from the times I went in there viewing them, Especially in 2 minutes!
By the way, I think both are correct in terms of those tick boxes. I did in fact untick them all, then waited, did some stuff, went back in, put a tick in all of the boxes, hit refresh, and there was very few entries. So, the dev's wording is not very accurate because I decifered it as filtering out unwanted entries not DISABLING logging for those particular entries. Since then, the performance had improved when I applied that certain web only policy, when people were running p2p applications.
It's something that really cannot be avoided since the prior connections, and we know how many connections p2p makes, will keep on DDoS'n the router in an attempt at trying to re-establish a connection to the p2p host. The only work-around for this, would be to apply such policies during a period where there is very little network usage.
