Printable View
I have been attempting to download the same file now over half a dozen times and I'm on a 10mb cable connection but download speed is less than 10k average. I have gotten as high as 97% before timing out and than I have to start over completely since resume is not supported. Even IE has resume support capability if the server can resume. So I have spent an average of 12 minutes per attempt that is over an hour of server time for a single file and still have not gotten the file yet.
We have no money to pay for more servers, and resume support is not coded into our scripts you would have to use a download manager, which we don't allow.Quote:
Originally Posted by molitar
Tomorrow we will be going public with our direct connect hubs and you might have better luck with that, but we are working on trying to find another way to gather funds for more machines.
IMO not having resume support is more harmful to bandwidth, for example sometimes I have to retry a file up to 20 times, sometimes only getting up to 5% and sometimes up to 90%, on a 5mb file with with an average of 50% completed that is a wasted 50mb transfer.
I can understand you don't want resume support due to multithreaded download managers, but personally I would do IP list checking, and throw a 403 to all 2nd connection attempts from the same IP.
yea but then the download mangers not setup right hammer the server with attempts to reconnect which is just as bad as active connections.
Its easier to just say no download managers. Keeps the people who don't know how to use it from screwing up.
That is why I said 403 specifically, as any program I have seen will realise that hammering on a 403 is pointless
The other option is 410, since that indicates the file is not there, and isn't comming back, only problem with that is that it is rather obscure, and any badly written programs wouldn't have a handler for it.Quote:
Originally Posted by w3.org - 10.4.4 403 Forbidden