Re: Potential DoS on Bittorrent
From: anthonyberet (nospam_at_me.invalid)
Date: 08/28/05
- Next message: tomstdenis_at_gmail.com: "Re: The importance of IVs"
- Previous message: anthonyberet: "Re: Potential DoS on Bittorrent"
- Maybe in reply to: anthonyberet: "Re: Potential DoS on Bittorrent"
- Next in thread: Joseph Ashwood: "Re: Potential DoS on Bittorrent"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Date: Sun, 28 Aug 2005 10:13:51 +0100
Twisted One wrote:
> Stephen Sprunk wrote:
>
>> Same for the one I'm reading, at least in the .torrent file. Perhaps
>> once a
>> piece is started, the sender directly transmits hashes of the
>> subpieces to
>> the receiver? See below on why I think this may happen.
>
>
> This would suck. If we assume the sender of subpieces is untrusted, then
> the hash they send is untrusted. It will match the data, even if the
> data is bogus and the big piece will hash-fail, just so they can't be
> fingered as the bad node, and so the downloader has to get the whole
> piece again and not just the bad subpiece.
>
Tut! - the HASH is in the .torrent file, which are usually to be trusted.
- Next message: tomstdenis_at_gmail.com: "Re: The importance of IVs"
- Previous message: anthonyberet: "Re: Potential DoS on Bittorrent"
- Maybe in reply to: anthonyberet: "Re: Potential DoS on Bittorrent"
- Next in thread: Joseph Ashwood: "Re: Potential DoS on Bittorrent"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Relevant Pages
|
|