From: "Robert A. Rosenberg" <hal9001@panix.com>
Date: Fri, 1 Sep 95 21:38:36 PDT
To: droelke@rdxsunhost.aud.alcatel.com (Daniel R. Oelke)
Subject: Re: SSL search attack
Message-ID: <v02130504ac6d39583ee9@[166.84.254.3]>
MIME-Version: 1.0
Content-Type: text/plain


At 07:25 9/1/95, Daniel R. Oelke wrote:
>>
>> I see nothing wrong with the concept of being allocated an initial chunk
>> and having the scan software attempt to ACK it when 50% of it has been
>> searched. A successful ACK would allow the releasing of a new chunk (in
>> response) equal in size to the returned chunk. A failure of the Server to
>> accept the ACK would trigger a retry at set intervals (such as 75% and 100%
>> or 60/70/80/90/100%) until the Server responds. Thus the scanner is always
>> in possession of a Full Sized Chuck to scan (so long as the Server accepts
>> an ACK before the 100% done mark) and temporary failures will not stop the
>> process of a scanner as currently happens.
>>
>
>The only way this can work is if the server is told it is a 50%/75%/etc
>size ACK, and then latter the server is ACKed for the full 100%.
>
>Why?  Because what happens if the client dies immediately after doing
>the ACK - maybe only 51% of that space has been searched, yet
>the server has already seen an ACK for it.

I thought that the ACK gives starting location and number of segments. If I
get 500 segments and ACK at the 50% point I am sending an ACK for the
Starting Point and 250 Segments (the unprocessed part would then ACK
Start+250 for 250 when done) Just as of I had only gotten 250 in the first
place and was also given the next 250 Segment Chunk (ie: I was "Next
Requester" after my original allocation of 250).

>IMO - a % ACK is to much complexity and extra work on the server,
>which is already having trouble keeping up.

No - It is the same load if you allow the first request to be twice the
size of the subsequent requests. If you ask people to request 30 minutes
worth of segments, there is no difference in load (if the Server responds
to each ACK when first attempted) if they start each run with a 1 hour
chunk (ie: 2X Chunk) and check in every 30 min to ACK a Chunk (and to get
the one to be worked on in a half an hour [and when you are 30 minutes away
from your shut down time, just ACK and do not request another chunk]) and
just getting a X sized chunk at your initial connection.

In the 2X method, you still have a X sized Chunk to work on for the next 30
Minutes if the Server is ignoring your ACK attempt (and when that Chunk has
been scanned you return both and get two more). This is hitting the Server
once every 30 minutes and NOT pounding away at it until you get an ACK
through (and more get more work) since you have no need for another chunk
immediately (as you would with the X sized Chunk every 30 minutes method)
and thus have no need to retry on a connect failure.