Re: [ftpext] COMB command IETF draft proposal

[Robert Oslin <[e-mail]>]

" So if the server is working sequntially it might give out some kind of information that helps seeing some progress, but it might be something completely different like a poem ;-)"

Agreed. Sending the 150 at regular intervals (and not dependent on some predetermined event, such as part1 combined), would work nicely.

A server NOOP if you will.

Robert


-----Original Message-----
From: Lothar Kimmeringer [mailto:[e-mail]] 
Sent: Thursday, June 09, 2011 4:51 PM
To: Robert Oslin
Cc: [e-mail]; [e-mail]
Subject: Re: [ftpext] COMB command IETF draft proposal

Am 09.06.2011 21:10, schrieb Robert Oslin:
> This assumes one method of recombining the file where each part is 
> sequentially
 > tacked on to the next, e.g. open, seek, write, open, seek, write, ad nauseam.
 > There are likely OS and File System specific techniques for combining multiple  > chunks (parts) into a single file that may only be able to provide an all or  > nothing "finished" vs. a partial success for each chunk appended in a sequential  > write operation. I'll provide more on that in a bit (currently writing my reply  > to the various comments)...

What the specific text after "150-" doesn't matter, it can also be 150-I'm still here, please wait or whatever you (the server) want. The only important thing is that something is given out at all that allows to reset the timeout on the client side.

So if the server is working sequntially it might give out some kind of information that helps seeing some progress, but it might be something completely different like a poem ;-)

In principle this technique is already in use, e.g. with "tarpits"
(http://en.wikipedia.org/wiki/Tarpit_%28networking%29).


Regards, Lothar
-- 
Lothar Kimmeringer                E-Mail: [e-mail]
                PGP-encrypted mails preferred (Key-ID: 0x8BC3CD81)

Always remember: The answer is forty-two, there can only be wrong
                  questions!