How a tools knows it's successfully decrypted a file?

www.EdmundKirwan.com <iamfractal@hotmail.com> wrote:
> My question is: how does a tool know that it's successfully decrypted a
> file?

By checksums.

Yours,
VB.
--
"Life was simple before World War II. After that, we had systems."
Grace Hopper

Hi, folks,

A newbie question (I couldn't find the right Googlism to hit a result,
nor did the FAQ shine a light).

I've used a couple of tools to encrypt and decrypt files for me based
on a password of my choice. When I try decrypting an encrypted file
with the wrong password, tools generally respectfully tell me that I've
made a tit of myself, and ask me to use the right password.

I hadn't thought much about this till the other day, using a new(ish)
tool to decrypt. I entered the wrong password (unknowingly, of course)
and it told me that the file was decrypted successfully; but a look at
the contents told me that this was not so. I think it just re-encrypted
the file using the, "New," password.

My question is: how does a tool know that it's successfully decrypted a
file?

I can think of a way of doing it, but it seems a little ... insecure:
the tool adds the password to the plaintext before encryption and then
encrypts it along with everything else. Then, after decryption, the
tool checks that the first X bytes match the password, and if so:
success.

Can anyone answer my question above?

Thanks for your time,

..ed

Volker Birk wrote:
> www.EdmundKirwan.com <iamfractal@hotmail.com> wrote:
> > My question is: how does a tool know that it's successfully decrypted a
> > file?
>
> By checksums.
>
> Yours,
> VB.
> --
> "Life was simple before World War II. After that, we had systems."
> Grace Hopper

Ah!

So if file novel.txt is being encrypted to produce novel.enc, then I
presume a checksum is generated for the plaintext and stored (for
example) as the first 8 bits of novel.enc, and then the encrypted data
is stored after the checksum. And when be de-crypted, the tool strips
away the first 8 bits of novel.enc before applying the decrption
algorithm to the rest.

Though if this storing of the checksum in the first 8 bits is not
standardised, then I suppose it's unlikely that a file encrypted with
blowfish using one tool could be decrypted by another tool using
blowfish: as one could store the checksum in the first 8 bits, and one
could store the checksum in the last 8 bits.

Thanks for the information,

..ed