On 9/27/2019 12:56 PM, Sage Gerard wrote:
I got sloppy here in a Stripe integration: https://github.com/zyrolasting/stripe-integration/blob/master/main.rkt#L31

I'm not an InfoSec expert, but I know I'd like to secure the secret key used here in memory instead of using a parameter.

I'd probably encrypt the value provided by a client module and store it (write-only from the client's perspective) using set-box!. But I have several other questions:

1. Does the garbage collector keep a clear text copy of the secret in memory before I encrypt it? If so, how can I make it easy for a client module to set the secret key AND make it such that a garbage collection pass will remove the clear secret from RAM?

Obviously the plaintext must be kept until the crypttext exists.   I assume you are asking if the plaintext will persist AFTER encryption, and that answer is "yes".  The plaintext string will persist until the collection following the point where the program abandons it - but program can overwrite the string immediately after encrypting to erase the original value.  There still would be a small window of vulnerability, before/during encryption, but you can minimize it as much as possible.


2. Are there any existing /cross-platform/ Racket projects that can proactively keep secrets away from the garbage collector and swap space? Nothing relevant comes up for "secret", "security" or "swap" on the package index.

Sorry, no clue.


3. Are there any other intermediaries in a Racket process that might keep a copy of a cleartext secret?

Not 100% certain, but I don't think so ... not unless the program itself makes copies.


George

--
You received this message because you are subscribed to the Google Groups "Racket 
Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/racket-users/0f0f6530-7b4e-f093-1fc3-a57abe5936ba%40comcast.net.

Reply via email to