On Tue, Sep 06, 2016 at 12:36:12PM -0700, Kevin J. McCarthy wrote: > On Mon, Sep 05, 2016 at 07:07:18PM -0500, Derek Martin wrote: > I need to think about this. I'm nervous about this approach, because > I'm not convinced all the possible string truncations are of sufficient > severity to warrant an abort. (e.g. some things may just be displayed > on the screen and are perfectly fine in truncated form). > > I would prefer a more measured approach. If there are places where we > lose data, then perhaps we should be allocating the string instead.
I understand your concern, but the problem is precisely that with the code the way it is, you'll never know if it's happening. That's why this bug is so (potentially) insidious. So, making this change is the first step to figuring that out, anyway. And as I said, it should be exceedingly rare that it could happen with legitimate mail (since SMTP basically limits the legal length of everything to 1K), so it's probably the case that it would only happen if a malicious sender is trying to do something funky--or if the sender's mailer is itself pretty broken. -- Derek D. Martin http://www.pizzashack.org/ GPG Key ID: 0xDFBEAD02 -=-=-=-=- This message is posted from an invalid address. Replying to it will result in undeliverable mail due to spam prevention. Sorry for the inconvenience.
pgp7UQCv79GgS.pgp
Description: PGP signature
