Tisdag den 11:e juli 2017, klockan 14:45, skrev Chris Severance detta: > Try compiling with -fstack-protector-strong. If that's not enough you'll > need to add -fstack-protector-strong to your glibc. The easiest way to > do that is to install an Arch based distro and install inetutils-git > from the AUR with the patch removed. > > https://aur.archlinux.org/packages/inetutils-git/ > > All you're going to see is a crash which isn't very interesting. > > The patch works. I see sizeof is why just changing the buffer size did > not produce a useful result. > > On Tue, Jul 11, 2017, at 12:18 PM, Alfred M. Szmidt wrote: > > The right way to fix this is to not repeat the same gibberish code 7 > > times, and then having a sensible test that checks the output. Which > > is exactly why I asked a means to reproduce the problem. That is not > > at all provided by the original bug report! > >
This latter comment is correct. It is my intention to implement the following when time permits: The present code collects and presents binary data using string functions like snprints(), this is doomed for failure and is substandard. This Guillems Jovers patch, while correct in itself, does not the resolve the issue at hand. Like Alfred M. Szmidt I am not able, nor were I ever able, to detect the crash found by Chris Severance. Regards, M E Andersson
