We are using a shared Linux account to run database scripts and CGI programs. I thought of the following method of hiding the $password in a script, but was not able to google its viability:
1. $password is obfuscated in a compiled C program that breaks up the password so it won't be visible via "strings".
2. The C program will only return the password to registered calling programs or scripts - and checks the registered inode value to ensure it was not altered.
Some code fragments follow to show how some of this information is gathered from C:
parentpid = (int) getppid();
sprintf(a,"cat /proc/%d/cmdline;echo", parentpid);
sprintf(b,"cat /proc/%d/comm;echo", parentpid);
Most of the posts I've seen say it is impossible to achieve unbreakable security in this respect. However, I was wondering if anyone has tried this technique or similar for Perl scripts?