Hi
On Thu, Oct 2, 2014 at 11:38 AM, Miloslav Trmač wrote:
OK, then; care to explicitly list the advantages you expect to see from such a switch, and why they outweigh the disadvantages and the migration costs?
I don't have a predrawn conclusion that I am advocating strongly for here but I am convinced it is worth a discussion. So far from the discussions I have seen the advantages as reduced size ( dash is significantly smaller than bash) and this can be a nice advantage for containers etc, POSIX correctness (/bin/sh should be only used by shell scripts that don't rely on Bash specific features), performance (Zdenek Kabelac posted some numbers)
On the flip side, there is probably some shell scripts that needs to be fixed to use /bin/bash explicitly as opposed to assuming /bin/sh will always be a symlink to bash. I can help write guidelines, check scripts, file bug reports etc but even if we do fix the scripts we include within the distribution (likely minimal since Debian/Ubuntu has switched over several years back), users might have to fix their scripts which we don't control. Any user ability to reconfigure the default system shell has some added complexity at the packaging level. Dash also might need a security review from someone competent to do it (ie) not me.
On the flip side, there is probably some shell scripts that needs to be fixed to use /bin/bash explicitly as opposed to assuming /bin/sh will always be a symlink to bash. I can help write guidelines, check scripts, file bug reports etc but even if we do fix the scripts we include within the distribution (likely minimal since Debian/Ubuntu has switched over several years back), users might have to fix their scripts which we don't control. Any user ability to reconfigure the default system shell has some added complexity at the packaging level. Dash also might need a security review from someone competent to do it (ie) not me.
The expected security improvement is essentially nonexistent. In the current case of importing functions from the environment (and we could have a looong philosophical conversation about whether this is a vulnerability in bash or in its callers, where the likely outcome is “not a vulnerability in bash but by far easiest to fix in bash”)
Why would this be a philosophical discussion when there were clearly bugs in the parser allowing things it shouldn't even if you consider the use cases valid otherwise?
Rahul
-- devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
