Re: Real-World Impact of Hash DoS in Lua

[Roberto Ierusalimschy <roberto@...>]

> Well, if these needs for "randomization for increased security" are
> gaining weight, maybe a secondary lua_newstate taking a seed argument
> would solve the problem, that seed being used for all (present and
> future) randomizations needs.

That seems a bit overshooting. Unlike memory allocation, I do not see
any need for an application to use different seed-creation methods for
different states. And we still have the problem for the stand-alone
interpreter.

I still think this whole problem is somewhat overstated. As Jay Carlson
put it (Jan 20), there are many other forms of DoS attacks that most
sites are not resistant to. The main reasons we are changing Lua
are that we like this idea of long non-internalized strings and for
propaganda. For propaganda purposes, the stand-alone interpreter is
very important. If it is too easy to simmulate an attack in a standard
interpreter running in a decent system (e.g., Linux), you will have a
hard time to counter-argument with "but you can solve that by passing an
appropriate seed to lua_newstate".

-- Roberto