I found that when I was defining a bug 309.
Here is the code:
import std.boxer;
import std.stdio;

interface A {
    public char[] message();
}
class B : A {
    char []info; 
    this(char []info){ this.info = info; }
    public char[] toString(){ return info; }
    public char[] message(){  return toString(); }
}

void main(char [][]args){
    A aa = new B( "I am A" );
    B bb = new B( "I am A too!");
    Box a = box( aa ), b = box( bb );

    if( unboxable!(A)( a ) ) writefln( "a is A! It says: "~a.toString() );
    else                     writefln( "a is not A! Despite it says: "~a.toString()  );
    
    if( unboxable!(A)( b ) ) writefln( "b is A! It says: "~b.toString() );
    else                     writefln( "b is not A! Despite it says: "~b.toString() );
}
Here is command lines:
> dmd -I/usr/local/dmd/src/phobos TestBoxError.d std/boxer.d
gcc TestBoxError.o boxer.o -o TestBoxError -m32 -lphobos -lpthread -lm 
Process dmd exited with code 0
> ./TestBoxError
Process TestBoxError exited with code 139

The reason is that interface A does not has toString() defined ?
The bug is connected with 309, I think.
It is not severe - workaround is obvious - especially taking in consideration 309.

GDC behaves differently:
> gdc -I/usr/local/dmd/src/phobos TestBoxError.d std/boxer.d
std/boxer.d:279: function std.format.doFormat (void delegate(dchar),TypeInfo[],char*,void*) does not match argument types (void delegate(dchar),TypeInfo[2],void[])
std/boxer.d:279: cannot implicitly convert expression (args) of type void[] to char*
Process gdc exited with code 1
But:
> gdc -I/usr/local/dmd/src/phobos TestBoxError.d 
Process gdc exited with code 0
>./a.out
Process a.out exited with code 139

What is confusing as well.

The bug is REALLY confusing, especially for Java programmers, because we know well, that EVERY object has toString.

I think that this bug can be relevant to  std.format.doFormat or to the way how
dmd handles interfaces.