I’ve just changed Hijack so that you can hijack any Ruby process - no
for your target process to require any code before it can be hijacked.
does this by first injecting a payload using gdb, then it signals the
process to start up a DRb server which the hijack client then connects
This should make it a lot more useful as most people probably wouldn’t
about using something like LiveConsole until a problem is known… at
point you have to enable LiveConsole in your app, restart it and then
for the problem to arise again.
The code is very new and fragile but it’d be great to get some feedback!
2009/7/11 Roger P. [email protected]