Henri Asseily <[EMAIL PROTECTED]> wrote: > The problem is not to know when a request is done processing. > The problem is killing requests that are processing for too long. > If you want kill them safely, you may not be able to kill them until > they're done, which defeats the purpose. > If you kill them "unsafely", then the Perl interpreter might be in a > dirty state, forcing you to thoroughly dispose of it if you want to > be 100% safe.
I could definately see this with things like SQLite... I havent tried this, but I would figure that with MySQL or Postgres or others that use a socket to communicate, closing the socket would be enough though... at worst, wouldn't the XS return to the perl code with a socket error? - Tyler