Hello Jeff:
[I deleted a half-typed-rant on this subject (and the related, growing
trend to think "virtualizing solves all my problems") before posting it... on
the theory that I was just being old and crotchety.... suffice it to say that
if you consider what is going to happen the day you have a billion requests
pending and you need to switch out you
interpreter/hardware/shared-libs/OS/whatever to one
that can't unpickle them anymore
because you aren't really sure of their data structure... (or, heaven forbid,
there is a subtle bug in your unpickle-and-upgrade code!). well...]
This topic has come up before. I would prefer to actually run tests to see what
the problems are? I need to look at pickling more in depth. Maybe it would help
if I read up on hot-swapping. I wonder if modulisation/layering and having an
interpreter dampen the effects of a change?
Currently my approach would be simple. Do a sanity check on the pickled
programme. If you can't properly depickle and continue, either the old
application or the new application, wakes up the tasklets, tells them something
bad has happened, clean-up and terminate. Sometimes unavoidable bad things
happen and cannot be mitigated.
Cheers,
Andrew
_______________________________________________
Stackless mailing list
[email protected]
http://www.stackless.com/mailman/listinfo/stackless
