Martin Pool
mbp at samba.org
Thu May 30 11:08:15 EST 2002
On 29 May 2002, Dustin Mitchell <dustin at ywlcs.org> wrote: > And the cost will be zero, if AutoReload is turned off :-) Yes, that's the nice thing. > True, but there are valid reasons to want to test and develop such > opportunistic caching mechanisms (since anything opportunistic is by at least > a little bit nondeterministic, it probably needs the most testing). Yes. You can test them (without changing the code) perfectly well with this patch. If you're changing the caching code then by definition you need to restart one way or another. Indeed, having objects in memory where the same datastructure is modified by different versions of the code (with different bugs) sounds like a recipe for confusion. Suppose an earlier version of the cache code actually broke one of the data structure invariants on the cache -- you *want* to dump and reinitialize the thing. > BTW: I assume you're not implying that opportunistic caching is overly > fragile! It's a great technique! Relying on not being reloaded, however, is > obviously stupid, since the Apache API *will* reload you :-) Yes, that's what I meant: it's only dumb if you *rely* on persistence, rather than just taking advantage of it when it happens. Indeed an important part of testing that kind of code would be to set MaxRequestsPerChild to a low number and make sure that it still works when it is always being flushed. -- Martin Every Python user scoffed at it once, saying significant whitespace is the past. -- Moshe Zadka (?)
|