another one weirdness

Antony Dovgal
Fri, 9 Apr 2004 10:43:20 +0400

On Thu, 8 Apr 2004 11:15:17 -0700 (PDT)
Brad Fitzpatrick <> wrote:

> Avva,
> It's because the flush_all just sets the "flush time", and then a new item
> is stored in the same second, then a subsequent "get" finds the item and
> sees it was created <= flush_time, therefore ignores it.
> Which is arguably safer behavior.

Sounds reasonable, but still confusing.

Maybe it's worth to add some flag "this-record-can-be-safely-replaced" (that's what Avva just told, as I understand).
In this case flush_all will mark the records as obsolete, but subsequent replace/set will take away this mark.

IMHO making granularity higher you will encounter the same problem after a while, there is no chance to make it unachievable high..

> My knee-jerk answer:  it's a cache for christ's sake... you're supposed to
> deal with empty answers.  The current behavior is fine.

Yeah, I know, that I can't rely on the cache, but I just want to clear up these things.
> But the better answer:  higher granularity.

Yep, this answer is better =)
But is it really the best?

Antony Dovgal aka tony2001 ||