On Tue, 20 Jun 2017, red 888 wrote:

> This doc says something about memcached I haven't read anywhere else:
> Which brings me to another point, items/chunks are not actively 
> reclaimed/expired. Memached does not
> have a background thread that explicitly expires items, reclaiming used 
> chunks for new items. When a
> slab needs a chunk and there are no more pages, Memcache will look at the 
> queue tail for items to
> evict. Memcache will make a best effort to evict expired items (items you've 
> explicitly set to expire
> after some time). In scenario 1, item 2, an expired item, is evicted. 
> However, in scenario 2, item 1,
> which has not yet expired, will be evicted, even though item 4 would seem 
> like the better candidate.
> But since item 4 is not near the tail, Memcached stops looking and just 
> expires item 1.
>
> So what is the memcache queue tail and why can't I find much info about it? 
> Seems pretty important
> because what he is saying is that memcache could evict a non expired item, if 
> it ran out of
> unassigned pages, had plenty of expire items for the slab classes, but 
> because an expired was nearer
> the "tail" (whatever that means) it evicted that one instead.

What is the problem you're trying to solve?

If you run the latest version and start it with "-o modern" option,
there's an LRU crawler that will reap expired items in the background.

-- 

--- 
You received this message because you are subscribed to the Google Groups 
"memcached" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to