Hello all,
I'm using Memcached version 1.4.4 and built a script to measure the
"wasted bytes" from the "stats slabs" information and draw it in Cacti.
But since a few days one of the slabs tells me this:
"stats slabs"
[...]
STAT 34:chunk_size 161720
STAT 34:chunks_per_page 6
STAT 34:total_pages 15
STAT 34:total_chunks 90
STAT 34:used_chunks 66
STAT 34:free_chunks 19
STAT 34:free_chunks_end 5
STAT 34:mem_requested 18446744073708916198
STAT 34:get_hits 265415
STAT 34:cmd_set 135045
STAT 34:delete_hits 52
STAT 34:incr_hits 0
STAT 34:decr_hits 0
STAT 34:cas_hits 0
STAT 34:cas_badval 0
[...]
"stats items"
[...]
STAT items:34:number 66
STAT items:34:age 1667880
STAT items:34:evicted 0
STAT items:34:evicted_nonzero 0
STAT items:34:evicted_time 0
STAT items:34:outofmemory 0
STAT items:34:tailrepairs 0
[...]
I calculate the "wasted bytes" as described in protocol.txt like this:
(total_chunks * chunk_size) - mem_requested
(90 * 161720) - 18446744073708916198 =
-18446744073694361398
^^^^^^^^^^^^^^^^^^^^^
And this seems to be a bit unlikely. Is this some kind of bug?
If you need any more information please let me know. Any help would be
much appreciated.
Thanks and regards,
Dieter
