Date: 2017-6-4

http://www.memcached.org/files/memcached-1.4.37.tar.gz

A good number of small community fixes (see below). Number of crash fixes and a 10 year old memory leak as well!

A new script (scripts/memcached-automove) is now included. This is a beta test of the slab page automove algorithm we will be moving to, and will be enabled by default when 1.5.0 lands soon. If you have time, please give it a test and send us feedback!

• LRU crawler: avoid running infinitely.
• fix very old memory leak in ASCII multigets. (when using multiget gets and keys after the first one are >255 characters)
• remove old slab mover example script.
• fix crash in page mover while using large items
• automover algo python script
• avoid segfault if idle_timeout value is missing.
• fix rare crash in LRU crawler
• sleep more aggressively in some threads
• don't overflow item refcount on get
• fix solaris compilation error
• usability fix for cache_memlimit command
• fix verbose print for idle-kicker
• disable refhang.t test due to flakiness
• fix ordering issue in conn dispatch (prevents potential hangups)

• LRU crawler scheduling improvements

The LRU crawler's internal run scheduler still predated the segmented LRU. It would run across each sub-LRU (HOT|WARM|etc), but schedule runs by slab class.

Now it schedules each sub-class independently. Now it will make more frequent targeted runs. IE: If you make use of TEMP and store many items that expire within a few minutes, it will wake up and re-run scans of each active TEMP LRU instead of timing based on the total size of the slab class.

It can also re-schedule runs while a previous run is still running. In the same example, it could continue to wake and scan TEMP LRU's, even if there's a COLD LRU with 300 million items in it which could take minutes to scan by itself.

• per-LRU hits breakdown under stats items (hits_to_(hot|warm|cold|temp))

The following people contributed to this release since 1.4.36.

Note that this is based on who contributed changes, not how they were done. In many cases, a code snippet on the mailing list or a bug report ended up as a commit with your name on it.

Note that this is just a summary of how many changes each person made which doesn't necessarily reflect how significant each change was. For details on what led up into a branch, either grab the git repo and look at the output of git log 1.4.36..1.4.37 or use a web view.

• Repo list: https://github.com/memcached/memcached/wiki/DevelopmentRepos
• Web View: http://github.com/memcached/memcached/commits/1.4.37

    17	dormando
     1	Fumihiro Ito
     1	Grant Mathews
     1	Josh Soref
     1	Peter (Stig) Edwards