diff options
author | dormando <dormando@rydia.net> | 2017-05-29 22:51:20 -0700 |
---|---|---|
committer | dormando <dormando@rydia.net> | 2017-05-29 22:51:20 -0700 |
commit | d184b4b08f92d0171c8a3a3f03fa22e33c6aaa55 (patch) | |
tree | 146248ce6468a6c5c4a6e19b24816adb823c5aa0 /crawler.h | |
parent | 7d72a92cf11bbea0aa2b40e277eb9c2c44b8ad41 (diff) | |
download | memcached-d184b4b08f92d0171c8a3a3f03fa22e33c6aaa55.tar.gz |
LRU crawler scheduling improvements
when trying to manually run a crawl, the internal autocrawler is now blocked
from restarting for 60 seconds.
the internal autocrawl now independently schedules LRU's, and can re-schedule
sub-LRU's while others are still running. should allow much better memory
control when some sub-lru's (such as TEMP or WARM) are small, or slab classes
are differently sized.
this also makes the crawler drop its lock frequently.. this fixes an issue
where a long crawl happening at the same time as a hash table expansion could
hang the server until the crawl finished.
to improve still:
- elapsed time can be wrong in the logger entry
- need to cap number of entries scanned. enough set pressure and a crawl may
never finish.
Diffstat (limited to 'crawler.h')
-rw-r--r-- | crawler.h | 2 |
1 files changed, 1 insertions, 1 deletions
@@ -14,7 +14,7 @@ typedef struct { struct crawler_expired_data { pthread_mutex_t lock; - crawlerstats_t crawlerstats[MAX_NUMBER_OF_SLAB_CLASSES]; + crawlerstats_t crawlerstats[POWER_LARGEST]; /* redundant with crawlerstats_t so we can get overall start/stop/done */ rel_time_t start_time; rel_time_t end_time; |