Details
-
Bug
-
Resolution: Fixed
-
1.6.0 beta3
-
None
-
Operating System: All
Platform: All
Description
I set the bucket size to 100 M
High and low water mark are set correctly.
Run memslap, filled up the server until I started getting "SERVER_ERROR a2b out of memory"
Checking the stats I don't see the mem_used goes down, it remains 100M. Running memslap after few minutes return errors immediately.
Checking after 10 min, mem_used is still 100M
Machine is 10.2.1.13
STAT ep_io_num_write 69614
STAT rejected_conns 0
STAT connection_structures 15
STAT limit_maxbytes 67108864
STAT decr_hits 0
STAT ep_pending_ops_max_duration 0 usec
STAT ep_too_young 0
STAT curr_connections 15
STAT rusage_system 14.455802
STAT ep_io_write_bytes 74763100
STAT ep_total_cache_size 79084270
STAT ep_storage_age 0
STAT ep_flush_duration_highwat 1
STAT ep_flush_duration 0
STAT cas_misses 0
STAT ep_flusher_todo 0
STAT ep_pending_ops 0
STAT mem_used 104857550
STAT get_misses 0
STAT ep_num_value_ejects 0
STAT ep_queue_size 0
STAT bytes_read 214860134
STAT get_hits 617310
STAT decr_misses 0
STAT rusage_user 11.377270
STAT bucket_conns 5
STAT ep_num_non_resident 0
STAT ep_tap_keepalive 0
STAT ep_oom_errors 74708
STAT ep_too_old 0
STAT cmd_flush 0
STAT ep_max_txn_size 50000
STAT ep_version 1.6.0beta3b_23_g2afb96f
STAT uptime 240
STAT ep_data_age_highwat 4
STAT ep_queue_age_cap 900
STAT incr_hits 0
STAT time 1282367323
STAT ep_total_persisted 68590
STAT daemon_connections 10
STAT ep_flusher_state running
STAT pointer_size 64
STAT version 1.4.4_266_ge579964
STAT ep_max_data_size 104857600