[CASSANDRA-599] The pending tasks listed for the compaction pool is confusing

• Type: Improvement
• Status: Resolved
• Priority: Low
• Resolution: Fixed
• Fix Version/s: 0.6
• Component/s: Legacy/Tools

Environment

RHEL 4

• Assignee: Jonathan Ellis
• Reporter: Tim Freeman

Authors

Jonathan Ellis

• Created: 03/Dec/09 22:45
• Updated: 16/Apr/19 09:33
• Resolved: 23/Dec/09 19:21

Description

Using the build dated 29 Nov 09, the output from "nodeprobe tpstats" is confusing. The following observations were made with an idle server that was compacting to clean up after some writes that were done in the past. The number of pending tasks for COMPACTION-POOL starts at 1673, and after doing some work the number of pending tasks has increased to 1675. I'd prefer that the amount of pending work reported by tpstats decreases as work is done.

> sh nodeprobe.sh tpstats
Pool Name Active Pending Completed
FILEUTILS-DELETE-POOL 0 0 872
MESSAGING-SERVICE-POOL 0 0 0
STREAM-STAGE 0 0 0
RESPONSE-STAGE 0 0 662309
ROW-READ-STAGE 0 0 662309
LB-OPERATIONS 0 0 0
COMMITLOG 1 0 1189659
MESSAGE-DESERIALIZER-POOL 0 0 0
GMFD 0 0 0
LB-TARGET 0 0 0
CONSISTENCY-MANAGER 0 0 0
ROW-MUTATION-STAGE 0 0 1101916
MESSAGE-STREAMING-POOL 0 0 0
LOAD-BALANCER-STAGE 0 0 0
FLUSH-SORTER-POOL 0 0 1723
MEMTABLE-POST-FLUSHER 0 0 1723
COMPACTION-POOL 1 1673 136
FLUSH-WRITER-POOL 0 0 1723
HINTED-HANDOFF-POOL 0 0 20
> sh nodeprobe.sh tpstats
Pool Name Active Pending Completed
FILEUTILS-DELETE-POOL 0 0 872
MESSAGING-SERVICE-POOL 0 0 0
STREAM-STAGE 0 0 0
RESPONSE-STAGE 0 0 662309
ROW-READ-STAGE 0 0 662309
LB-OPERATIONS 0 0 0
COMMITLOG 1 0 1191173
MESSAGE-DESERIALIZER-POOL 0 0 0
GMFD 0 0 0
LB-TARGET 0 0 0
CONSISTENCY-MANAGER 0 0 0
ROW-MUTATION-STAGE 0 0 1101916
MESSAGE-STREAMING-POOL 0 0 0
LOAD-BALANCER-STAGE 0 0 0
FLUSH-SORTER-POOL 0 0 1723
MEMTABLE-POST-FLUSHER 0 0 1723
COMPACTION-POOL 1 1675 139
FLUSH-WRITER-POOL 0 0 1723
HINTED-HANDOFF-POOL 0 0 20
(^{[[01;32mroot@localhost}[[00m)^{[[01;34m~/rackspace-stuff/cassandra-0.5-pre}[[00m>