cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Joel Knighton (JIRA)" <j...@apache.org>
Subject [jira] [Comment Edited] (CASSANDRA-10018) Stats for several pools removed from nodetool tpstats output
Date Mon, 30 Nov 2015 21:00:12 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-10018?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15032444#comment-15032444
] 

Joel Knighton edited comment on CASSANDRA-10018 at 11/30/15 8:59 PM:
---------------------------------------------------------------------

This regression was introduced because of a change in metrics. When we bumped the version
of metrics in [CASSANDRA-5657], changes in the metrics library changed the mapping between
metrics and JMX.

Because this broke the old way we were querying for threadpool information in JMX, we switched
to storing information related to the query with each {{Stage}}, as we needed to store some
information in able to get the correct metrics.

I've reworked this so that we directly query JMX for threadpools and then query the related
mBeans. This is a little more complicated than the old approach, but it means we know longer
rely on information associated with each {{Stage}}, so it works for all threadpools.

Due to minor formatting changes, the patch for 2.2 does not apply cleanly up.  I've made patches
to 2.2 and 3.0. The 3.0 patch merges cleanly up.

||branch||testall||dtest||
|[10018-2.2|https://github.com/jkni/cassandra/tree/10018-2.2]|[testall|http://cassci.datastax.com/view/Dev/view/jkni/job/jkni-10018-2.2-testall]|[dtest|http://cassci.datastax.com/view/Dev/view/jkni/job/jkni-10018-2.2-dtest]|
|[10018-3.0|https://github.com/jkni/cassandra/tree/10018-3.0]|[testall|http://cassci.datastax.com/view/Dev/view/jkni/job/jkni-10018-3.0-testall]|[dtest|http://cassci.datastax.com/view/Dev/view/jkni/job/jkni-10018-3.0-dtest]|

CI is clean relative to upstream.

EDIT: I should note that I've manually verified that all threadpools are visible after this
in tpstats and {{StatusLogger}}.


was (Author: jkni):
This regression was introduced because of a change in metrics. When we bumped the version
of metrics in [CASSANDRA-5657], changes in the metrics library changed the mapping between
metrics and JMX.

Because this broke the old way we were querying for threadpool information in JMX, we switched
to storing information related to the query with each {{Stage}}, as we needed to store some
information in able to get the correct metrics.

I've reworked this so that we directly query JMX for threadpools and then query the related
mBeans. This is a little more complicated than the old approach, but it means we know longer
rely on information associated with each {{Stage}}, so it works for all threadpools.

Due to minor formatting changes, the patch for 2.2 does not apply cleanly up.  I've made patches
to 2.2 and 3.0. The 3.0 patch merges cleanly up.

||branch||testall||dtest||
|[10018-2.2|https://github.com/jkni/cassandra/tree/10018-2.2]|[testall|http://cassci.datastax.com/view/Dev/view/jkni/job/jkni-10018-2.2-testall]|[dtest|http://cassci.datastax.com/view/Dev/view/jkni/job/jkni-10018-2.2-dtest]|
|[10018-3.0|https://github.com/jkni/cassandra/tree/10018-3.0]|[testall|http://cassci.datastax.com/view/Dev/view/jkni/job/jkni-10018-3.0-testall]|[dtest|http://cassci.datastax.com/view/Dev/view/jkni/job/jkni-10018-3.0-dtest]|

CI is clean relative to upstream.

> Stats for several pools removed from nodetool tpstats output 
> -------------------------------------------------------------
>
>                 Key: CASSANDRA-10018
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-10018
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Observability
>            Reporter: Sam Tunnicliffe
>            Assignee: Joel Knighton
>             Fix For: 2.2.x
>
>
> With CASSANDRA-5657, the output of nodetool tpstats changed to only include threadpool
info for actual Stages. There are a number of
> JMX enabled thread pool executors which we used to include in tpstats and that are still
in use but no longer show up.
> Before CASSANDRA-5657
> {noformat}
> Pool Name                    Active   Pending      Completed   Blocked  All time blocked
> MutationStage                     0         0              0         0              
  0
> ReadStage                         0         0              0         0              
  0
> RequestResponseStage              0         0              0         0              
  0
> ReadRepairStage                   0         0              0         0              
  0
> CounterMutationStage              0         0              0         0              
  0
> MiscStage                         0         0              0         0              
  0
> HintedHandoff                     0         1              0         0              
  0
> GossipStage                       0         0              0         0              
  0
> CacheCleanupExecutor              0         0              0         0              
  0
> InternalResponseStage             0         0              0         0              
  0
> CommitLogArchiver                 0         0              0         0              
  0
> CompactionExecutor                0         0             48         0              
  0
> ValidationExecutor                0         0              0         0              
  0
> MigrationStage                    0         0              2         0              
  0
> AntiEntropyStage                  0         0              0         0              
  0
> PendingRangeCalculator            0         0              1         0              
  0
> Sampler                           0         0              0         0              
  0
> MemtableFlushWriter               0         0             14         0              
  0
> MemtablePostFlush                 0         0             20         0              
  0
> MemtableReclaimMemory             0         0             14         0              
  0
> Message type           Dropped
> READ                         0
> RANGE_SLICE                  0
> _TRACE                       0
> MUTATION                     0
> COUNTER_MUTATION             0
> BINARY                       0
> REQUEST_RESPONSE             0
> PAGED_RANGE                  0
> READ_REPAIR                  0
> {noformat}
> After CASSANDRA-5657
> {noformat}
> Pool Name                    Active   Pending      Completed   Blocked  All time blocked
> ReadStage                         0         0              0         0              
  0
> MutationStage                     0         0              0         0              
  0
> CounterMutationStage              0         0              0         0              
  0
> GossipStage                       0         0              0         0              
  0
> RequestResponseStage              0         0              0         0              
  0
> AntiEntropyStage                  0         0              0         0              
  0
> MigrationStage                    0         0              2         0              
  0
> MiscStage                         0         0              0         0              
  0
> InternalResponseStage             0         0              0         0              
  0
> ReadRepairStage                   0         0              0         0              
  0
> Message type           Dropped
> READ                         0
> RANGE_SLICE                  0
> _TRACE                       0
> MUTATION                     0
> COUNTER_MUTATION             0
> BINARY                       0
> REQUEST_RESPONSE             0
> PAGED_RANGE                  0
> READ_REPAIR                  0
> {noformat}



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message