[
https://issues.apache.org/jira/browse/HBASE-27617?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17692775#comment-17692775
]
Bryan Beaudreault commented on HBASE-27617:
-------------------------------------------
Hi [~tangtianhang], sorry. My main concern is whether the complexity is worth
it here. To me this seems like it could be worked around in the client
dashboards. At my company we also use these metrics, but we don't have the
problems you say.
TableLatencies metric has counters and histograms. The histograms are reset
after each query. So if a table stops getting requests, it should show 0's.
The counters don't get reset, but they stop incrementing. Counters are usually
turned into Rates, and the Rate would be 0/sec when the counter stops
incrementing.
So once a table moves off a server, all of the table metrics should quickly
become 0. They aren't removed, but should be easily filterable in dashboards.
> Make metrics from MetricsTableLatenciesImpl removable
> -----------------------------------------------------
>
> Key: HBASE-27617
> URL: https://issues.apache.org/jira/browse/HBASE-27617
> Project: HBase
> Issue Type: Bug
> Reporter: tianhang tang
> Assignee: tianhang tang
> Priority: Major
>
> Now the table latency metrics will last forever on a RS.
> Could reproduce this issue easily:
> 1. Create a test table and the region opened on RS_A;
> 2. Do a query;
> 3. Move the table to another group;
> 4. See JMX on RS_A, table latency metrics still exists and will not deleted.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
