[
https://issues.apache.org/jira/browse/HDFS-16488?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
tomscut updated HDFS-16488:
---------------------------
Description:
Currently, external SPS has no monitoring metrics. We do not know how many
blocks are waiting to be processed, how many blocks are waiting to be retried,
and how many blocks have been migrated.
We can expose these metrics in JMX for easy collection and display by
monitoring systems.
!image-2022-02-26-22-15-25-543.png|width=631,height=170!
For example, in our cluster, we exposed these metrics to JMX, collected by
JMX-Exporter and combined with Prometheus, and finally display by Grafana.
was:
Currently, external SPS has no monitoring metrics. We do not know how many
blocks are waiting to be processed, how many blocks are waiting to be retried,
and how many blocks have been migrated.
We can expose these metrics in JMX for easy collection and display by
monitoring systems.
!image-2022-02-26-22-15-09-432.png|width=593,height=160!
For example, in our cluster, we exposed these metrics to JMX, collected by
JMX-Exporter and combined with Prometheus, and finally display by Grafana.
> [SPS]: Expose metrics to JMX for external SPS
> ---------------------------------------------
>
> Key: HDFS-16488
> URL: https://issues.apache.org/jira/browse/HDFS-16488
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Reporter: tomscut
> Assignee: tomscut
> Priority: Major
> Labels: pull-request-available
> Attachments: image-2022-02-26-22-15-25-543.png
>
> Time Spent: 10m
> Remaining Estimate: 0h
>
> Currently, external SPS has no monitoring metrics. We do not know how many
> blocks are waiting to be processed, how many blocks are waiting to be
> retried, and how many blocks have been migrated.
> We can expose these metrics in JMX for easy collection and display by
> monitoring systems.
> !image-2022-02-26-22-15-25-543.png|width=631,height=170!
> For example, in our cluster, we exposed these metrics to JMX, collected by
> JMX-Exporter and combined with Prometheus, and finally display by Grafana.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
