[
https://issues.apache.org/jira/browse/YARN-4851?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Li Lu updated YARN-4851:
------------------------
Attachment: YARN-4851-trunk.003.patch
Thanks [~hitesh]! I addressed most of the comments except the following items:
bq. updates to the timeline server docs for these metrics seems missing.
Sure. Let's include this part in YARN-4694 since this is something incremental
over ATS v1.5 documentation?
bq. should be there 2 levels of metrics - one group enabled by default and a
second group for more detailed monitoring to reduce load on the metrics system?
This is a very nice suggestion, but I feel a better way to fix is to improve
the existing metrics system instead of introducing special cases for ATS v1.5
only?
bq. at the overall ATS level, an overall avg latency across all reqs might be
useful for a general health check
This appears to be a general improvement to TimelineDataManager metrics for the
whole ATS. Shall we address this in a JIRA for ATS as a whole?
Other points all addresses in the latest patch. Now the metrics will be
published as:
{code}
name:
"Hadoop:service=ApplicationHistoryServer,name=EntityGroupFSTimelineStoreMetrics",
modelerType: "EntityGroupFSTimelineStoreMetrics",
tag.Context: "yarn",
tag.Hostname: "local",
GetEntityToSummaryOps: 0,
GetEntityToDetailOps: 0,
SummaryLogReadNumOps: 0,
SummaryLogReadAvgTimeMs: 0,
EntitiesReadToSummary: 0,
CacheHits: 0,
CacheStaleRefreshes: 0,
CacheEvicts: 0,
CacheRefreshNumOps: 0,
CacheRefreshAvgTimeMs: 0,
ActiveLogDirScanNumOps: 0,
ActiveLogDirScanAvgTimeMs: 0,
LogCleanNumOps: 0,
LogCleanAvgTimeMs: 0,
LogsDirsCleaned: 0
{code}
> Metric improvements for ATS v1.5 storage components
> ---------------------------------------------------
>
> Key: YARN-4851
> URL: https://issues.apache.org/jira/browse/YARN-4851
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: timelineserver
> Reporter: Li Lu
> Assignee: Li Lu
> Attachments: YARN-4851-trunk.001.patch, YARN-4851-trunk.002.patch,
> YARN-4851-trunk.003.patch
>
>
> We can add more metrics to the ATS v1.5 storage systems, including purging,
> cache hit/misses, read latency, etc.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
