[jira] [Updated] (SOLR-11201) Implement trigger for arbitrary metrics

["Shalin Shekhar Mangar (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/SOLR-11201?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Shalin Shekhar Mangar updated SOLR-11201:
-----------------------------------------
    Attachment: SOLR-11201.patch

This patch adds a MetricTrigger that creates MetricBreachedEvents. The intention is to watch container level metrics and perform move replica actions if they are breached. But any arbitrary metric (even core level) can be used as well. It supports "below" and "above" threshold values and can limit operations to a specific collection, shard or a single node.

Example set-trigger call that fires the trigger whenever the total usable space on a node having replicas of "mycollection" falls below 100GB. The computed plan will then move replicas of mycollection away from such nodes.
{code}
{
  "set-trigger": {
    "name": "metric_trigger",
    "event": "metric",
    "waitFor": "5s",
    "metric": "metric:solr.node:CONTAINER.fs.coreRoot.usableSpace"
    "below": 107374182400,
    "collection": "mycollection",
    "shard": "shard1"    
  }
}
{code}

> Implement trigger for arbitrary metrics
> ---------------------------------------
>
>                 Key: SOLR-11201
>                 URL: https://issues.apache.org/jira/browse/SOLR-11201
>             Project: Solr
>          Issue Type: Sub-task
>      Security Level: Public(Default Security Level. Issues are Public) 
>          Components: AutoScaling, SolrCloud
>            Reporter: Shalin Shekhar Mangar
>            Assignee: Shalin Shekhar Mangar
>             Fix For: 7.2
>
>         Attachments: SOLR-11201.patch
>
>
> It should be possible to set a trigger on any metrics exposed by the Metrics API using a threshold value. Supporting {{waitFor}} may not be possible or useful for all metrics. For those we will implement proper trigger support (such as searchRate) However, a naive implementation might be to just poll the value of the metric frequently and if it is consistently above the threshold, fire the trigger.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org