[jira] [Updated] (SOLR-10339) Implement set-trigger and remove-trigger APIs

2017-09-20 Thread Shalin Shekhar Mangar (JIRA)

 [ 
https://issues.apache.org/jira/browse/SOLR-10339?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Shalin Shekhar Mangar updated SOLR-10339:
-
Fix Version/s: (was: 7.0)
   7.1
   master (8.0)

> Implement set-trigger and remove-trigger APIs
> -
>
> Key: SOLR-10339
> URL: https://issues.apache.org/jira/browse/SOLR-10339
> Project: Solr
>  Issue Type: Sub-task
>  Security Level: Public(Default Security Level. Issues are Public) 
>  Components: SolrCloud
>Reporter: Shalin Shekhar Mangar
>Assignee: Shalin Shekhar Mangar
>  Labels: autoscaling
> Fix For: master (8.0), 7.1
>
> Attachments: SOLR-10339.patch, SOLR-10339.patch
>
>
> Implement set-trigger and remove-trigger API to add, update and remove 
> triggers for autoscaling.
> The following events are supported:
> # nodeAdded
> # nodeLost
> # replicaLost
> # schedule
> # searchRate
> # indexRate
> Each trigger has the following properties:
> # ‘name’ - a unique string to identify the trigger so that it can be read, 
> updated or removed later
> # ‘state’ - the state of the event (ENABLED or DISABLED), default is ENABLED. 
> This allows one to add a trigger which is disabled until a RESUME_TRIGGER API 
> is called.
> # ‘actions’ - a list of actions to be performed in the order specified. The 
> default list of actions for every trigger are to compute the plan, execute 
> the plan and save the plan. If an empty list of actions is explicitly 
> specified or null is specified when creating/updating the trigger then no 
> actions are performed at all.
> Here's an example of an API invocation:
> {code}
> {
>   "set-trigger" : {
>   "name" : "node_lost_trigger",
>   "event" : "nodeLost",
>   "waitFor" : "10m",
>   "state" : "ENABLED",
>   "actions" : [
>   {
>   "name" : "compute_plan",
>   "class" : "solr.ComputePlanAction"
>   },
>   {
>   "name" : "execute_plan",
>   "class" : "solr.ExecutePlanAction"
>   },
>   {
>   "name" : "log_plan",
>   "class" : "solr.LogPlanAction",
>   "collection" : ".system"
>   }
>   ]
>   }
> }
> {code}
> Note this issue is only about implementation of the user-facing APIs and not 
> the actual trigger mechanism itself.



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

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



[jira] [Updated] (SOLR-10339) Implement set-trigger and remove-trigger APIs

2017-03-24 Thread Shalin Shekhar Mangar (JIRA)

 [ 
https://issues.apache.org/jira/browse/SOLR-10339?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Shalin Shekhar Mangar updated SOLR-10339:
-
Attachment: SOLR-10339.patch

Added a marker interface for TriggerAction and empty stubs for ComputeAction, 
ExecuteAction and LogAction. These will be filled out in future issues. I'll 
commit this to the autoscaling branch.

> Implement set-trigger and remove-trigger APIs
> -
>
> Key: SOLR-10339
> URL: https://issues.apache.org/jira/browse/SOLR-10339
> Project: Solr
>  Issue Type: Sub-task
>  Security Level: Public(Default Security Level. Issues are Public) 
>  Components: SolrCloud
>Reporter: Shalin Shekhar Mangar
>Assignee: Shalin Shekhar Mangar
>  Labels: autoscaling
> Fix For: master (7.0)
>
> Attachments: SOLR-10339.patch, SOLR-10339.patch
>
>
> Implement set-trigger and remove-trigger API to add, update and remove 
> triggers for autoscaling.
> The following events are supported:
> # nodeAdded
> # nodeLost
> # replicaLost
> # schedule
> # searchRate
> # indexRate
> Each trigger has the following properties:
> # ‘name’ - a unique string to identify the trigger so that it can be read, 
> updated or removed later
> # ‘state’ - the state of the event (ENABLED or DISABLED), default is ENABLED. 
> This allows one to add a trigger which is disabled until a RESUME_TRIGGER API 
> is called.
> # ‘actions’ - a list of actions to be performed in the order specified. The 
> default list of actions for every trigger are to compute the plan, execute 
> the plan and save the plan. If an empty list of actions is explicitly 
> specified or null is specified when creating/updating the trigger then no 
> actions are performed at all.
> Here's an example of an API invocation:
> {code}
> {
>   "set-trigger" : {
>   "name" : "node_lost_trigger",
>   "event" : "nodeLost",
>   "waitFor" : "10m",
>   "state" : "ENABLED",
>   "actions" : [
>   {
>   "name" : "compute_plan",
>   "class" : "solr.ComputePlanAction"
>   },
>   {
>   "name" : "execute_plan",
>   "class" : "solr.ExecutePlanAction"
>   },
>   {
>   "name" : "log_plan",
>   "class" : "solr.LogPlanAction",
>   "collection" : ".system"
>   }
>   ]
>   }
> }
> {code}
> Note this issue is only about implementation of the user-facing APIs and not 
> the actual trigger mechanism itself.



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

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



[jira] [Updated] (SOLR-10339) Implement set-trigger and remove-trigger APIs

2017-03-22 Thread Shalin Shekhar Mangar (JIRA)

 [ 
https://issues.apache.org/jira/browse/SOLR-10339?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Shalin Shekhar Mangar updated SOLR-10339:
-
Attachment: SOLR-10339.patch

First cut for the new APIs. There are still a few nocommits.

> Implement set-trigger and remove-trigger APIs
> -
>
> Key: SOLR-10339
> URL: https://issues.apache.org/jira/browse/SOLR-10339
> Project: Solr
>  Issue Type: Sub-task
>  Security Level: Public(Default Security Level. Issues are Public) 
>  Components: SolrCloud
>Reporter: Shalin Shekhar Mangar
>Assignee: Shalin Shekhar Mangar
>  Labels: autoscaling
> Fix For: master (7.0)
>
> Attachments: SOLR-10339.patch
>
>
> Implement set-trigger and remove-trigger API to add, update and remove 
> triggers for autoscaling.
> The following events are supported:
> # nodeAdded
> # nodeLost
> # replicaLost
> # schedule
> # searchRate
> # indexRate
> Each trigger has the following properties:
> # ‘name’ - a unique string to identify the trigger so that it can be read, 
> updated or removed later
> # ‘state’ - the state of the event (ENABLED or DISABLED), default is ENABLED. 
> This allows one to add a trigger which is disabled until a RESUME_TRIGGER API 
> is called.
> # ‘actions’ - a list of actions to be performed in the order specified. The 
> default list of actions for every trigger are to compute the plan, execute 
> the plan and save the plan. If an empty list of actions is explicitly 
> specified or null is specified when creating/updating the trigger then no 
> actions are performed at all.
> Here's an example of an API invocation:
> {code}
> {
>   "set-trigger" : {
>   "name" : "node_lost_trigger",
>   "event" : "nodeLost",
>   "waitFor" : "10m",
>   "state" : "ENABLED",
>   "actions" : [
>   {
>   "name" : "compute_plan",
>   "class" : "solr.ComputePlanAction"
>   },
>   {
>   "name" : "execute_plan",
>   "class" : "solr.ExecutePlanAction"
>   },
>   {
>   "name" : "log_plan",
>   "class" : "solr.LogPlanAction",
>   "collection" : ".system"
>   }
>   ]
>   }
> }
> {code}
> Note this issue is only about implementation of the user-facing APIs and not 
> the actual trigger mechanism itself.



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

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



[jira] [Updated] (SOLR-10339) Implement set-trigger and remove-trigger APIs

2017-03-22 Thread Shalin Shekhar Mangar (JIRA)

 [ 
https://issues.apache.org/jira/browse/SOLR-10339?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Shalin Shekhar Mangar updated SOLR-10339:
-
Description: 
Implement set-trigger and remove-trigger API to add, update and remove triggers 
for autoscaling.

The following events are supported:
# nodeAdded
# nodeLost
# replicaLost
# schedule
# searchRate
# indexRate

Each trigger has the following properties:
# ‘name’ - a unique string to identify the trigger so that it can be read, 
updated or removed later
# ‘state’ - the state of the event (ENABLED or DISABLED), default is ENABLED. 
This allows one to add a trigger which is disabled until a RESUME_TRIGGER API 
is called.
# ‘actions’ - a list of actions to be performed in the order specified. The 
default list of actions for every trigger are to compute the plan, execute the 
plan and save the plan. If an empty list of actions is explicitly specified or 
null is specified when creating/updating the trigger then no actions are 
performed at all.

Here's an example of an API invocation:
{code}
{
"set-trigger" : {
"name" : "node_lost_trigger",
"event" : "nodeLost",
"waitFor" : "10m",
"state" : "ENABLED",
"actions" : [
{
"name" : "compute_plan",
"class" : "solr.ComputePlanAction"
},
{
"name" : "execute_plan",
"class" : "solr.ExecutePlanAction"
},
{
"name" : "log_plan",
"class" : "solr.LogPlanAction",
"collection" : ".system"
}
]
}
}
{code}

Note this issue is only about implementation of the user-facing APIs and not 
the actual trigger mechanism itself.

  was:Implement set-trigger and remove-trigger API to add, update and remove 
triggers for autoscaling.


> Implement set-trigger and remove-trigger APIs
> -
>
> Key: SOLR-10339
> URL: https://issues.apache.org/jira/browse/SOLR-10339
> Project: Solr
>  Issue Type: Sub-task
>  Security Level: Public(Default Security Level. Issues are Public) 
>  Components: SolrCloud
>Reporter: Shalin Shekhar Mangar
>Assignee: Shalin Shekhar Mangar
>  Labels: autoscaling
> Fix For: master (7.0)
>
>
> Implement set-trigger and remove-trigger API to add, update and remove 
> triggers for autoscaling.
> The following events are supported:
> # nodeAdded
> # nodeLost
> # replicaLost
> # schedule
> # searchRate
> # indexRate
> Each trigger has the following properties:
> # ‘name’ - a unique string to identify the trigger so that it can be read, 
> updated or removed later
> # ‘state’ - the state of the event (ENABLED or DISABLED), default is ENABLED. 
> This allows one to add a trigger which is disabled until a RESUME_TRIGGER API 
> is called.
> # ‘actions’ - a list of actions to be performed in the order specified. The 
> default list of actions for every trigger are to compute the plan, execute 
> the plan and save the plan. If an empty list of actions is explicitly 
> specified or null is specified when creating/updating the trigger then no 
> actions are performed at all.
> Here's an example of an API invocation:
> {code}
> {
>   "set-trigger" : {
>   "name" : "node_lost_trigger",
>   "event" : "nodeLost",
>   "waitFor" : "10m",
>   "state" : "ENABLED",
>   "actions" : [
>   {
>   "name" : "compute_plan",
>   "class" : "solr.ComputePlanAction"
>   },
>   {
>   "name" : "execute_plan",
>   "class" : "solr.ExecutePlanAction"
>   },
>   {
>   "name" : "log_plan",
>   "class" : "solr.LogPlanAction",
>   "collection" : ".system"
>   }
>   ]
>   }
> }
> {code}
> Note this issue is only about implementation of the user-facing APIs and not 
> the actual trigger mechanism itself.



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

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