Module org.elasticsearch.server

Package org.elasticsearch.cluster.routing.allocation.allocator

Class UndesiredAllocationsTracker

java.lang.Object

org.elasticsearch.cluster.routing.allocation.allocator.UndesiredAllocationsTracker

public class UndesiredAllocationsTracker
extends Object

Keeps track of a limited number of shards that are currently in undesired allocations. If the shards remain in undesired allocations for longer than a configurable threshold, it will log why those shards can't be allocated to desired nodes.

Field Summary

Fields

Modifier and Type	Field	Description
static final Setting<Integer>	MAX_UNDESIRED_ALLOCATIONS_TO_TRACK	The maximum number of undesired allocations to track.
static final Setting<TimeValue>	UNDESIRED_ALLOCATION_DURATION_LOG_INTERVAL_SETTING	The minimum amount of time between warnings about persistent undesired allocations
static final Setting<TimeValue>	UNDESIRED_ALLOCATION_DURATION_LOG_THRESHOLD_SETTING	Warning logs will be periodically written if we see a shard that's been in an undesired allocation for this long

Method Summary

Modifier and Type	Method	Description
void	cleanup(RoutingNodes routingNodes)	Clear any ShardRouting that are no longer present in the routing nodes
void	clear()	Clear all tracked allocations
Releasable	disableMissingAllocationAssertions()
void	maybeLogUndesiredShardsWarning(RoutingNodes routingNodes, RoutingAllocation routingAllocation, DesiredBalance desiredBalance)	If there are shards that have been in undesired allocations for longer than the configured threshold, log a warning
void	removeTracking(ShardRouting shardRouting)	Remove any tracking of the specified allocation (a no-op if the allocation isn't being tracked)
void	trackUndesiredAllocation(ShardRouting shardRouting)	Track an allocation as being undesired

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Details

UNDESIRED_ALLOCATION_DURATION_LOG_THRESHOLD_SETTING

public static final Setting<TimeValue> UNDESIRED_ALLOCATION_DURATION_LOG_THRESHOLD_SETTING

Warning logs will be periodically written if we see a shard that's been in an undesired allocation for this long

UNDESIRED_ALLOCATION_DURATION_LOG_INTERVAL_SETTING

public static final Setting<TimeValue> UNDESIRED_ALLOCATION_DURATION_LOG_INTERVAL_SETTING

The minimum amount of time between warnings about persistent undesired allocations

MAX_UNDESIRED_ALLOCATIONS_TO_TRACK

public static final Setting<Integer> MAX_UNDESIRED_ALLOCATIONS_TO_TRACK

The maximum number of undesired allocations to track. We expect this to be relatively small.

Method Details

trackUndesiredAllocation

public void trackUndesiredAllocation(ShardRouting shardRouting)

Track an allocation as being undesired

removeTracking

public void removeTracking(ShardRouting shardRouting)

Remove any tracking of the specified allocation (a no-op if the allocation isn't being tracked)

cleanup

public void cleanup(RoutingNodes routingNodes)

Clear any ShardRouting that are no longer present in the routing nodes

clear

public void clear()

Clear all tracked allocations

maybeLogUndesiredShardsWarning

public void maybeLogUndesiredShardsWarning(RoutingNodes routingNodes,
 RoutingAllocation routingAllocation,
 DesiredBalance desiredBalance)

If there are shards that have been in undesired allocations for longer than the configured threshold, log a warning

disableMissingAllocationAssertions

public Releasable disableMissingAllocationAssertions()