better logging when not allocating a shard due to required number of existing shards
This commit is contained in:
parent
574c455203
commit
2eb8278a80
|
@ -209,6 +209,9 @@ public class LocalGatewayNodeAllocation extends NodeAllocation {
|
|||
// we can't really allocate, so ignore it and continue
|
||||
unassignedIterator.remove();
|
||||
routingNodes.ignoredUnassigned().add(shard);
|
||||
if (logger.isDebugEnabled()) {
|
||||
logger.debug("[{}][{}]: not allocating, number_of_allocated_shards_found [{}], required_number [{}]", shard.index(), shard.id(), numberOfAllocationsFound, requiredAllocation);
|
||||
}
|
||||
continue;
|
||||
}
|
||||
|
||||
|
@ -216,7 +219,7 @@ public class LocalGatewayNodeAllocation extends NodeAllocation {
|
|||
// check if we need to throttle, NOTE, we don't check on NO since it does not apply
|
||||
// since this is our master data!
|
||||
if (nodeAllocations.canAllocate(shard, node, allocation) == NodeAllocation.Decision.THROTTLE) {
|
||||
if (logger.isTraceEnabled()) {
|
||||
if (logger.isDebugEnabled()) {
|
||||
logger.debug("[{}][{}]: throttling allocation [{}] to [{}] on primary allocation", shard.index(), shard.id(), shard, nodeWithHighestVersion);
|
||||
}
|
||||
// we are throttling this, but we have enough to allocate to this node, ignore it for now
|
||||
|
|
Loading…
Reference in New Issue