[
https://issues.apache.org/jira/browse/PHOENIX-1457?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14380489#comment-14380489
]
ASF GitHub Bot commented on PHOENIX-1457:
-----------------------------------------
Github user jyates commented on a diff in the pull request:
https://github.com/apache/phoenix/pull/55#discussion_r27153366
--- Diff:
phoenix-core/src/main/java/org/apache/phoenix/hbase/index/ipc/PhoenixRpcSchedulerFactory.java
---
@@ -62,30 +62,59 @@ public RpcScheduler create(Configuration conf,
RegionServerServices services) {
return delegate;
}
+ // get the index priority configs
int indexHandlerCount =
conf.getInt(QueryServices.INDEX_HANDLER_COUNT_ATTRIB,
QueryServicesOptions.DEFAULT_INDEX_HANDLER_COUNT);
- int minPriority = getMinPriority(conf);
- int maxPriority =
conf.getInt(QueryServices.MAX_INDEX_PRIOIRTY_ATTRIB,
QueryServicesOptions.DEFAULT_INDEX_MAX_PRIORITY);
- // make sure the ranges are outside the warning ranges
- Preconditions.checkArgument(maxPriority > minPriority, "Max index
priority (" + maxPriority
- + ") must be larger than min priority (" + minPriority +
")");
- boolean allSmaller =
- minPriority < HConstants.REPLICATION_QOS
- && maxPriority < HConstants.REPLICATION_QOS;
- boolean allLarger = minPriority > HConstants.HIGH_QOS;
- Preconditions.checkArgument(allSmaller || allLarger, "Index
priority range (" + minPriority
- + ", " + maxPriority + ") must be outside HBase priority
range ("
- + HConstants.REPLICATION_QOS + ", " + HConstants.HIGH_QOS
+ ")");
+ int indexMinPriority = getIndexMinPriority(conf);
+ int indexMaxPriority =
conf.getInt(QueryServices.MAX_INDEX_PRIOIRTY_ATTRIB,
QueryServicesOptions.DEFAULT_INDEX_MAX_PRIORITY);
+ validatePriority(indexMinPriority, indexMaxPriority);
+
+ // get the metadata priority configs
+ int metadataHandlerCount =
conf.getInt(QueryServices.INDEX_HANDLER_COUNT_ATTRIB,
QueryServicesOptions.DEFAULT_METADATA_HANDLER_COUNT);
+ int metadataMinPriority = getMetadataMinPriority(conf);
+ int metadataMaxPriority =
conf.getInt(QueryServices.MAX_INDEX_PRIOIRTY_ATTRIB,
QueryServicesOptions.DEFAULT_METADATA_MAX_PRIORITY);
--- End diff --
I can see us wanting a reserved range for metadata updates - say 10
different numbers - so we don't have to worry about conflicting with anything
else that needs its own priority in the future. Its easy to get into
situations, especially if everyone is carefully watching these numbers and
fully know whats going on, to get into situations where you have META = 1,
SOME_FEATURE=2, OTHER_META=3, etc.
A reserved range is just safer for the future and is likely to never be
changed in the configuration, unless people already to priority changing, which
is unlikely given how hard it was for us :). However, having the configuration
is necessary, regardless of range or a single value, to ensure people can make
it work with their existing installations (and as proof against HBase invading
our range space by accident).
> Use high priority queue for metadata endpoint calls
> ---------------------------------------------------
>
> Key: PHOENIX-1457
> URL: https://issues.apache.org/jira/browse/PHOENIX-1457
> Project: Phoenix
> Issue Type: Bug
> Reporter: James Taylor
> Assignee: Thomas D'Silva
> Labels: 4.3.1
>
> If the RS hosting the system table gets swamped, then we'd be bottlenecked
> waiting for the response back before running a query when we check if the
> metadata is in sync. We should run endpoint coprocessor calls for
> MetaDataService at a high priority to avoid that.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
