[
https://issues.apache.org/jira/browse/DRILL-3171?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15014474#comment-15014474
]
ASF GitHub Bot commented on DRILL-3171:
---------------------------------------
Github user hnfgns commented on a diff in the pull request:
https://github.com/apache/drill/pull/260#discussion_r45407026
--- Diff:
exec/java-exec/src/main/java/org/apache/drill/exec/store/sys/zk/ZkEStore.java
---
@@ -17,40 +17,25 @@
*/
package org.apache.drill.exec.store.sys.zk;
+import java.io.IOException;
+
import org.apache.curator.framework.CuratorFramework;
import org.apache.drill.exec.store.sys.EStore;
import org.apache.drill.exec.store.sys.PStoreConfig;
import org.apache.zookeeper.CreateMode;
-import java.io.IOException;
-
/**
* Implementation of EStore using Zookeeper's EPHEMERAL node.
* @param <V>
*/
-public class ZkEStore<V> extends ZkAbstractStore<V> implements EStore<V>{
+public class ZkEStore<V> extends ZkAbstractStore<V> implements EStore<V> {
- public ZkEStore(CuratorFramework framework, PStoreConfig<V> config)
throws IOException{
+ public ZkEStore(CuratorFramework framework, PStoreConfig<V> config)
throws IOException {
super(framework,config);
}
@Override
- public void delete(String key) {
- try {
- if (framework.checkExists().forPath(p(key)) != null) {
- framework.delete().forPath(p(key));
--- End diff --
This ensures that the cache is updated after deletion takes place.
> Storage Plugins : Two processes tried to update the storage plugin at the
> same time
> -----------------------------------------------------------------------------------
>
> Key: DRILL-3171
> URL: https://issues.apache.org/jira/browse/DRILL-3171
> Project: Apache Drill
> Issue Type: Bug
> Components: Storage - Information Schema
> Affects Versions: 1.0.0
> Reporter: Rahul Challapalli
> Assignee: Hanifi Gunes
> Labels: test
> Fix For: Future
>
>
> Commit Id# : bd8ac4fca03ad5043bca27fbc7e0dec5a35ac474
> We have seen this issue happen with the below steps
> 1. Clear out the zookeeper
> 2. Update the storage plugin using the rest API on one of the node
> 3. Submit 10 queries concurrently
> With randomized foreman node selection, the node executing the query might
> not have the updated storage plugins info. This could be causing the issue.
> - Rahul
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
