You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@phoenix.apache.org by "Chinmay Kulkarni (Jira)" <ji...@apache.org> on 2019/11/13 09:22:00 UTC

[jira] [Commented] (PHOENIX-5545) DropChildViews Task fails for a base table when its child view has an index

    [ https://issues.apache.org/jira/browse/PHOENIX-5545?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16973163#comment-16973163 ] 

Chinmay Kulkarni commented on PHOENIX-5545:
-------------------------------------------

When dropping a child view that has an index, if DROP_METADATA_ATTRIB is true, then we try to flush the physical HBase table (see [flushParentPhysicalTable|https://github.com/apache/phoenix/blob/4becb2b2aa95aa74f5339493d97ed9b3dd0c5396/phoenix-core/src/main/java/org/apache/phoenix/query/ConnectionQueryServicesImpl.java#L1901]) in order to avoid stale WAL edits of indexes. 

As per the 4.15 design of drop table cascade, since we drop child views at a later point, this flush fails with an _org.apache.hadoop.hbase.TableNotFoundException_. We should just swallow the exception in this case and not fail the DropChildViewsTask.
Please review the PR/patch [~kozdemir] [~larsh] [~tdsilva]




> DropChildViews Task fails for a base table when its child view has an index
> ---------------------------------------------------------------------------
>
>                 Key: PHOENIX-5545
>                 URL: https://issues.apache.org/jira/browse/PHOENIX-5545
>             Project: Phoenix
>          Issue Type: Bug
>    Affects Versions: 4.15.0, 5.1.0
>            Reporter: Chinmay Kulkarni
>            Assignee: Chinmay Kulkarni
>            Priority: Major
>             Fix For: 4.15.0, 5.1.0
>
>
> 1) Start the HBase server with 4.15.0 Phoenix
> 2) Connect with a 4.15.0 client
> 3) Create a base table and a view on top of the base table and create an index on top of this view:
> {code:sql}
> CREATE TABLE IF NOT EXISTS Z_BASE_TABLE (ID INTEGER NOT NULL PRIMARY KEY, HOST VARCHAR(10), FLAG BOOLEAN);
> CREATE VIEW Z_VIEW1 (col1 INTEGER, col2 INTEGER, col3 INTEGER, col4 INTEGER, col5 INTEGER) AS SELECT * FROM Z_BASE_TABLE WHERE ID>10;
> CREATE INDEX V_I ON Z_VIEW1(COL4);
> {code}
> 4) Drop the base table with the cascade option:
> {code:sql}
> DROP TABLE Z_BASE_TABLE CASCADE;
> {code}
> 5) DropChildViewsTask gets added to the SYSTEM.TASK table as expected, but when it gets scheduled and subsequently executed, it fails with the following exception:
> {code:java}
> 2019-10-24 22:44:34,214 ERROR [pool-121-thread-1] tasks.DropChildViewsTask: Exception while dropping a child view task. null.Z_BASE_TABLE with tenant id  IS NULL and timestamp 292278994-08-16 23:12:55.807
> org.apache.phoenix.exception.PhoenixIOException: Z_BASE_TABLE
> 	at org.apache.phoenix.query.ConnectionQueryServicesImpl.flushTable(ConnectionQueryServicesImpl.java:4442)
> 	at org.apache.phoenix.query.ConnectionQueryServicesImpl.flushParentPhysicalTable(ConnectionQueryServicesImpl.java:1922)
> 	at org.apache.phoenix.query.ConnectionQueryServicesImpl.dropTable(ConnectionQueryServicesImpl.java:1888)
> 	at org.apache.phoenix.schema.MetaDataClient.dropTable(MetaDataClient.java:3152)
> 	at org.apache.phoenix.schema.MetaDataClient.dropTable(MetaDataClient.java:3060)
> 	at org.apache.phoenix.util.ViewUtil.dropChildViews(ViewUtil.java:207)
> 	at org.apache.phoenix.coprocessor.tasks.DropChildViewsTask.run(DropChildViewsTask.java:60)
> 	at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
> 	at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
> 	at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
> 	at java.lang.reflect.Method.invoke(Method.java:498)
> 	at org.apache.phoenix.coprocessor.TaskRegionObserver$SelfHealingTask.run(TaskRegionObserver.java:204)
> 	at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
> 	at java.util.concurrent.FutureTask.runAndReset(FutureTask.java:308)
> 	at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$301(ScheduledThreadPoolExecutor.java:180)
> 	at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:294)
> 	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> 	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> 	at java.lang.Thread.run(Thread.java:748)
> Caused by: org.apache.hadoop.hbase.TableNotFoundException: Z_BASE_TABLE
> 	at org.apache.hadoop.hbase.client.HBaseAdmin.checkTableExists(HBaseAdmin.java:2863)
> 	at org.apache.hadoop.hbase.client.HBaseAdmin.flush(HBaseAdmin.java:1904)
> 	at org.apache.hadoop.hbase.client.HBaseAdmin.flush(HBaseAdmin.java:1949)
> 	at org.apache.phoenix.query.ConnectionQueryServicesImpl.flushTable(ConnectionQueryServicesImpl.java:4440)
> 	... 18 more
> {code}
> Note that this marks the DropChildViews task for the base table as failed and so *if the base table has any other views, those views will also not be dropped*.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)