[
https://issues.apache.org/jira/browse/PHOENIX-5545?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16974693#comment-16974693
]
Chinmay Kulkarni commented on PHOENIX-5545:
-------------------------------------------
[~larsh] [~tdsilva] Okay, v4 should solve the issue. When we get a connection
in the DropChildView task, we get a connection on the server so the test setup
should pass in "phoenix.schema.dropMetaData" to true in the server properties
as well.
On top of this, there is an issue where in the case of dropping a view, we
don't retrieve the dropMetaData property correctly (see
[this|https://github.com/apache/phoenix/blob/a55ba2a861303de3dd30edf369acdc1c873a2d59/phoenix-core/src/main/java/org/apache/phoenix/schema/MetaDataClient.java#L3180])
and it is always false.
The tests were flapping because of a race condition between the scheduled
dropChildViews task and the one initiated by the test via
[runDropChildViewsTask|https://github.com/apache/phoenix/blob/a55ba2a861303de3dd30edf369acdc1c873a2d59/phoenix-core/src/it/java/org/apache/phoenix/end2end/ViewMetadataIT.java#L281]
where we have the dropmetadata property set in 1 case and unset in the other.
> 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
>
> Attachments: PHOENIX-5545-4.x-HBase-1.3-v1.patch,
> PHOENIX-5545-4.x-HBase-1.3-v2.patch, PHOENIX-5545-4.x-HBase-1.3-v3.patch,
> PHOENIX-5545-4.x-HBase-1.3-v4.patch
>
> Time Spent: 50m
> Remaining Estimate: 0h
>
> 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)
