This is an automated email from the ASF dual-hosted git repository.
lijibing pushed a commit to branch branch-2.1
in repository https://gitbox.apache.org/repos/asf/doris.git
The following commit(s) were added to refs/heads/branch-2.1 by this push:
new c176c9feb02 [fix](statistics)Fix empty table keep auto analyze bug.
(#40811) (#40896)
c176c9feb02 is described below
commit c176c9feb02ec26bb36c01c3fce571ff83051e58
Author: Jibing-Li <[email protected]>
AuthorDate: Wed Sep 18 12:51:44 2024 +0800
[fix](statistics)Fix empty table keep auto analyze bug. (#40811) (#40896)
backport: https://github.com/apache/doris/pull/40811
---
.../doris/statistics/AnalysisInfoBuilder.java | 2 +-
.../apache/doris/statistics/TableStatsMeta.java | 16 +++++---
.../doris/statistics/util/StatisticsUtil.java | 28 ++++++-------
.../statistics/test_drop_stats_and_truncate.groovy | 48 ++++++++++++++++++++++
4 files changed, 73 insertions(+), 21 deletions(-)
diff --git
a/fe/fe-core/src/main/java/org/apache/doris/statistics/AnalysisInfoBuilder.java
b/fe/fe-core/src/main/java/org/apache/doris/statistics/AnalysisInfoBuilder.java
index 6541027538a..2e7c4078ca1 100644
---
a/fe/fe-core/src/main/java/org/apache/doris/statistics/AnalysisInfoBuilder.java
+++
b/fe/fe-core/src/main/java/org/apache/doris/statistics/AnalysisInfoBuilder.java
@@ -63,7 +63,7 @@ public class AnalysisInfoBuilder {
private boolean usingSqlForPartitionColumn;
private long tblUpdateTime;
private boolean emptyJob;
- private boolean userInject;
+ private boolean userInject = false;
private long rowCount;
public AnalysisInfoBuilder() {
diff --git
a/fe/fe-core/src/main/java/org/apache/doris/statistics/TableStatsMeta.java
b/fe/fe-core/src/main/java/org/apache/doris/statistics/TableStatsMeta.java
index f2af3e88274..ca918c950bb 100644
--- a/fe/fe-core/src/main/java/org/apache/doris/statistics/TableStatsMeta.java
+++ b/fe/fe-core/src/main/java/org/apache/doris/statistics/TableStatsMeta.java
@@ -26,6 +26,7 @@ import org.apache.doris.common.io.Writable;
import org.apache.doris.persist.gson.GsonPostProcessable;
import org.apache.doris.persist.gson.GsonUtils;
import org.apache.doris.statistics.AnalysisInfo.JobType;
+import org.apache.doris.statistics.util.StatisticsUtil;
import com.google.common.annotations.VisibleForTesting;
import com.google.gson.annotations.SerializedName;
@@ -159,7 +160,9 @@ public class TableStatsMeta implements Writable,
GsonPostProcessable {
public void update(AnalysisInfo analyzedJob, TableIf tableIf) {
updatedTime = analyzedJob.tblUpdateTime;
- userInjected = analyzedJob.userInject;
+ if (analyzedJob.userInject) {
+ userInjected = true;
+ }
for (Pair<String, String> colPair : analyzedJob.jobColumns) {
ColStatsMeta colStatsMeta = colToColStatsMeta.get(colPair);
if (colStatsMeta == null) {
@@ -179,15 +182,18 @@ public class TableStatsMeta implements Writable,
GsonPostProcessable {
indexesRowCount.putAll(analyzedJob.indexesRowCount);
clearStaleIndexRowCount((OlapTable) tableIf);
}
- if (analyzedJob.emptyJob) {
- return;
- }
if (analyzedJob.jobColumns.containsAll(
tableIf.getColumnIndexPairs(
-
tableIf.getSchemaAllIndexes(false).stream().map(Column::getName).collect(Collectors.toSet()))))
{
+ tableIf.getSchemaAllIndexes(false).stream()
+ .filter(c ->
!StatisticsUtil.isUnsupportedType(c.getType()))
+
.map(Column::getName).collect(Collectors.toSet())))) {
updatedRows.set(0);
newPartitionLoaded.set(false);
}
+ // Set userInject back to false after manual analyze.
+ if (JobType.MANUAL.equals(jobType) && !analyzedJob.userInject) {
+ userInjected = false;
+ }
}
}
diff --git
a/fe/fe-core/src/main/java/org/apache/doris/statistics/util/StatisticsUtil.java
b/fe/fe-core/src/main/java/org/apache/doris/statistics/util/StatisticsUtil.java
index c2c7de48a9c..0794b356595 100644
---
a/fe/fe-core/src/main/java/org/apache/doris/statistics/util/StatisticsUtil.java
+++
b/fe/fe-core/src/main/java/org/apache/doris/statistics/util/StatisticsUtil.java
@@ -43,7 +43,6 @@ import org.apache.doris.catalog.Partition;
import org.apache.doris.catalog.PrimitiveType;
import org.apache.doris.catalog.ScalarType;
import org.apache.doris.catalog.StructType;
-import org.apache.doris.catalog.TableAttributes;
import org.apache.doris.catalog.TableIf;
import org.apache.doris.catalog.Type;
import org.apache.doris.catalog.VariantType;
@@ -878,27 +877,26 @@ public class StatisticsUtil {
}
public static boolean isEmptyTable(TableIf table,
AnalysisInfo.AnalysisMethod method) {
- int waitRowCountReportedTime = 90;
+ int waitRowCountReportedTime = 120;
if (!(table instanceof OlapTable) ||
method.equals(AnalysisInfo.AnalysisMethod.FULL)) {
return false;
}
OlapTable olapTable = (OlapTable) table;
+ long rowCount = 0;
for (int i = 0; i < waitRowCountReportedTime; i++) {
- if (olapTable.getRowCount() > 0) {
- return false;
- }
- // If visible version is 2, table is probably not empty. So we
wait row count to be reported.
- // If visible version is not 2 and getRowCount return 0, we assume
it is an empty table.
- if (olapTable.getVisibleVersion() !=
TableAttributes.TABLE_INIT_VERSION + 1) {
- return true;
- }
- try {
- Thread.sleep(1000);
- } catch (InterruptedException e) {
- LOG.info("Sleep interrupted.", e);
+ rowCount =
olapTable.getRowCountForIndex(olapTable.getBaseIndexId(), true);
+ // rowCount == -1 means new table or first load row count not
fully reported, need to wait.
+ if (rowCount == -1) {
+ try {
+ Thread.sleep(1000);
+ } catch (InterruptedException e) {
+ LOG.info("Sleep interrupted.");
+ }
+ continue;
}
+ break;
}
- return true;
+ return rowCount == 0;
}
}
diff --git
a/regression-test/suites/statistics/test_drop_stats_and_truncate.groovy
b/regression-test/suites/statistics/test_drop_stats_and_truncate.groovy
index 7eb13e9a855..969e03cb295 100644
--- a/regression-test/suites/statistics/test_drop_stats_and_truncate.groovy
+++ b/regression-test/suites/statistics/test_drop_stats_and_truncate.groovy
@@ -174,6 +174,54 @@ suite("test_drop_stats_and_truncate") {
columns = all_columns.split(",");
assertEquals(9, columns.size())
+ sql """drop table part"""
+ sql """CREATE TABLE `part` (
+ `id` INT NULL,
+ `colint` INT NULL,
+ `coltinyint` tinyint NULL,
+ `colsmallint` smallINT NULL,
+ `colbigint` bigINT NULL,
+ `collargeint` largeINT NULL,
+ `colfloat` float NULL,
+ `coldouble` double NULL,
+ `coldecimal` decimal(27, 9) NULL
+ ) ENGINE=OLAP
+ DUPLICATE KEY(`id`)
+ COMMENT 'OLAP'
+ PARTITION BY RANGE(`id`)
+ (
+ PARTITION p1 VALUES [("-2147483648"), ("10000")),
+ PARTITION p2 VALUES [("10000"), ("20000")),
+ PARTITION p3 VALUES [("20000"), ("30000"))
+ )
+ DISTRIBUTED BY HASH(`id`) BUCKETS 3
+ PROPERTIES (
+ "replication_allocation" = "tag.location.default: 1"
+ )
+ """
+ sql """analyze table part with sync"""
+ sql """Insert into part values (1, 1, 1, 1, 1, 1, 1.1, 1.1, 1.1)"""
+ result = sql """show table stats part"""
+ assertEquals("true", result[0][6])
+ sql """truncate table part partition(p1)"""
+ result = sql """show table stats part"""
+ assertEquals("true", result[0][6])
+ sql """analyze table part with sample rows 100 with sync"""
+ result = sql """show table stats part"""
+ if (result[0][6].equals("true")) {
+ result = """show index stats part part"""
+ logger.info("Report not ready. index stats: " + result)
+ sql """analyze table part with sample rows 100 with sync"""
+ result = sql """show table stats part"""
+ }
+ if (result[0][6].equals("true")) {
+ result = """show index stats part part"""
+ logger.info("Report not ready. index stats: " + result)
+ sql """analyze table part with sample rows 100 with sync"""
+ result = sql """show table stats part"""
+ }
+ assertEquals("false", result[0][6])
+
sql """drop database if exists test_drop_stats_and_truncate"""
}
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
