github-actions[bot] commented on code in PR #68180:
URL: https://github.com/apache/doris/pull/68180#discussion_r4063714837
##########
fe/fe-core/src/main/java/org/apache/doris/catalog/MTMV.java:
##########
@@ -682,33 +687,217 @@ public void invalidateIvmBaseline() {
editLogItem.await();
}
- public void invalidateIvmBaseline(BaseTableInfo baseTableInfo, Map<String,
Long> changedPartitions) {
+ /**
+ * Mark the MV partitions that may hold rows read from the changed base
table partitions as needing a
+ * rebuild. When those partitions cannot be determined, the whole MV is
marked instead.
+ */
+ /**
+ * @return whether a barrier was recorded. The caller reports the two
outcomes differently: a change
+ * that no MV partition reads leaves nothing to rebuild and must
not be logged as one.
+ */
+ public boolean invalidateIvmBaseline(BaseTableInfo baseTableInfo,
Map<String, Long> changedPartitions) {
+ // Computed before the MV lock is taken, not inside it: the mapping
reads the partition items of the
+ // MV and of every PCT table, so it takes those tables' locks, and the
MV lock has to stay a leaf
+ // (nothing may be acquired under it) the way the rest of this class
assumes. The selection does not
+ // need to be atomic with the barrier it produces: the barrier is
recorded under the lock below, and
+ // the names it carries are intersected with the live partition names
when they are consumed
+ // (MTMVTask).
+ Optional<Set<String>> affectedMvPartitions =
selectAffectedMvPartitions(baseTableInfo,
+ changedPartitions);
+ if (affectedMvPartitions.isPresent() &&
affectedMvPartitions.get().isEmpty()) {
+ // No MV partition reads any of the changed base partitions, so
this change cannot leave
+ // anything behind here: there is no barrier to persist, and
skipping the version bump
+ // keeps it from discarding the result of a task that is already
running.
+ LOG.debug("No MV partition is affected by changed base partitions,
mv={}, baseTable={}, "
+ + "changedPartitions={}", name, baseTableInfo,
changedPartitions);
+ return false;
+ }
EditLogItem editLogItem;
writeMvLock();
try {
if (ivmInfo == null) {
ivmInfo = new IvmInfo();
}
- if (mvPartitionInfo.getPartitionType() !=
MTMVPartitionType.SELF_MANAGE
- && mvPartitionInfo.getPctInfos().stream()
- .anyMatch(pctInfo ->
pctInfo.getTableInfo().equals(baseTableInfo))) {
- Optional<Set<String>> mvPartitionNames =
refreshSnapshot.getMvPartitionNames(baseTableInfo,
- changedPartitions);
- if (mvPartitionNames.isPresent()) {
-
ivmInfo.addPendingBaselineRebuildPartitions(mvPartitionNames.get());
- } else {
- // Without a snapshot for every changed base partition, a
PARTITIONS rebuild is unsafe.
- ivmInfo.requireCompleteBaselineRebuild();
- }
- } else {
+ if (!affectedMvPartitions.isPresent()) {
+ // A narrower rebuild could leave a partition holding rows of
the changed base partition
+ // untouched, and those rows cannot be repaired later: the
change emitted no row binlog.
ivmInfo.requireCompleteBaselineRebuild();
+ } else {
+
ivmInfo.addPendingBaselineRebuildPartitions(affectedMvPartitions.get());
}
schemaChangeVersion++;
editLogItem = submitIvmInfoChange();
} finally {
writeMvUnlock();
}
editLogItem.await();
+ return true;
+ }
+
+ /**
+ * Select the MV partitions that may hold rows read from the changed base
table partitions.
+ *
+ * <p>This asks which MV partitions read the changed base partitions at
all, instead of (as the
+ * refresh snapshot based selection did) which of them had already seen
them. The snapshot is a lower
+ * bound that is allowed to lag: a base partition that was added after the
snapshot was captured never
+ * appears in it, so it can report "this partition never read the changed
base partition" about a
+ * partition that does hold its rows. Missing a partition here is not
repaired by a later refresh --
+ * dropping or truncating a base partition emits no row binlog, so the
incremental path never learns
+ * about those orphan rows and they stay in the MV forever.
+ *
+ * <p>Three cases have no answer in the mapping, and each of them must
rebuild the whole MV instead:
+ * a SELF_MANAGE MV (the mapping API answers nothing for it, although its
single partition reads every
+ * base partition); a base table that is not one of the MV's PCT tables
(the mapping is seeded from
+ * {@code getPctTables()} and never gains a table later, so a joined
partition table that the MV's
+ * partition column does not reach is not described at all); and a changed
partition that is not in
+ * the base table's metadata right now, which is how RECOVER PARTITION
arrives here -- it marks before
+ * the partition is added back, so at this point the partition is still in
the recycle bin.
+ *
+ * <p>Locking is the fourth way to end up rebuilding everything, but it is
contention rather than a
+ * property of the MV: the tables whose partition items the mapping reads
are locked with a bounded
+ * tryLock, and the MV is rebuilt only while one of them is being written.
See the comment at that
+ * loop.
+ *
+ * <p>An empty result is meaningful, on the other hand: the mapping lists
every base partition read by
+ * the MV, so a changed base partition that no MV partition maps to is
read by none of them.
+ *
+ * @param changedBasePartitions base partition name to partition id, never
empty
+ * @return {@link Optional#empty()} when the affected MV partitions cannot
be determined, otherwise the
+ * (possibly empty) set of MV partition names that must be rebuilt
+ */
+ private Optional<Set<String>> selectAffectedMvPartitions(BaseTableInfo
baseTableInfo,
+ Map<String, Long> changedBasePartitions) {
+ if (mvPartitionInfo.getPartitionType() ==
MTMVPartitionType.SELF_MANAGE) {
+ return Optional.empty();
+ }
+ MTMVRelatedTableIf pctTable = findPctTable(baseTableInfo);
+ if (pctTable == null) {
+ return Optional.empty();
+ }
+ // Computing the mapping reads the partition items of the MV and of
every PCT table, which means
+ // taking their read locks. The caller already holds the changed
table's write lock (a partition DDL
+ // marks before it releases it), so these other reads must not block:
two partition DDLs on two PCT
+ // tables of this MV would otherwise each hold the write lock the
other one needs, and acquiring in
+ // id order cannot break a cycle whose first lock is already held.
They are taken with a bounded
+ // tryLock instead, the way the stream cleanup treats a busy table: a
busy table means a writer is
+ // involved, and then the whole MV is rebuilt. The list is still
sorted by id so that the acquisition
+ // order matches the rest of the code base.
+ List<TableIf> tablesToRead =
Lists.newArrayListWithCapacity(mvPartitionInfo.getPctInfos().size() + 1);
+ tablesToRead.add(this);
+ for (BaseColInfo pctInfo : mvPartitionInfo.getPctInfos()) {
+ if (pctInfo.getTableInfo().equals(baseTableInfo)) {
+ continue;
+ }
+ try {
+ tablesToRead.add(MTMVUtil.getTable(pctInfo.getTableInfo()));
+ } catch (Exception e) {
+ LOG.warn("Failed to resolve PCT table {}, rebuild the whole
MV. mv={}",
+ pctInfo.getTableInfo(), name, e);
+ return Optional.empty();
+ }
+ }
+ tablesToRead.sort(Comparator.comparing(TableIf::getId));
+ if (!MetaLockUtils.tryReadLockTables(tablesToRead,
Table.TRY_LOCK_TIMEOUT_MS, TimeUnit.MILLISECONDS)) {
+ LOG.warn("A PCT table is busy, rebuild the whole MV {} instead of
selecting part of it", name);
+ return Optional.empty();
+ }
+ try {
+ // A partition that is missing from the metadata here is invisible
to the mapping as well, so
+ // an empty answer below would be indistinguishable from "no MV
partition reads it". The match
+ // is deliberately exact: the mapping is keyed by the metadata's
spelling, so a name that only
+ // differs in case must take the whole-MV path too, or the lookup
below would quietly select
+ // nothing for a partition that some MV partition does read. Base
tables that do not implement
+ // getPartitionNames -- the external ones -- report no partition
at all, so a partition change
+ // on them always rebuilds the whole MV. That matches what the
refresh-snapshot selection
+ // answered for them, and the mapping has never been exercised for
external tables (IVM does
+ // not support them as base tables yet): revisit before taking the
narrow path for them.
+ if
(!pctTable.getPartitionNames().containsAll(changedBasePartitions.keySet())) {
Review Comment:
[P1] Validate the changed partition IDs before trusting this name-only
membership check. `RECOVER PARTITION p3 AS p6` may run while a newly added,
different-range `p3` is already live (the recycle-bin regression suite
exercises this supported flow). The recovery marker passes `{p3 -> recycledId}`
before inserting `p6`; this check accepts the replacement `p3`, and the mapping
narrows the barrier to that replacement's MV partition. Partition sync can then
add the recovered range while the baseline pre-step rebuilds only the wrong
partition; recovery has no row binlog to restore the recovered rows. Please
require each live partition's ID to match the supplied ID and fall back to
COMPLETE on a missing or mismatched incarnation.
##########
fe/fe-core/src/test/java/org/apache/doris/mtmv/ivm/IvmBaselineRebuildTest.java:
##########
@@ -122,35 +127,267 @@ public void
testDropColumnMarksBaselineRebuildOnlyWhenReferenced() throws Except
Assertions.assertTrue(mtmv.getIvmInfo().isBaselineRebuildRequired());
}
+ /**
+ * Which MV partitions must be rebuilt is decided by the MV's partition
mapping, not by what the
+ * refresh snapshot happens to record. This test publishes no snapshot at
all: an MV whose partitions
+ * follow the base table's still narrows the rebuild down to the
partitions that read the dropped one.
+ */
@Test
- public void testPublishedPctPartitionUsesPartitionsBaselineRebuild()
throws Exception {
+ public void testDropPartitionMarksOnlyMvPartitionsThatReadIt() throws
Exception {
String db = "ivm_partitions_baseline_rebuild";
- createPartitionedIvmTableAndMv(db);
+ createPartitionedIvmTableAndPartitionedMv(db);
MTMV mtmv = getMtmv(db);
- OlapTable baseTable = getBaseTable(db);
- publishPctPartitionSnapshot(mtmv, baseTable, "p202001");
+ Assertions.assertEquals(2, mtmv.getPartitionNames().size());
+ Set<String> expected = mvPartitionsWithSameRange(mtmv,
getBaseTable(db), "p202001");
+ Assertions.assertEquals(1, expected.size());
executeSql("ALTER TABLE ivm_base DROP PARTITION p202001");
Assertions.assertFalse(mtmv.getIvmInfo().requiresCompleteBaselineRebuild());
- Assertions.assertEquals(Collections.singleton("mv_partition"),
- mtmv.getIvmInfo().getPendingBaselineRebuildPartitions());
+ Assertions.assertEquals(expected,
mtmv.getIvmInfo().getPendingBaselineRebuildPartitions());
}
+ /**
+ * A base partition that no MV partition reads: dropping it cannot leave
any of its rows in the MV, so
+ * there is nothing to rebuild. The previous selection could not tell this
apart from "the snapshot does
+ * not know this partition" and rebuilt the whole MV instead.
+ */
@Test
- public void testMissingPctSnapshotRequiresCompleteBaselineRebuild() throws
Exception {
- String db = "ivm_complete_baseline_rebuild";
- createPartitionedIvmTableAndMv(db);
+ public void testDropPartitionOutsideMvPartitionsMarksNothing() throws
Exception {
+ String db = "ivm_partition_outside_mv";
+ createPartitionedIvmTableAndPartitionedMv(db);
+ MTMV mtmv = getMtmv(db);
+ // Added after the MV was created and never synced into it, so no MV
partition reads it.
+ executeSql("ALTER TABLE ivm_base ADD PARTITION p202003 "
+ + "VALUES [('2020-03-01'), ('2020-04-01'))");
+ Assertions.assertTrue(mvPartitionsWithSameRange(mtmv,
getBaseTable(db), "p202003").isEmpty());
+
+ executeSql("ALTER TABLE ivm_base DROP PARTITION p202003");
+
+ Assertions.assertFalse(mtmv.getIvmInfo().isBaselineRebuildRequired());
+ }
+
+ /**
+ * A base partition the partition_sync_limit window no longer covers can
still have its rows in an MV
+ * partition: the MV was built while that partition was inside the window,
shrinking the window does not
+ * touch the MV's own partitions, and widening it again makes partition
sync keep the partition holding
+ * those rows. TRUNCATE emits no binlog, so nothing incremental can repair
them -- the whole MV has to be
+ * rebuilt rather than a partition being guessed at.
+ */
+ @Test
+ public void testChangedPartitionOutsideTheSyncWindowRebuildsTheWholeMv()
throws Exception {
+ String db = "ivm_baseline_sync_window";
+ String thisYear = LocalDate.now().withDayOfYear(1).toString();
+ String nextYear =
LocalDate.now().withDayOfYear(1).plusYears(1).toString();
+ createDatabaseAndUse(db);
+ createTable("CREATE TABLE " + db + ".ivm_base (\n"
+ + " dt date NOT NULL,\n"
+ + " k1 int,\n"
+ + " v1 int\n"
+ + ")\n"
+ + "DUPLICATE KEY(dt, k1)\n"
+ + "PARTITION BY RANGE(dt) (\n"
+ + " PARTITION p202001 VALUES [('2020-01-01'),
('2020-02-01')),\n"
+ + " PARTITION p202002 VALUES [('2020-02-01'),
('2020-03-01')),\n"
+ + " PARTITION pThisYear VALUES [('" + thisYear + "'), ('" +
nextYear + "'))\n"
+ + ")\n"
+ + "DISTRIBUTED BY HASH(k1) BUCKETS 1\n"
+ + "PROPERTIES ('replication_num' = '1', 'binlog.enable' =
'true', 'binlog.format' = 'ROW')");
+ createMvByNereids("CREATE MATERIALIZED VIEW ivm_mv\n"
+ + "BUILD DEFERRED REFRESH INCREMENTAL ON MANUAL\n"
+ + "PARTITION BY(dt)\n"
+ + "DISTRIBUTED BY RANDOM BUCKETS 1\n"
+ + "PROPERTIES ('replication_num' = '1')\n"
+ + "AS SELECT dt, k1, v1 FROM ivm_base");
+ MTMV mtmv = getMtmv(db);
+ Assertions.assertEquals(3, mtmv.getPartitionNames().size());
+
+ // The window now keeps only this year's partition, so p202001 leaves
the mapping while the MV's own
+ // partition for it stays. TRUNCATE leaves the base partition in
place, so partition sync would keep
+ // that MV partition too -- the rows it still holds are exactly what
the rebuild has to remove.
+ executeSql("ALTER MATERIALIZED VIEW ivm_mv SET ('partition_sync_limit'
= '1',"
+ + " 'partition_sync_time_unit' = 'YEAR')");
+ executeSql("TRUNCATE TABLE ivm_base PARTITION(p202001)");
+
+
Assertions.assertTrue(mtmv.getIvmInfo().requiresCompleteBaselineRebuild());
+ }
+
+ /**
+ * The same window, with a change that touches a partition inside it and
one outside it at once: the
+ * partition inside fills the selection, and the one outside contributes
nothing because the window
+ * left it out of the mapping. Judging the change by "was anything
selected" would mark only the MV
+ * partition backed by the inside half, and the rows of the outside half
-- which the MV partition for
+ * it still holds -- would never be rebuilt.
+ */
+ @Test
+ public void
testChangeThatMixesInWindowAndOutOfWindowPartitionsRebuildsTheWholeMv() throws
Exception {
+ String db = "ivm_baseline_sync_window_mixed";
+ String thisYear = LocalDate.now().withDayOfYear(1).toString();
Review Comment:
[P2] Keep these sync-window cases valid across a year rollover. `pThisYear`
ends at January 1 of the next year, but production derives its YEAR cutoff
later in the Doris session zone. If setup crosses January 1, or the JVM and
session zones straddle it, the range's upper endpoint equals the new cutoff, so
both `p202001` and `pThisYear` are filtered. The old `res.isEmpty()`
implementation then also records COMPLETE and this mixed regression passes
without exercising the non-empty incomplete selection it was added for. Please
make the recent partition in both tests extend safely past an immediate
rollover, or assert the intended mapping shape before the TRUNCATE.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
