Skip to content

HBASE-30016 Cached table regions locations are not using in TableSnapshotInputFormatImpl#getSplits #7975

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
srinireddy2020 wants to merge 1 commit into
base: master
Choose a base branch
from
Open

HBASE-30016 Cached table regions locations are not using in TableSnapshotInputFormatImpl#getSplits #7975

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
106 changes: 54 additions & 52 deletions ...preduce/src/main/java/org/apache/hadoop/hbase/mapreduce/TableSnapshotInputFormatImpl.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -424,74 +424,76 @@ public static List<InputSplit> getSplits(Scan scan, SnapshotManifest manifest,

Connection connection = null;
RegionLocator regionLocator = null;
if (localityEnabled && useRegionLoc) {
Configuration newConf = new Configuration(conf);
newConf.setInt("hbase.hconnection.threads.max", 1);
try {
List<InputSplit> splits = new ArrayList<>();
try {
Copy link
Contributor

@NihalJain NihalJain Mar 25, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

can we do try-with-resources here

if (localityEnabled && useRegionLoc) {
Configuration newConf = new Configuration(conf);
newConf.setInt("hbase.hconnection.threads.max", 1);

connection = ConnectionFactory.createConnection(newConf);
regionLocator = connection.getRegionLocator(htd.getTableName());

/* Get all locations for the table and cache it */
regionLocator.getAllRegionLocations();
} finally {
if (connection != null) {
connection.close();
}
}
}

List<InputSplit> splits = new ArrayList<>();
for (RegionInfo hri : regionManifests) {
// load region descriptor
List<String> hosts = null;
if (localityEnabled) {
if (regionLocator != null) {
/* Get Location from the local cache */
HRegionLocation location = regionLocator.getRegionLocation(hri.getStartKey(), false);

hosts = new ArrayList<>(1);
hosts.add(location.getHostname());
} else {
hosts = calculateLocationsForInputSplit(conf, htd, hri, tableDir);
for (RegionInfo hri : regionManifests) {
// load region descriptor
List<String> hosts = null;
if (localityEnabled) {
if (regionLocator != null) {
/* Get Location from the local cache */
HRegionLocation location = regionLocator.getRegionLocation(hri.getStartKey(), false);

hosts = new ArrayList<>(1);
hosts.add(location.getHostname());
} else {
hosts = calculateLocationsForInputSplit(conf, htd, hri, tableDir);
}
}
}

if (numSplits > 1) {
byte[][] sp = sa.split(hri.getStartKey(), hri.getEndKey(), numSplits, true);
for (int i = 0; i < sp.length - 1; i++) {
if (numSplits > 1) {
byte[][] sp = sa.split(hri.getStartKey(), hri.getEndKey(), numSplits, true);
for (int i = 0; i < sp.length - 1; i++) {
if (
PrivateCellUtil.overlappingKeys(scan.getStartRow(), scan.getStopRow(), sp[i],
sp[i + 1])
) {

Scan boundedScan = new Scan(scan);
if (scan.getStartRow().length == 0) {
boundedScan.withStartRow(sp[i]);
} else {
boundedScan.withStartRow(
Bytes.compareTo(scan.getStartRow(), sp[i]) > 0 ? scan.getStartRow() : sp[i]);
}

if (scan.getStopRow().length == 0) {
boundedScan.withStopRow(sp[i + 1]);
} else {
boundedScan.withStopRow(Bytes.compareTo(scan.getStopRow(), sp[i + 1]) < 0
? scan.getStopRow()
: sp[i + 1]);
}

splits.add(new InputSplit(htd, hri, hosts, boundedScan, restoreDir));
}
}
} else {
if (
PrivateCellUtil.overlappingKeys(scan.getStartRow(), scan.getStopRow(), sp[i], sp[i + 1])
PrivateCellUtil.overlappingKeys(scan.getStartRow(), scan.getStopRow(),
hri.getStartKey(), hri.getEndKey())
) {

Scan boundedScan = new Scan(scan);
if (scan.getStartRow().length == 0) {
boundedScan.withStartRow(sp[i]);
} else {
boundedScan.withStartRow(
Bytes.compareTo(scan.getStartRow(), sp[i]) > 0 ? scan.getStartRow() : sp[i]);
}

if (scan.getStopRow().length == 0) {
boundedScan.withStopRow(sp[i + 1]);
} else {
boundedScan.withStopRow(
Bytes.compareTo(scan.getStopRow(), sp[i + 1]) < 0 ? scan.getStopRow() : sp[i + 1]);
}

splits.add(new InputSplit(htd, hri, hosts, boundedScan, restoreDir));
splits.add(new InputSplit(htd, hri, hosts, scan, restoreDir));
}
}
} else {
if (
PrivateCellUtil.overlappingKeys(scan.getStartRow(), scan.getStopRow(), hri.getStartKey(),
hri.getEndKey())
) {

splits.add(new InputSplit(htd, hri, hosts, scan, restoreDir));
}
}
} finally {
if (connection != null) {
Copy link
Contributor

@NihalJain NihalJain Mar 25, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

i believe all changes are just because we added an extra block {} and code got shifted?

connection.close();
}
}

return splits;
}

Expand Down