chandrasekhar-188k commented on code in PR #5833:
URL: https://github.com/apache/hbase/pull/5833#discussion_r2365013900
##########
hbase-server/src/main/java/org/apache/hadoop/hbase/mob/MobFileCleanerChore.java:
##########
@@ -83,29 +112,43 @@ protected void chore() {
LOG.error("MobFileCleanerChore failed", e);
return;
}
+ List<Future> futureList = new ArrayList<>(map.size());
for (TableDescriptor htd : map.values()) {
- for (ColumnFamilyDescriptor hcd : htd.getColumnFamilies()) {
- if (hcd.isMobEnabled() && hcd.getMinVersions() == 0) {
- try {
- cleaner.cleanExpiredMobFiles(htd, hcd);
- } catch (IOException e) {
- LOG.error("Failed to clean the expired mob files table={}
family={}",
- htd.getTableName().getNameAsString(), hcd.getNameAsString(), e);
- }
- }
- }
+ Future<?> future = threadPool.submit(() -> handleOneTable(htd));
+ futureList.add(future);
+ }
+
+ for (Future future : futureList) {
try {
- // Now clean obsolete files for a table
- LOG.info("Cleaning obsolete MOB files from table={}",
htd.getTableName());
- try (final Admin admin = master.getConnection().getAdmin()) {
-
MobFileCleanupUtil.cleanupObsoleteMobFiles(master.getConfiguration(),
htd.getTableName(),
- admin);
+ future.get(cleanerFutureTimeout, TimeUnit.SECONDS);
+ } catch (InterruptedException | ExecutionException | TimeoutException e)
{
Review Comment:
added code to restore the state.
Since the cleaning of multiple mob files already submitted to pool, I think
we can continue to wait for other mob file clearing to finish even if we get
InterruptedException for one of them.
--
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]
