chaijunjie0101 commented on code in PR #7117:
URL: https://github.com/apache/hbase/pull/7117#discussion_r2192876668
##########
hbase-client/src/main/java/org/apache/hadoop/hbase/client/AsyncNonMetaRegionLocator.java:
##########
@@ -91,6 +95,9 @@ class AsyncNonMetaRegionLocator {
private final ConcurrentMap<TableName, TableCache> cache = new
ConcurrentHashMap<>();
+ // A chore service to invalidate table cache which table is not exist or
disabled.
+ private ChoreService metaCacheInvalidateChoreService;
Review Comment:
sorry for replaying too late..Thank you for reviewing...
I checked org.apache.hadoop.hbase.client.AsyncConnectionImpl#RETRY_TIMER,
for HashedWheelTimer, it seems not support schedule task, just like
**scheduleWithFixedDelay** in **ScheduledExecutorService**, it only support
call org.apache.hbase.thirdparty.io.netty.util.Timer#newTimeout to execute one
time(delay time task), we need check all table state in the meta cache
regularly, there one way is in the ending of TimeTask to call newTimeout add
itself...but we are not Implement a common tool to use HashedWheelTimer to do
schedule logic...and if this timetask cost much more time, it may also cause
other task timeout...(these tasks execute serial at the same tick...and there
are so many client logic using the same Timer...)
so, sir, should us need use HashedWheelTimer to do schedule logic? or there
are more suggesting?
--
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]
