Duo Zhang created HBASE-29380:
---------------------------------
Summary: Two concurrent remove peer requests may hang
Key: HBASE-29380
URL: https://issues.apache.org/jira/browse/HBASE-29380
Project: HBase
Issue Type: Bug
Components: proc-v2, Replication
Reporter: Duo Zhang
TestAsyncReplicationAdminApi fails about 75% on flaky dashboard.
Tried locally, sometimes testRemovePeerTableCFs may hang.
The situation here is that, at the end of this test method, we will call
admin.removeReplicationPeer(ID_ONE);
The admin here is an AsyncAdmin, so it will return immediately, and then in the
Junit's AFter methods, we will call
admin.removeReplicationPeer(ID_ONE).join();
Sometimes, if the first request runs first, the remove peer procedure for the
second one will not wake up so we hang there forever.
Need to dig more.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
