[
https://issues.apache.org/jira/browse/HBASE-7966?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13591334#comment-13591334
]
stack commented on HBASE-7966:
------------------------------
Here is failure w/ stdout:
https://builds.apache.org/job/hbase-0.95/2/testReport/junit/org.apache.hadoop.hbase.security.access/TestAccessControlFilter/testQualifierAccess/
As soon as master is initialized, some one is trying to scan a not-yet-created
_acl_ table.
2013-02-26 21:42:50,488 DEBUG [Master:0;vesta.apache.org,50372,1361914960929]
client.ClientScanner(209): Advancing internal scanner to startKey at '_acl_,,'
200ms later we start the creation of _acl_
2013-02-26 21:42:50,644 DEBUG [Master:0;vesta.apache.org,50372,1361914960929]
master.TableLockManager$ZKTableLockManager$TableLockImpl(225): Attempt to
acquire table write lock on :_acl_ for:C_M_CREATE_TABLE
In same second just after acquiring lock because we are going to start in for
real on table creation, someone is trying again to scan...
2013-02-26 21:42:50,680 DEBUG [Master:0;vesta.apache.org,50372,1361914960929]
client.ClientScanner(209): Advancing internal scanner to startKey at '_acl_,,'
Seems like its the ACL#addUserPermission running in the postCreateTable.
But its #92 inside in testQualifiedAccess. Again we have blown past
waitTableEnabled on _acl_ though it not yet present.
Adding a bit of debug log...
> ACL tests fail on trunk (flaky)
> -------------------------------
>
> Key: HBASE-7966
> URL: https://issues.apache.org/jira/browse/HBASE-7966
> Project: HBase
> Issue Type: Bug
> Components: security
> Reporter: Enis Soztutar
> Assignee: Enis Soztutar
> Fix For: 0.95.0, 0.98.0
>
>
> I've noticed that these tests fail for me quite frequently:
> {code}
> org.apache.hadoop.hbase.security.access.TestTablePermissions:
> org.apache.hadoop.hbase.exceptions.TableNotFoundException: _acl_
>
> testQualifierAccess(org.apache.hadoop.hbase.security.access.TestAccessControlFilter):
> org.apache.hadoop.hbase.exceptions.TableNotFoundException: _acl_
> {code}
> The root cause seems to be that, the AccessController only creates the table
> in HMaster.postStartMaster(), but before _acl_ is created, master start
> accepting other create table requests, hence race.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
