[
https://issues.apache.org/jira/browse/HBASE-12581?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14227545#comment-14227545
]
stack commented on HBASE-12581:
-------------------------------
[~ram_krish] Thanks for taking a look!
bq. I still need to see whether the recent Connection related changes is
slowing things in any way.
I was thinking that because the Connection instance varies, we now may 'miss'
notification of the acl update.
I've seen versions of our missing notification of acl update in a few guises
during the move to unmanaged connections. One interesting one was over in
TestAccessController.testTruncatePerms. It does truncate. Truncate is kinda
broke in that it copy/pastes what create table handler does only its missing
pieces. I was seeing that after putting back the table, the acl perms were not
also put back for this 'new' table. I changed the test back to how it was
before just to get it passing again. Let me file an issue on truncate and this
test now (HBASE-12592).
bq. Introducing a sleep after setUp() method of the test makes the test pass
consistently.
We need to add something like the waitFor in HBaseTestingUtility where we wait
for the acl stuff to show up before we proceed with the test? Would need to add
it in to these tests here in this test suite but also to any test making use of
acl table.
For acl perms, we always read from zk, never from the table?
Thanks for helping [~ram_krish] That helped a bunch.
> TestCellACLWithMultipleVersions failing since task 5 HBASE-12404 (HBASE-12404
> addendum)
> ---------------------------------------------------------------------------------------
>
> Key: HBASE-12581
> URL: https://issues.apache.org/jira/browse/HBASE-12581
> Project: HBase
> Issue Type: Bug
> Components: test
> Reporter: stack
> Assignee: stack
> Priority: Critical
> Fix For: 2.0.0, 0.99.2
>
> Attachments: 12581.addendum.txt, 12581.txt, acls.txt
>
>
> TestCellACLWithMultipleVersions failed after HBASE-12404 went in (though it
> passed twice on hadoopqa!). Fails locally too. Here, make a Connection when
> we go to check perms. That seems to fix it. Going to commit since patch is
> just more of hbase-12404... this is in essence and addendum.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
