The only rush for me is the uncertainty of time to fix this later vs
time to fix it now. :)
a couple weeks will give me a chance to get the functional testing
stuff I wanted to use working. since not much is expected to change
perf wise between the code now and the next RC maybe I can find time
Thanks.
Probably the earliest I'll have available for running the next RC is week
after next. Plenty of time to get this sorted out. Also, there is still
that testing I've promised that has been hard to find time for recently.
On Tue, Mar 12, 2019 at 11:08 AM Sean Busbey wrote:
> file
file HBASE-22044. I'll start putting up patches shortly.
On Tue, Mar 12, 2019 at 11:52 AM Andrew Purtell
wrote:
>
> Could you file a JIRA for this?
>
> I don’t understand why a class like that should be part of our public API.
> What about it offers HBase functionality? IMHO - none. It’s a
Could you file a JIRA for this?
I don’t understand why a class like that should be part of our public API. What
about it offers HBase functionality? IMHO - none. It’s a utility class for
internal implementation detail.
If this is a concern make this class Private.
Making a release candidate
-1, but maybe it's fine.
I would like to discuss a complaint from the compatibility report. I
believe it is against our promises and I don't see it discussed on the
relevant jiras nor noted above.
ByteBufferUtils is IA.Public and had a public method removed in
HBASE-20716 (the jira is not marked
Thank you for the clarification, Sean!
On Fri, Mar 8, 2019 at 5:03 PM Sean Busbey wrote:
> I count two binding (Josh, Peter) and one non-binding (Xu)
>
> On Fri, Mar 8, 2019, 15:48 Andrew Purtell wrote:
>
> > We have one binding vote and two nonbinding votes.
> >
> > I'm extending this vote
I count two binding (Josh, Peter) and one non-binding (Xu)
On Fri, Mar 8, 2019, 15:48 Andrew Purtell wrote:
> We have one binding vote and two nonbinding votes.
>
> I'm extending this vote another week because $dayjob stuff preempted time
> for the testing as promised. I will try to close it
We have one binding vote and two nonbinding votes.
I'm extending this vote another week because $dayjob stuff preempted time
for the testing as promised. I will try to close it end of next week.
On Thu, Feb 28, 2019 at 10:16 AM Andrew Purtell wrote:
> We have one binding vote so far, need two
+1
Checksums, signatures: OK
Build from source: OK
Unit tests: OK
Rat check: OK
LTT 1M rows: OK
Basic shell commands: OK
Master, RegionServer UI: OK
Logs: OK
On Thu, Feb 28, 2019 at 7:17 PM Andrew Purtell wrote:
> We have one binding vote so far, need two more.
> Thanks for the nonbinding
We have one binding vote so far, need two more.
Thanks for the nonbinding vote, Xu.
I have not yet had time to do the performance testing as promised, so let
me extend the vote for a week. I will try to close it end of next week.
On Thu, Feb 21, 2019 at 3:44 PM Andrew Purtell wrote:
> The
+1 (binding)
Checked out src archive, things look OK with it.
Thanks, Andrew!
On 2/21/19 6:44 PM, Andrew Purtell wrote:
The third HBase 1.5.0 release candidate (RC2) is available for download at
https://dist.apache.org/repos/dist/dev/hbase/hbase-1.5.0RC2/ and Maven
artifacts are available in
+1
RAT check passes
2 Jars' sha512 verified.
Unit test suite passes (8u191)
Except this one:
https://issues.apache.org/jira/browse/HBASE-21952
and this one:
TestSnapshotFromMaster.testAsyncSnapshotWillNotBlockSnapshotHFileCleaner ,
This is in our Flaky test
12 matches
Mail list logo
