Stack
2012-01-08, 21:39
Stack
2012-01-14, 01:00
Stack
2012-01-16, 13:07
yuzhihong@...
2012-01-16, 14:33
Ramkrishna.S.Vasudevan
2012-01-16, 15:54
Stack
2012-01-16, 23:59
Stack
2012-01-16, 23:59
Jonathan Hsieh
2012-01-17, 00:03
Ted Yu
2012-01-17, 00:08
Stack
2012-01-17, 00:46
Andrew Purtell
2012-01-16, 17:14
Jean-Daniel Cryans
2012-01-14, 01:10
Ted Yu
2012-01-14, 01:15
Jean-Daniel Cryans
2012-01-14, 01:21
Ted Yu
2012-01-14, 01:27
Stack
2012-01-14, 05:23
Ted Yu
2012-01-14, 05:35
lars hofhansl
2012-01-14, 05:47
Stack
2012-01-14, 06:16
Ramkrishna.S.Vasudevan
2012-01-14, 06:41
lars hofhansl
2012-01-14, 22:09
Stack
2012-01-14, 06:15
Ted Yu
2012-01-14, 11:15
Stack
2012-01-15, 02:42
yuzhihong@...
2012-01-15, 03:05
Stack
2012-01-15, 03:42
Andrew Purtell
2012-01-14, 17:45
Jonathan Hsieh
2012-01-14, 18:11
Roman Shaposhnik
2012-01-14, 23:59
Lars George
2012-01-15, 13:31
Ted Yu
2012-01-15, 15:13
Lars George
2012-01-16, 03:01
-
ANN: The fourth hbase 0.92.0 release candidate is available for downloadStack 2012-01-08, 21:39
The fourth hbase 0.92.0 release candidate is available for download:
http://people.apache.org/~stack/hbase-0.92.0-candidate-3/
I've posted a secure and an insecure tarball built by jenkins.
This RC bundles hadoop 1.0.0 and zookeeper 3.4.2.
As said previous, HBase 0.92.0 includes a wagon-load of new features
including coprocessors, security, a new (self-migrating) file format,
distributed log splitting, etc. There's been > 660 fixes since 0.90.
See the list here: http://su.pr/1VZzl5
See the hbase manual for the low-down on what hadoop version this
release will run on: http://hbase.apache.org/book.html#hadoop
There is no migration necessary. A shutdown and restart after putting
in place the new HBase should be all thats involved (A rolling restart
from 0.90.x to 0.92.0 will not work). That said, once you have moved
to 0.92.x, there is no going back to 0.90.x after the transition has
been made.
Should we release this candidate as hbase 0.92.0? Take it for a spin.
Check out the doc. Vote +1/-1 by Martin Luther King Day, January 16th, 2012.
Yours,
The HBasistas
P.S. This candidate is looking pretty good. If you are thinking of using 0.92.0
any time soon, we would suggest that you dig in now and try it. Thanks.
http://people.apache.org/~stack/hbase-0.92.0-candidate-3/
I've posted a secure and an insecure tarball built by jenkins.
This RC bundles hadoop 1.0.0 and zookeeper 3.4.2.
As said previous, HBase 0.92.0 includes a wagon-load of new features
including coprocessors, security, a new (self-migrating) file format,
distributed log splitting, etc. There's been > 660 fixes since 0.90.
See the list here: http://su.pr/1VZzl5
See the hbase manual for the low-down on what hadoop version this
release will run on: http://hbase.apache.org/book.html#hadoop
There is no migration necessary. A shutdown and restart after putting
in place the new HBase should be all thats involved (A rolling restart
from 0.90.x to 0.92.0 will not work). That said, once you have moved
to 0.92.x, there is no going back to 0.90.x after the transition has
been made.
Should we release this candidate as hbase 0.92.0? Take it for a spin.
Check out the doc. Vote +1/-1 by Martin Luther King Day, January 16th, 2012.
Yours,
The HBasistas
P.S. This candidate is looking pretty good. If you are thinking of using 0.92.0
any time soon, we would suggest that you dig in now and try it. Thanks.
+
Stack 2012-01-08, 21:39
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadStack 2012-01-14, 01:00
+1 on releasing this as 0.92.0
I've done some spotty testing of it over the last week and it seems to
basically work (I was doing mostly write loads with aggressive
splitting in small heaps killing a node from time to time then doing
hbck to see if stuff survived...)
I've downloaded it and checked out the doc. All seems fine.
St.Ack
On Sun, Jan 8, 2012 at 1:39 PM, Stack <[EMAIL PROTECTED]> wrote:
> The fourth hbase 0.92.0 release candidate is available for download:
>
> http://people.apache.org/~stack/hbase-0.92.0-candidate-3/
>
> I've posted a secure and an insecure tarball built by jenkins.
>
> This RC bundles hadoop 1.0.0 and zookeeper 3.4.2.
>
> As said previous, HBase 0.92.0 includes a wagon-load of new features
> including coprocessors, security, a new (self-migrating) file format,
> distributed log splitting, etc. There's been > 660 fixes since 0.90.
> See the list here: http://su.pr/1VZzl5
>
> See the hbase manual for the low-down on what hadoop version this
> release will run on: http://hbase.apache.org/book.html#hadoop
>
> There is no migration necessary. A shutdown and restart after putting
> in place the new HBase should be all thats involved (A rolling restart
> from 0.90.x to 0.92.0 will not work). That said, once you have moved
> to 0.92.x, there is no going back to 0.90.x after the transition has
> been made.
>
> Should we release this candidate as hbase 0.92.0? Take it for a spin.
> Check out the doc. Vote +1/-1 by Martin Luther King Day, January 16th, 2012.
>
> Yours,
> The HBasistas
>
> P.S. This candidate is looking pretty good. If you are thinking of using 0.92.0
> any time soon, we would suggest that you dig in now and try it. Thanks.
I've done some spotty testing of it over the last week and it seems to
basically work (I was doing mostly write loads with aggressive
splitting in small heaps killing a node from time to time then doing
hbck to see if stuff survived...)
I've downloaded it and checked out the doc. All seems fine.
St.Ack
On Sun, Jan 8, 2012 at 1:39 PM, Stack <[EMAIL PROTECTED]> wrote:
> The fourth hbase 0.92.0 release candidate is available for download:
>
> http://people.apache.org/~stack/hbase-0.92.0-candidate-3/
>
> I've posted a secure and an insecure tarball built by jenkins.
>
> This RC bundles hadoop 1.0.0 and zookeeper 3.4.2.
>
> As said previous, HBase 0.92.0 includes a wagon-load of new features
> including coprocessors, security, a new (self-migrating) file format,
> distributed log splitting, etc. There's been > 660 fixes since 0.90.
> See the list here: http://su.pr/1VZzl5
>
> See the hbase manual for the low-down on what hadoop version this
> release will run on: http://hbase.apache.org/book.html#hadoop
>
> There is no migration necessary. A shutdown and restart after putting
> in place the new HBase should be all thats involved (A rolling restart
> from 0.90.x to 0.92.0 will not work). That said, once you have moved
> to 0.92.x, there is no going back to 0.90.x after the transition has
> been made.
>
> Should we release this candidate as hbase 0.92.0? Take it for a spin.
> Check out the doc. Vote +1/-1 by Martin Luther King Day, January 16th, 2012.
>
> Yours,
> The HBasistas
>
> P.S. This candidate is looking pretty good. If you are thinking of using 0.92.0
> any time soon, we would suggest that you dig in now and try it. Thanks.
+
Stack 2012-01-14, 01:00
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadStack 2012-01-16, 13:07
On Fri, Jan 13, 2012 at 5:00 PM, Stack <[EMAIL PROTECTED]> wrote:
> +1 on releasing this as 0.92.0
>
HBASE-5204 "Backward compatibility fixes for 0.92" looks like a
blocker to me. Without HBASE-5204, asynchbase/opentsdb will not work
on hbase 0.92.0. HBASE-5204 cannot go into a 0.92.1 because we would
not be able to do a rolling restart going from 0.92.0 to 0.92.1 (See
change in HbaseObjectWritable).
St.Ack
> +1 on releasing this as 0.92.0
>
HBASE-5204 "Backward compatibility fixes for 0.92" looks like a
blocker to me. Without HBASE-5204, asynchbase/opentsdb will not work
on hbase 0.92.0. HBASE-5204 cannot go into a 0.92.1 because we would
not be able to do a rolling restart going from 0.92.0 to 0.92.1 (See
change in HbaseObjectWritable).
St.Ack
+
Stack 2012-01-16, 13:07
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadyuzhihong@... 2012-01-16, 14:33
Thanks for the update, Stack.
Do you want to freeze patch submission to 0.92 branch until you cut rc4 ?
I think hbase-5153 and hbase-5179 are close to the quality of integration.
Cheers
On Jan 16, 2012, at 5:07 AM, Stack <[EMAIL PROTECTED]> wrote:
> On Fri, Jan 13, 2012 at 5:00 PM, Stack <[EMAIL PROTECTED]> wrote:
>> +1 on releasing this as 0.92.0
>>
>
> HBASE-5204 "Backward compatibility fixes for 0.92" looks like a
> blocker to me. Without HBASE-5204, asynchbase/opentsdb will not work
> on hbase 0.92.0. HBASE-5204 cannot go into a 0.92.1 because we would
> not be able to do a rolling restart going from 0.92.0 to 0.92.1 (See
> change in HbaseObjectWritable).
>
> St.Ack
Do you want to freeze patch submission to 0.92 branch until you cut rc4 ?
I think hbase-5153 and hbase-5179 are close to the quality of integration.
Cheers
On Jan 16, 2012, at 5:07 AM, Stack <[EMAIL PROTECTED]> wrote:
> On Fri, Jan 13, 2012 at 5:00 PM, Stack <[EMAIL PROTECTED]> wrote:
>> +1 on releasing this as 0.92.0
>>
>
> HBASE-5204 "Backward compatibility fixes for 0.92" looks like a
> blocker to me. Without HBASE-5204, asynchbase/opentsdb will not work
> on hbase 0.92.0. HBASE-5204 cannot go into a 0.92.1 because we would
> not be able to do a rolling restart going from 0.92.0 to 0.92.1 (See
> change in HbaseObjectWritable).
>
> St.Ack
+
yuzhihong@... 2012-01-16, 14:33
-
RE: ANN: The fourth hbase 0.92.0 release candidate is available for downloadRamkrishna.S.Vasudevan 2012-01-16, 15:54
Hi Stack
Are you planning for a RC4? If so just see if HBASE-5120 can go in?
Regards
Ram
-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
Sent: Monday, January 16, 2012 8:03 PM
To: [EMAIL PROTECTED]
Cc: HBase Dev List
Subject: Re: ANN: The fourth hbase 0.92.0 release candidate is available for
download
Thanks for the update, Stack.
Do you want to freeze patch submission to 0.92 branch until you cut rc4 ?
I think hbase-5153 and hbase-5179 are close to the quality of integration.
Cheers
On Jan 16, 2012, at 5:07 AM, Stack <[EMAIL PROTECTED]> wrote:
> On Fri, Jan 13, 2012 at 5:00 PM, Stack <[EMAIL PROTECTED]> wrote:
>> +1 on releasing this as 0.92.0
>>
>
> HBASE-5204 "Backward compatibility fixes for 0.92" looks like a
> blocker to me. Without HBASE-5204, asynchbase/opentsdb will not work
> on hbase 0.92.0. HBASE-5204 cannot go into a 0.92.1 because we would
> not be able to do a rolling restart going from 0.92.0 to 0.92.1 (See
> change in HbaseObjectWritable).
>
> St.Ack
Are you planning for a RC4? If so just see if HBASE-5120 can go in?
Regards
Ram
-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
Sent: Monday, January 16, 2012 8:03 PM
To: [EMAIL PROTECTED]
Cc: HBase Dev List
Subject: Re: ANN: The fourth hbase 0.92.0 release candidate is available for
download
Thanks for the update, Stack.
Do you want to freeze patch submission to 0.92 branch until you cut rc4 ?
I think hbase-5153 and hbase-5179 are close to the quality of integration.
Cheers
On Jan 16, 2012, at 5:07 AM, Stack <[EMAIL PROTECTED]> wrote:
> On Fri, Jan 13, 2012 at 5:00 PM, Stack <[EMAIL PROTECTED]> wrote:
>> +1 on releasing this as 0.92.0
>>
>
> HBASE-5204 "Backward compatibility fixes for 0.92" looks like a
> blocker to me. Without HBASE-5204, asynchbase/opentsdb will not work
> on hbase 0.92.0. HBASE-5204 cannot go into a 0.92.1 because we would
> not be able to do a rolling restart going from 0.92.0 to 0.92.1 (See
> change in HbaseObjectWritable).
>
> St.Ack
+
Ramkrishna.S.Vasudevan 2012-01-16, 15:54
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadStack 2012-01-16, 23:59
On Mon, Jan 16, 2012 at 7:54 AM, Ramkrishna.S.Vasudevan
<[EMAIL PROTECTED]> wrote:
> Hi Stack
>
> Are you planning for a RC4? If so just see if HBASE-5120 can go in?
>
See my note to Ted. We can put it into a 0.92.1.
Good on you Ram,
St.Ack
<[EMAIL PROTECTED]> wrote:
> Hi Stack
>
> Are you planning for a RC4? If so just see if HBASE-5120 can go in?
>
See my note to Ted. We can put it into a 0.92.1.
Good on you Ram,
St.Ack
+
Stack 2012-01-16, 23:59
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadStack 2012-01-16, 23:59
On Mon, Jan 16, 2012 at 6:33 AM, <[EMAIL PROTECTED]> wrote:
> Thanks for the update, Stack.
>
> Do you want to freeze patch submission to 0.92 branch until you cut rc4 ?
>
Its ok. I just tagged rc4. 0.92 branch is open again.
> I think hbase-5153 and hbase-5179 are close to the quality of integration.
>
I don't want to wait nor do I want to have rc4 too different from rc3.
RC3 almost made it. I'd like rc4 to get pulled out by rc3 good
feeling.
The above can go into a 0.92.1.
St.Ack
> Thanks for the update, Stack.
>
> Do you want to freeze patch submission to 0.92 branch until you cut rc4 ?
>
Its ok. I just tagged rc4. 0.92 branch is open again.
> I think hbase-5153 and hbase-5179 are close to the quality of integration.
>
I don't want to wait nor do I want to have rc4 too different from rc3.
RC3 almost made it. I'd like rc4 to get pulled out by rc3 good
feeling.
The above can go into a 0.92.1.
St.Ack
+
Stack 2012-01-16, 23:59
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadJonathan Hsieh 2012-01-17, 00:03
I lean towards minimizing the number of patches (maybe just this blocker?).
If we do that we could have an shorter window for this hopefully final
release candidate...
Jon
On Mon, Jan 16, 2012 at 3:59 PM, Stack <[EMAIL PROTECTED]> wrote:
> On Mon, Jan 16, 2012 at 6:33 AM, <[EMAIL PROTECTED]> wrote:
> > Thanks for the update, Stack.
> >
> > Do you want to freeze patch submission to 0.92 branch until you cut rc4 ?
> >
>
> Its ok. I just tagged rc4. 0.92 branch is open again.
>
> > I think hbase-5153 and hbase-5179 are close to the quality of
> integration.
> >
>
> I don't want to wait nor do I want to have rc4 too different from rc3.
> RC3 almost made it. I'd like rc4 to get pulled out by rc3 good
> feeling.
>
> The above can go into a 0.92.1.
>
> St.Ack
>
--
// Jonathan Hsieh (shay)
// Software Engineer, Cloudera
// [EMAIL PROTECTED]
If we do that we could have an shorter window for this hopefully final
release candidate...
Jon
On Mon, Jan 16, 2012 at 3:59 PM, Stack <[EMAIL PROTECTED]> wrote:
> On Mon, Jan 16, 2012 at 6:33 AM, <[EMAIL PROTECTED]> wrote:
> > Thanks for the update, Stack.
> >
> > Do you want to freeze patch submission to 0.92 branch until you cut rc4 ?
> >
>
> Its ok. I just tagged rc4. 0.92 branch is open again.
>
> > I think hbase-5153 and hbase-5179 are close to the quality of
> integration.
> >
>
> I don't want to wait nor do I want to have rc4 too different from rc3.
> RC3 almost made it. I'd like rc4 to get pulled out by rc3 good
> feeling.
>
> The above can go into a 0.92.1.
>
> St.Ack
>
--
// Jonathan Hsieh (shay)
// Software Engineer, Cloudera
// [EMAIL PROTECTED]
+
Jonathan Hsieh 2012-01-17, 00:03
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadTed Yu 2012-01-17, 00:08
I think we should keep HBASE-5121 in RC4.
As Stack acknowledged, 5121 is of importance.
Since Jan. 8th, there have been several bug fixes going into 0.92 branch.
RC4 should include them.
Cheers
On Mon, Jan 16, 2012 at 4:03 PM, Jonathan Hsieh <[EMAIL PROTECTED]> wrote:
> I lean towards minimizing the number of patches (maybe just this blocker?).
> If we do that we could have an shorter window for this hopefully final
> release candidate...
>
> Jon
>
> On Mon, Jan 16, 2012 at 3:59 PM, Stack <[EMAIL PROTECTED]> wrote:
>
> > On Mon, Jan 16, 2012 at 6:33 AM, <[EMAIL PROTECTED]> wrote:
> > > Thanks for the update, Stack.
> > >
> > > Do you want to freeze patch submission to 0.92 branch until you cut
> rc4 ?
> > >
> >
> > Its ok. I just tagged rc4. 0.92 branch is open again.
> >
> > > I think hbase-5153 and hbase-5179 are close to the quality of
> > integration.
> > >
> >
> > I don't want to wait nor do I want to have rc4 too different from rc3.
> > RC3 almost made it. I'd like rc4 to get pulled out by rc3 good
> > feeling.
> >
> > The above can go into a 0.92.1.
> >
> > St.Ack
> >
>
>
>
> --
> // Jonathan Hsieh (shay)
> // Software Engineer, Cloudera
> // [EMAIL PROTECTED]
>
As Stack acknowledged, 5121 is of importance.
Since Jan. 8th, there have been several bug fixes going into 0.92 branch.
RC4 should include them.
Cheers
On Mon, Jan 16, 2012 at 4:03 PM, Jonathan Hsieh <[EMAIL PROTECTED]> wrote:
> I lean towards minimizing the number of patches (maybe just this blocker?).
> If we do that we could have an shorter window for this hopefully final
> release candidate...
>
> Jon
>
> On Mon, Jan 16, 2012 at 3:59 PM, Stack <[EMAIL PROTECTED]> wrote:
>
> > On Mon, Jan 16, 2012 at 6:33 AM, <[EMAIL PROTECTED]> wrote:
> > > Thanks for the update, Stack.
> > >
> > > Do you want to freeze patch submission to 0.92 branch until you cut
> rc4 ?
> > >
> >
> > Its ok. I just tagged rc4. 0.92 branch is open again.
> >
> > > I think hbase-5153 and hbase-5179 are close to the quality of
> > integration.
> > >
> >
> > I don't want to wait nor do I want to have rc4 too different from rc3.
> > RC3 almost made it. I'd like rc4 to get pulled out by rc3 good
> > feeling.
> >
> > The above can go into a 0.92.1.
> >
> > St.Ack
> >
>
>
>
> --
> // Jonathan Hsieh (shay)
> // Software Engineer, Cloudera
> // [EMAIL PROTECTED]
>
+
Ted Yu 2012-01-17, 00:08
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadStack 2012-01-17, 00:46
On Mon, Jan 16, 2012 at 4:08 PM, Ted Yu <[EMAIL PROTECTED]> wrote:
> I think we should keep HBASE-5121 in RC4.
Yes. Small important change w/ a test. Otherwise, I am with Jon that
we keep diff between rc3 and rc4 small (I tagged the 0.92 branch an
hour or so ago so there are few other minors that made it in -- they
are listed above in this thread).
Waiting on successful jenkins security branch 0.92.0 build before
putting up RC (insecure built successfully).
St.Ack
> I think we should keep HBASE-5121 in RC4.
Yes. Small important change w/ a test. Otherwise, I am with Jon that
we keep diff between rc3 and rc4 small (I tagged the 0.92 branch an
hour or so ago so there are few other minors that made it in -- they
are listed above in this thread).
Waiting on successful jenkins security branch 0.92.0 build before
putting up RC (insecure built successfully).
St.Ack
+
Stack 2012-01-17, 00:46
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadAndrew Purtell 2012-01-16, 17:14
+1 concur
Changing my previous vote on the RC to -1.
Best regards,
- Andy
Problems worthy of attack prove their worth by hitting back. - Piet Hein (via Tom White)
>________________________________
> From: Stack <[EMAIL PROTECTED]>
>To: HBase Dev List <[EMAIL PROTECTED]>
>Sent: Monday, January 16, 2012 5:07 AM
>Subject: Re: ANN: The fourth hbase 0.92.0 release candidate is available for download
>
>On Fri, Jan 13, 2012 at 5:00 PM, Stack <[EMAIL PROTECTED]> wrote:
>> +1 on releasing this as 0.92.0
>>
>
>HBASE-5204 "Backward compatibility fixes for 0.92" looks like a
>blocker to me. Without HBASE-5204, asynchbase/opentsdb will not work
>on hbase 0.92.0. HBASE-5204 cannot go into a 0.92.1 because we would
>not be able to do a rolling restart going from 0.92.0 to 0.92.1 (See
>change in HbaseObjectWritable).
>
>St.Ack
>
>
>
Changing my previous vote on the RC to -1.
Best regards,
- Andy
Problems worthy of attack prove their worth by hitting back. - Piet Hein (via Tom White)
>________________________________
> From: Stack <[EMAIL PROTECTED]>
>To: HBase Dev List <[EMAIL PROTECTED]>
>Sent: Monday, January 16, 2012 5:07 AM
>Subject: Re: ANN: The fourth hbase 0.92.0 release candidate is available for download
>
>On Fri, Jan 13, 2012 at 5:00 PM, Stack <[EMAIL PROTECTED]> wrote:
>> +1 on releasing this as 0.92.0
>>
>
>HBASE-5204 "Backward compatibility fixes for 0.92" looks like a
>blocker to me. Without HBASE-5204, asynchbase/opentsdb will not work
>on hbase 0.92.0. HBASE-5204 cannot go into a 0.92.1 because we would
>not be able to do a rolling restart going from 0.92.0 to 0.92.1 (See
>change in HbaseObjectWritable).
>
>St.Ack
>
>
>
+
Andrew Purtell 2012-01-16, 17:14
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadJean-Daniel Cryans 2012-01-14, 01:10
+1
I've been testing this under heavy write loads and I also looked at
replication. On previous RCs I did many read tests and AFAIK that
didn't change so I didn't go through that again. I think this is a
solid release, and 0.92.1 will take care of bugs that didn't make it
in time.
J-D
On Sun, Jan 8, 2012 at 1:39 PM, Stack <[EMAIL PROTECTED]> wrote:
> The fourth hbase 0.92.0 release candidate is available for download:
>
> http://people.apache.org/~stack/hbase-0.92.0-candidate-3/
>
> I've posted a secure and an insecure tarball built by jenkins.
>
> This RC bundles hadoop 1.0.0 and zookeeper 3.4.2.
>
> As said previous, HBase 0.92.0 includes a wagon-load of new features
> including coprocessors, security, a new (self-migrating) file format,
> distributed log splitting, etc. There's been > 660 fixes since 0.90.
> See the list here: http://su.pr/1VZzl5
>
> See the hbase manual for the low-down on what hadoop version this
> release will run on: http://hbase.apache.org/book.html#hadoop
>
> There is no migration necessary. A shutdown and restart after putting
> in place the new HBase should be all thats involved (A rolling restart
> from 0.90.x to 0.92.0 will not work). That said, once you have moved
> to 0.92.x, there is no going back to 0.90.x after the transition has
> been made.
>
> Should we release this candidate as hbase 0.92.0? Take it for a spin.
> Check out the doc. Vote +1/-1 by Martin Luther King Day, January 16th, 2012.
>
> Yours,
> The HBasistas
>
> P.S. This candidate is looking pretty good. If you are thinking of using 0.92.0
> any time soon, we would suggest that you dig in now and try it. Thanks.
I've been testing this under heavy write loads and I also looked at
replication. On previous RCs I did many read tests and AFAIK that
didn't change so I didn't go through that again. I think this is a
solid release, and 0.92.1 will take care of bugs that didn't make it
in time.
J-D
On Sun, Jan 8, 2012 at 1:39 PM, Stack <[EMAIL PROTECTED]> wrote:
> The fourth hbase 0.92.0 release candidate is available for download:
>
> http://people.apache.org/~stack/hbase-0.92.0-candidate-3/
>
> I've posted a secure and an insecure tarball built by jenkins.
>
> This RC bundles hadoop 1.0.0 and zookeeper 3.4.2.
>
> As said previous, HBase 0.92.0 includes a wagon-load of new features
> including coprocessors, security, a new (self-migrating) file format,
> distributed log splitting, etc. There's been > 660 fixes since 0.90.
> See the list here: http://su.pr/1VZzl5
>
> See the hbase manual for the low-down on what hadoop version this
> release will run on: http://hbase.apache.org/book.html#hadoop
>
> There is no migration necessary. A shutdown and restart after putting
> in place the new HBase should be all thats involved (A rolling restart
> from 0.90.x to 0.92.0 will not work). That said, once you have moved
> to 0.92.x, there is no going back to 0.90.x after the transition has
> been made.
>
> Should we release this candidate as hbase 0.92.0? Take it for a spin.
> Check out the doc. Vote +1/-1 by Martin Luther King Day, January 16th, 2012.
>
> Yours,
> The HBasistas
>
> P.S. This candidate is looking pretty good. If you are thinking of using 0.92.0
> any time soon, we would suggest that you dig in now and try it. Thanks.
+
Jean-Daniel Cryans 2012-01-14, 01:10
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadTed Yu 2012-01-14, 01:15
Since last Sunday, 6 checkins have gone in under 0.92.0 section in
CHANGES.txt
I am fine with re-categorizing them to go under 0.92.1
Or should we evaluate the 6 checkins ? I personally think the current 0.92
Jenkins build would be more robust than RC3.
Cheers
On Fri, Jan 13, 2012 at 5:10 PM, Jean-Daniel Cryans <[EMAIL PROTECTED]>wrote:
> +1
>
> I've been testing this under heavy write loads and I also looked at
> replication. On previous RCs I did many read tests and AFAIK that
> didn't change so I didn't go through that again. I think this is a
> solid release, and 0.92.1 will take care of bugs that didn't make it
> in time.
>
> J-D
>
> On Sun, Jan 8, 2012 at 1:39 PM, Stack <[EMAIL PROTECTED]> wrote:
> > The fourth hbase 0.92.0 release candidate is available for download:
> >
> > http://people.apache.org/~stack/hbase-0.92.0-candidate-3/
> >
> > I've posted a secure and an insecure tarball built by jenkins.
> >
> > This RC bundles hadoop 1.0.0 and zookeeper 3.4.2.
> >
> > As said previous, HBase 0.92.0 includes a wagon-load of new features
> > including coprocessors, security, a new (self-migrating) file format,
> > distributed log splitting, etc. There's been > 660 fixes since 0.90.
> > See the list here: http://su.pr/1VZzl5
> >
> > See the hbase manual for the low-down on what hadoop version this
> > release will run on: http://hbase.apache.org/book.html#hadoop
> >
> > There is no migration necessary. A shutdown and restart after putting
> > in place the new HBase should be all thats involved (A rolling restart
> > from 0.90.x to 0.92.0 will not work). That said, once you have moved
> > to 0.92.x, there is no going back to 0.90.x after the transition has
> > been made.
> >
> > Should we release this candidate as hbase 0.92.0? Take it for a spin.
> > Check out the doc. Vote +1/-1 by Martin Luther King Day, January 16th,
> 2012.
> >
> > Yours,
> > The HBasistas
> >
> > P.S. This candidate is looking pretty good. If you are thinking of
> using 0.92.0
> > any time soon, we would suggest that you dig in now and try it. Thanks.
>
CHANGES.txt
I am fine with re-categorizing them to go under 0.92.1
Or should we evaluate the 6 checkins ? I personally think the current 0.92
Jenkins build would be more robust than RC3.
Cheers
On Fri, Jan 13, 2012 at 5:10 PM, Jean-Daniel Cryans <[EMAIL PROTECTED]>wrote:
> +1
>
> I've been testing this under heavy write loads and I also looked at
> replication. On previous RCs I did many read tests and AFAIK that
> didn't change so I didn't go through that again. I think this is a
> solid release, and 0.92.1 will take care of bugs that didn't make it
> in time.
>
> J-D
>
> On Sun, Jan 8, 2012 at 1:39 PM, Stack <[EMAIL PROTECTED]> wrote:
> > The fourth hbase 0.92.0 release candidate is available for download:
> >
> > http://people.apache.org/~stack/hbase-0.92.0-candidate-3/
> >
> > I've posted a secure and an insecure tarball built by jenkins.
> >
> > This RC bundles hadoop 1.0.0 and zookeeper 3.4.2.
> >
> > As said previous, HBase 0.92.0 includes a wagon-load of new features
> > including coprocessors, security, a new (self-migrating) file format,
> > distributed log splitting, etc. There's been > 660 fixes since 0.90.
> > See the list here: http://su.pr/1VZzl5
> >
> > See the hbase manual for the low-down on what hadoop version this
> > release will run on: http://hbase.apache.org/book.html#hadoop
> >
> > There is no migration necessary. A shutdown and restart after putting
> > in place the new HBase should be all thats involved (A rolling restart
> > from 0.90.x to 0.92.0 will not work). That said, once you have moved
> > to 0.92.x, there is no going back to 0.90.x after the transition has
> > been made.
> >
> > Should we release this candidate as hbase 0.92.0? Take it for a spin.
> > Check out the doc. Vote +1/-1 by Martin Luther King Day, January 16th,
> 2012.
> >
> > Yours,
> > The HBasistas
> >
> > P.S. This candidate is looking pretty good. If you are thinking of
> using 0.92.0
> > any time soon, we would suggest that you dig in now and try it. Thanks.
>
+
Ted Yu 2012-01-14, 01:15
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadJean-Daniel Cryans 2012-01-14, 01:21
Are all 6 required? If not, which ones are and how important are they
in your opinion so that they need to be in 0.92.0? If haven't looked
over them in too great details but you seem in the know so that's why
I ask #lazyfriday.
Regarding their "fix version", they should have been committed under
0.92.1 AFAIK (and changed if there's a new RC) but no biggie.
J-D
On Fri, Jan 13, 2012 at 5:15 PM, Ted Yu <[EMAIL PROTECTED]> wrote:
> Since last Sunday, 6 checkins have gone in under 0.92.0 section in
> CHANGES.txt
>
> I am fine with re-categorizing them to go under 0.92.1
>
> Or should we evaluate the 6 checkins ? I personally think the current 0.92
> Jenkins build would be more robust than RC3.
>
> Cheers
>
> On Fri, Jan 13, 2012 at 5:10 PM, Jean-Daniel Cryans <[EMAIL PROTECTED]>wrote:
>
>> +1
>>
>> I've been testing this under heavy write loads and I also looked at
>> replication. On previous RCs I did many read tests and AFAIK that
>> didn't change so I didn't go through that again. I think this is a
>> solid release, and 0.92.1 will take care of bugs that didn't make it
>> in time.
>>
>> J-D
>>
>> On Sun, Jan 8, 2012 at 1:39 PM, Stack <[EMAIL PROTECTED]> wrote:
>> > The fourth hbase 0.92.0 release candidate is available for download:
>> >
>> > http://people.apache.org/~stack/hbase-0.92.0-candidate-3/
>> >
>> > I've posted a secure and an insecure tarball built by jenkins.
>> >
>> > This RC bundles hadoop 1.0.0 and zookeeper 3.4.2.
>> >
>> > As said previous, HBase 0.92.0 includes a wagon-load of new features
>> > including coprocessors, security, a new (self-migrating) file format,
>> > distributed log splitting, etc. There's been > 660 fixes since 0.90.
>> > See the list here: http://su.pr/1VZzl5
>> >
>> > See the hbase manual for the low-down on what hadoop version this
>> > release will run on: http://hbase.apache.org/book.html#hadoop
>> >
>> > There is no migration necessary. A shutdown and restart after putting
>> > in place the new HBase should be all thats involved (A rolling restart
>> > from 0.90.x to 0.92.0 will not work). That said, once you have moved
>> > to 0.92.x, there is no going back to 0.90.x after the transition has
>> > been made.
>> >
>> > Should we release this candidate as hbase 0.92.0? Take it for a spin.
>> > Check out the doc. Vote +1/-1 by Martin Luther King Day, January 16th,
>> 2012.
>> >
>> > Yours,
>> > The HBasistas
>> >
>> > P.S. This candidate is looking pretty good. If you are thinking of
>> using 0.92.0
>> > any time soon, we would suggest that you dig in now and try it. Thanks.
>>
in your opinion so that they need to be in 0.92.0? If haven't looked
over them in too great details but you seem in the know so that's why
I ask #lazyfriday.
Regarding their "fix version", they should have been committed under
0.92.1 AFAIK (and changed if there's a new RC) but no biggie.
J-D
On Fri, Jan 13, 2012 at 5:15 PM, Ted Yu <[EMAIL PROTECTED]> wrote:
> Since last Sunday, 6 checkins have gone in under 0.92.0 section in
> CHANGES.txt
>
> I am fine with re-categorizing them to go under 0.92.1
>
> Or should we evaluate the 6 checkins ? I personally think the current 0.92
> Jenkins build would be more robust than RC3.
>
> Cheers
>
> On Fri, Jan 13, 2012 at 5:10 PM, Jean-Daniel Cryans <[EMAIL PROTECTED]>wrote:
>
>> +1
>>
>> I've been testing this under heavy write loads and I also looked at
>> replication. On previous RCs I did many read tests and AFAIK that
>> didn't change so I didn't go through that again. I think this is a
>> solid release, and 0.92.1 will take care of bugs that didn't make it
>> in time.
>>
>> J-D
>>
>> On Sun, Jan 8, 2012 at 1:39 PM, Stack <[EMAIL PROTECTED]> wrote:
>> > The fourth hbase 0.92.0 release candidate is available for download:
>> >
>> > http://people.apache.org/~stack/hbase-0.92.0-candidate-3/
>> >
>> > I've posted a secure and an insecure tarball built by jenkins.
>> >
>> > This RC bundles hadoop 1.0.0 and zookeeper 3.4.2.
>> >
>> > As said previous, HBase 0.92.0 includes a wagon-load of new features
>> > including coprocessors, security, a new (self-migrating) file format,
>> > distributed log splitting, etc. There's been > 660 fixes since 0.90.
>> > See the list here: http://su.pr/1VZzl5
>> >
>> > See the hbase manual for the low-down on what hadoop version this
>> > release will run on: http://hbase.apache.org/book.html#hadoop
>> >
>> > There is no migration necessary. A shutdown and restart after putting
>> > in place the new HBase should be all thats involved (A rolling restart
>> > from 0.90.x to 0.92.0 will not work). That said, once you have moved
>> > to 0.92.x, there is no going back to 0.90.x after the transition has
>> > been made.
>> >
>> > Should we release this candidate as hbase 0.92.0? Take it for a spin.
>> > Check out the doc. Vote +1/-1 by Martin Luther King Day, January 16th,
>> 2012.
>> >
>> > Yours,
>> > The HBasistas
>> >
>> > P.S. This candidate is looking pretty good. If you are thinking of
>> using 0.92.0
>> > any time soon, we would suggest that you dig in now and try it. Thanks.
>>
+
Jean-Daniel Cryans 2012-01-14, 01:21
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadTed Yu 2012-01-14, 01:27
All are important correctness fixes, except for HBASE-5163 which would make
0.92 Jenkins build more stable - considering the build takes around two
hours.
Maybe we can declare Jenkins build #243 which is ongoing as 0.92.0 ?
Assuming it passes.
Just my two cents.
On Fri, Jan 13, 2012 at 5:21 PM, Jean-Daniel Cryans <[EMAIL PROTECTED]>wrote:
> Are all 6 required? If not, which ones are and how important are they
> in your opinion so that they need to be in 0.92.0? If haven't looked
> over them in too great details but you seem in the know so that's why
> I ask #lazyfriday.
>
> Regarding their "fix version", they should have been committed under
> 0.92.1 AFAIK (and changed if there's a new RC) but no biggie.
>
> J-D
>
> On Fri, Jan 13, 2012 at 5:15 PM, Ted Yu <[EMAIL PROTECTED]> wrote:
> > Since last Sunday, 6 checkins have gone in under 0.92.0 section in
> > CHANGES.txt
> >
> > I am fine with re-categorizing them to go under 0.92.1
> >
> > Or should we evaluate the 6 checkins ? I personally think the current
> 0.92
> > Jenkins build would be more robust than RC3.
> >
> > Cheers
> >
> > On Fri, Jan 13, 2012 at 5:10 PM, Jean-Daniel Cryans <[EMAIL PROTECTED]
> >wrote:
> >
> >> +1
> >>
> >> I've been testing this under heavy write loads and I also looked at
> >> replication. On previous RCs I did many read tests and AFAIK that
> >> didn't change so I didn't go through that again. I think this is a
> >> solid release, and 0.92.1 will take care of bugs that didn't make it
> >> in time.
> >>
> >> J-D
> >>
> >> On Sun, Jan 8, 2012 at 1:39 PM, Stack <[EMAIL PROTECTED]> wrote:
> >> > The fourth hbase 0.92.0 release candidate is available for download:
> >> >
> >> > http://people.apache.org/~stack/hbase-0.92.0-candidate-3/
> >> >
> >> > I've posted a secure and an insecure tarball built by jenkins.
> >> >
> >> > This RC bundles hadoop 1.0.0 and zookeeper 3.4.2.
> >> >
> >> > As said previous, HBase 0.92.0 includes a wagon-load of new features
> >> > including coprocessors, security, a new (self-migrating) file format,
> >> > distributed log splitting, etc. There's been > 660 fixes since 0.90.
> >> > See the list here: http://su.pr/1VZzl5
> >> >
> >> > See the hbase manual for the low-down on what hadoop version this
> >> > release will run on: http://hbase.apache.org/book.html#hadoop
> >> >
> >> > There is no migration necessary. A shutdown and restart after putting
> >> > in place the new HBase should be all thats involved (A rolling restart
> >> > from 0.90.x to 0.92.0 will not work). That said, once you have moved
> >> > to 0.92.x, there is no going back to 0.90.x after the transition has
> >> > been made.
> >> >
> >> > Should we release this candidate as hbase 0.92.0? Take it for a spin.
> >> > Check out the doc. Vote +1/-1 by Martin Luther King Day, January
> 16th,
> >> 2012.
> >> >
> >> > Yours,
> >> > The HBasistas
> >> >
> >> > P.S. This candidate is looking pretty good. If you are thinking of
> >> using 0.92.0
> >> > any time soon, we would suggest that you dig in now and try it.
> Thanks.
> >>
>
0.92 Jenkins build more stable - considering the build takes around two
hours.
Maybe we can declare Jenkins build #243 which is ongoing as 0.92.0 ?
Assuming it passes.
Just my two cents.
On Fri, Jan 13, 2012 at 5:21 PM, Jean-Daniel Cryans <[EMAIL PROTECTED]>wrote:
> Are all 6 required? If not, which ones are and how important are they
> in your opinion so that they need to be in 0.92.0? If haven't looked
> over them in too great details but you seem in the know so that's why
> I ask #lazyfriday.
>
> Regarding their "fix version", they should have been committed under
> 0.92.1 AFAIK (and changed if there's a new RC) but no biggie.
>
> J-D
>
> On Fri, Jan 13, 2012 at 5:15 PM, Ted Yu <[EMAIL PROTECTED]> wrote:
> > Since last Sunday, 6 checkins have gone in under 0.92.0 section in
> > CHANGES.txt
> >
> > I am fine with re-categorizing them to go under 0.92.1
> >
> > Or should we evaluate the 6 checkins ? I personally think the current
> 0.92
> > Jenkins build would be more robust than RC3.
> >
> > Cheers
> >
> > On Fri, Jan 13, 2012 at 5:10 PM, Jean-Daniel Cryans <[EMAIL PROTECTED]
> >wrote:
> >
> >> +1
> >>
> >> I've been testing this under heavy write loads and I also looked at
> >> replication. On previous RCs I did many read tests and AFAIK that
> >> didn't change so I didn't go through that again. I think this is a
> >> solid release, and 0.92.1 will take care of bugs that didn't make it
> >> in time.
> >>
> >> J-D
> >>
> >> On Sun, Jan 8, 2012 at 1:39 PM, Stack <[EMAIL PROTECTED]> wrote:
> >> > The fourth hbase 0.92.0 release candidate is available for download:
> >> >
> >> > http://people.apache.org/~stack/hbase-0.92.0-candidate-3/
> >> >
> >> > I've posted a secure and an insecure tarball built by jenkins.
> >> >
> >> > This RC bundles hadoop 1.0.0 and zookeeper 3.4.2.
> >> >
> >> > As said previous, HBase 0.92.0 includes a wagon-load of new features
> >> > including coprocessors, security, a new (self-migrating) file format,
> >> > distributed log splitting, etc. There's been > 660 fixes since 0.90.
> >> > See the list here: http://su.pr/1VZzl5
> >> >
> >> > See the hbase manual for the low-down on what hadoop version this
> >> > release will run on: http://hbase.apache.org/book.html#hadoop
> >> >
> >> > There is no migration necessary. A shutdown and restart after putting
> >> > in place the new HBase should be all thats involved (A rolling restart
> >> > from 0.90.x to 0.92.0 will not work). That said, once you have moved
> >> > to 0.92.x, there is no going back to 0.90.x after the transition has
> >> > been made.
> >> >
> >> > Should we release this candidate as hbase 0.92.0? Take it for a spin.
> >> > Check out the doc. Vote +1/-1 by Martin Luther King Day, January
> 16th,
> >> 2012.
> >> >
> >> > Yours,
> >> > The HBasistas
> >> >
> >> > P.S. This candidate is looking pretty good. If you are thinking of
> >> using 0.92.0
> >> > any time soon, we would suggest that you dig in now and try it.
> Thanks.
> >>
>
+
Ted Yu 2012-01-14, 01:27
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadStack 2012-01-14, 05:23
On Fri, Jan 13, 2012 at 5:27 PM, Ted Yu <[EMAIL PROTECTED]> wrote:
> Maybe we can declare Jenkins build #243 which is ongoing as 0.92.0 ?
>
Thats not how it works. RC3 is what we are voting on. +1 or -1 it.
If -1, it should be a good reason.
Regards the issues committed since RC3 was put up last Sunday, none
are marked critical or even blocker.
Is this the set (I searched all resolved since last Sunday)?
HBASE-5143 HBASE-3274 Fix config typo in pluggable load balancer factory
HBASE-5163 TestLogRolling#testLogRollOnDatanodeDeath fails sometimes
on Jenkins or hadoop QA ("The directory is already locked.")
HBASE-5088 A concurrency issue on SoftValueSortedMap
HBASE-5137 MasterFileSystem.splitLog() should abort even if
waitOnSafeMode() throws IOException
HBASE-5041 Major compaction on non existing table does not throw error
None of the above merit sinking the release in my opinion (The first
is minor, the second an occasional test failure, the third and fourth
something we've had for ages and rare is their incidence and the fifth
is an improvement).
St.Ack
> Maybe we can declare Jenkins build #243 which is ongoing as 0.92.0 ?
>
Thats not how it works. RC3 is what we are voting on. +1 or -1 it.
If -1, it should be a good reason.
Regards the issues committed since RC3 was put up last Sunday, none
are marked critical or even blocker.
Is this the set (I searched all resolved since last Sunday)?
HBASE-5143 HBASE-3274 Fix config typo in pluggable load balancer factory
HBASE-5163 TestLogRolling#testLogRollOnDatanodeDeath fails sometimes
on Jenkins or hadoop QA ("The directory is already locked.")
HBASE-5088 A concurrency issue on SoftValueSortedMap
HBASE-5137 MasterFileSystem.splitLog() should abort even if
waitOnSafeMode() throws IOException
HBASE-5041 Major compaction on non existing table does not throw error
None of the above merit sinking the release in my opinion (The first
is minor, the second an occasional test failure, the third and fourth
something we've had for ages and rare is their incidence and the fifth
is an improvement).
St.Ack
+
Stack 2012-01-14, 05:23
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadTed Yu 2012-01-14, 05:35
The following went in after last Sunday:
HBASE-5196 Failure in region split after PONR could cause region hole
HBASE-5121 MajorCompaction may affect scan's correctness
HBASE-5152 Region is on service before completing initialization when
doing rollback of split,
it will affect read correctness
All are bug fixes.
Cheers
On Fri, Jan 13, 2012 at 9:23 PM, Stack <[EMAIL PROTECTED]> wrote:
> On Fri, Jan 13, 2012 at 5:27 PM, Ted Yu <[EMAIL PROTECTED]> wrote:
> > Maybe we can declare Jenkins build #243 which is ongoing as 0.92.0 ?
> >
>
> Thats not how it works. RC3 is what we are voting on. +1 or -1 it.
> If -1, it should be a good reason.
>
> Regards the issues committed since RC3 was put up last Sunday, none
> are marked critical or even blocker.
>
> Is this the set (I searched all resolved since last Sunday)?
>
> HBASE-5143 HBASE-3274 Fix config typo in pluggable load balancer
> factory
> HBASE-5163 TestLogRolling#testLogRollOnDatanodeDeath fails
> sometimes
> on Jenkins or hadoop QA ("The directory is already locked.")
> HBASE-5088 A concurrency issue on SoftValueSortedMap
> HBASE-5137 MasterFileSystem.splitLog() should abort even if
> waitOnSafeMode() throws IOException
> HBASE-5041 Major compaction on non existing table does not
> throw error
>
> None of the above merit sinking the release in my opinion (The first
> is minor, the second an occasional test failure, the third and fourth
> something we've had for ages and rare is their incidence and the fifth
> is an improvement).
>
> St.Ack
>
HBASE-5196 Failure in region split after PONR could cause region hole
HBASE-5121 MajorCompaction may affect scan's correctness
HBASE-5152 Region is on service before completing initialization when
doing rollback of split,
it will affect read correctness
All are bug fixes.
Cheers
On Fri, Jan 13, 2012 at 9:23 PM, Stack <[EMAIL PROTECTED]> wrote:
> On Fri, Jan 13, 2012 at 5:27 PM, Ted Yu <[EMAIL PROTECTED]> wrote:
> > Maybe we can declare Jenkins build #243 which is ongoing as 0.92.0 ?
> >
>
> Thats not how it works. RC3 is what we are voting on. +1 or -1 it.
> If -1, it should be a good reason.
>
> Regards the issues committed since RC3 was put up last Sunday, none
> are marked critical or even blocker.
>
> Is this the set (I searched all resolved since last Sunday)?
>
> HBASE-5143 HBASE-3274 Fix config typo in pluggable load balancer
> factory
> HBASE-5163 TestLogRolling#testLogRollOnDatanodeDeath fails
> sometimes
> on Jenkins or hadoop QA ("The directory is already locked.")
> HBASE-5088 A concurrency issue on SoftValueSortedMap
> HBASE-5137 MasterFileSystem.splitLog() should abort even if
> waitOnSafeMode() throws IOException
> HBASE-5041 Major compaction on non existing table does not
> throw error
>
> None of the above merit sinking the release in my opinion (The first
> is minor, the second an occasional test failure, the third and fourth
> something we've had for ages and rare is their incidence and the fifth
> is an improvement).
>
> St.Ack
>
+
Ted Yu 2012-01-14, 05:35
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadlars hofhansl 2012-01-14, 05:47
My $0.02:
Let's get 0.92.0 out the door (these all are problems that exist in 0.90 as well).
Then do a 0.92.1 relatively soon.
-- Lars
----- Original Message -----
From: Ted Yu <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
Cc:
Sent: Friday, January 13, 2012 9:35 PM
Subject: Re: ANN: The fourth hbase 0.92.0 release candidate is available for download
The following went in after last Sunday:
HBASE-5196 Failure in region split after PONR could cause region hole
HBASE-5121 MajorCompaction may affect scan's correctness
HBASE-5152 Region is on service before completing initialization when
doing rollback of split,
it will affect read correctness
All are bug fixes.
Cheers
On Fri, Jan 13, 2012 at 9:23 PM, Stack <[EMAIL PROTECTED]> wrote:
> On Fri, Jan 13, 2012 at 5:27 PM, Ted Yu <[EMAIL PROTECTED]> wrote:
> > Maybe we can declare Jenkins build #243 which is ongoing as 0.92.0 ?
> >
>
> Thats not how it works. RC3 is what we are voting on. +1 or -1 it.
> If -1, it should be a good reason.
>
> Regards the issues committed since RC3 was put up last Sunday, none
> are marked critical or even blocker.
>
> Is this the set (I searched all resolved since last Sunday)?
>
> HBASE-5143 HBASE-3274 Fix config typo in pluggable load balancer
> factory
> HBASE-5163 TestLogRolling#testLogRollOnDatanodeDeath fails
> sometimes
> on Jenkins or hadoop QA ("The directory is already locked.")
> HBASE-5088 A concurrency issue on SoftValueSortedMap
> HBASE-5137 MasterFileSystem.splitLog() should abort even if
> waitOnSafeMode() throws IOException
> HBASE-5041 Major compaction on non existing table does not
> throw error
>
> None of the above merit sinking the release in my opinion (The first
> is minor, the second an occasional test failure, the third and fourth
> something we've had for ages and rare is their incidence and the fifth
> is an improvement).
>
> St.Ack
>
Let's get 0.92.0 out the door (these all are problems that exist in 0.90 as well).
Then do a 0.92.1 relatively soon.
-- Lars
----- Original Message -----
From: Ted Yu <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
Cc:
Sent: Friday, January 13, 2012 9:35 PM
Subject: Re: ANN: The fourth hbase 0.92.0 release candidate is available for download
The following went in after last Sunday:
HBASE-5196 Failure in region split after PONR could cause region hole
HBASE-5121 MajorCompaction may affect scan's correctness
HBASE-5152 Region is on service before completing initialization when
doing rollback of split,
it will affect read correctness
All are bug fixes.
Cheers
On Fri, Jan 13, 2012 at 9:23 PM, Stack <[EMAIL PROTECTED]> wrote:
> On Fri, Jan 13, 2012 at 5:27 PM, Ted Yu <[EMAIL PROTECTED]> wrote:
> > Maybe we can declare Jenkins build #243 which is ongoing as 0.92.0 ?
> >
>
> Thats not how it works. RC3 is what we are voting on. +1 or -1 it.
> If -1, it should be a good reason.
>
> Regards the issues committed since RC3 was put up last Sunday, none
> are marked critical or even blocker.
>
> Is this the set (I searched all resolved since last Sunday)?
>
> HBASE-5143 HBASE-3274 Fix config typo in pluggable load balancer
> factory
> HBASE-5163 TestLogRolling#testLogRollOnDatanodeDeath fails
> sometimes
> on Jenkins or hadoop QA ("The directory is already locked.")
> HBASE-5088 A concurrency issue on SoftValueSortedMap
> HBASE-5137 MasterFileSystem.splitLog() should abort even if
> waitOnSafeMode() throws IOException
> HBASE-5041 Major compaction on non existing table does not
> throw error
>
> None of the above merit sinking the release in my opinion (The first
> is minor, the second an occasional test failure, the third and fourth
> something we've had for ages and rare is their incidence and the fifth
> is an improvement).
>
> St.Ack
>
+
lars hofhansl 2012-01-14, 05:47
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadStack 2012-01-14, 06:16
On Fri, Jan 13, 2012 at 9:47 PM, lars hofhansl <[EMAIL PROTECTED]> wrote:
> My $0.02:
>
> Let's get 0.92.0 out the door (these all are problems that exist in 0.90 as well).
> Then do a 0.92.1 relatively soon.
>
Are you +1'ing RC3 Lars (smile)?
St.Ack
> My $0.02:
>
> Let's get 0.92.0 out the door (these all are problems that exist in 0.90 as well).
> Then do a 0.92.1 relatively soon.
>
Are you +1'ing RC3 Lars (smile)?
St.Ack
+
Stack 2012-01-14, 06:16
-
RE: ANN: The fourth hbase 0.92.0 release candidate is available for downloadRamkrishna.S.Vasudevan 2012-01-14, 06:41
Hi Stack
AS TM is any way not going to be included in 0.92.0 I am +1 on it.
If you can take a look at HBASE-5120 once and see if we can take it in. No
problem
Even if it is taken into 0.92.1. (smile).
As Lars said we can do a 0.92.1 sooner.
Regards
Ram
-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] On Behalf Of Stack
Sent: Saturday, January 14, 2012 11:46 AM
To: [EMAIL PROTECTED]; lars hofhansl
Subject: Re: ANN: The fourth hbase 0.92.0 release candidate is available for
download
On Fri, Jan 13, 2012 at 9:47 PM, lars hofhansl <[EMAIL PROTECTED]> wrote:
> My $0.02:
>
> Let's get 0.92.0 out the door (these all are problems that exist in 0.90
as well).
> Then do a 0.92.1 relatively soon.
>
Are you +1'ing RC3 Lars (smile)?
St.Ack
AS TM is any way not going to be included in 0.92.0 I am +1 on it.
If you can take a look at HBASE-5120 once and see if we can take it in. No
problem
Even if it is taken into 0.92.1. (smile).
As Lars said we can do a 0.92.1 sooner.
Regards
Ram
-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] On Behalf Of Stack
Sent: Saturday, January 14, 2012 11:46 AM
To: [EMAIL PROTECTED]; lars hofhansl
Subject: Re: ANN: The fourth hbase 0.92.0 release candidate is available for
download
On Fri, Jan 13, 2012 at 9:47 PM, lars hofhansl <[EMAIL PROTECTED]> wrote:
> My $0.02:
>
> Let's get 0.92.0 out the door (these all are problems that exist in 0.90
as well).
> Then do a 0.92.1 relatively soon.
>
Are you +1'ing RC3 Lars (smile)?
St.Ack
+
Ramkrishna.S.Vasudevan 2012-01-14, 06:41
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadlars hofhansl 2012-01-14, 22:09
+1
I have been following all the jiras.Took it for a quick spin:
Created some table, inserted some data, set some delete markers, tested TTL with MIN_VERSIONS, disabled/droped some tables, etc.
Also visually inspected the content of the tar file.
All looks good.
I noticed one small strange thing. Upon first startup ROOT and META are opened, then closed, then opened again.
This is repeatable if the HBase data directory is removed between restarts.
Sat Jan 14 13:59:23 PST 2012 Initializing region .META.,,1.1028785192 COMPLETE (since 2sec ago) Region opened successfully (since 2sec ago)
Sat Jan 14 13:59:22 PST 2012 Initializing region -ROOT-,,0.70236052 COMPLETE (since 3sec ago) Region opened successfully (since 3sec ago)
Sat Jan 14 13:59:16 PST 2012 Closing region .META.,,1.1028785192 COMPLETE (since 10sec ago) Closed (since 10sec ago)
Sat Jan 14 13:59:16 PST 2012 Closing region -ROOT-,,0.70236052 COMPLETE (since 10sec ago) Closed (since 10sec ago)
Sat Jan 14 13:59:16 PST 2012 Initializing region .META.,,1.1028785192 COMPLETE (since 10sec ago) Region opened successfully (since 10sec ago)
Sat Jan 14 13:59:16 PST 2012 Initializing region -ROOT-,,0.70236052 COMPLETE (since 10sec ago) Region opened successfully (since 10sec ago)
Sat Jan 14 13:59:15 PST 2012 Master startup COMPLETE (since 2sec ago) Initialization successful (since 2sec ago)
-- Lars
----- Original Message -----
From: Stack <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]; lars hofhansl <[EMAIL PROTECTED]>
Cc:
Sent: Friday, January 13, 2012 10:16 PM
Subject: Re: ANN: The fourth hbase 0.92.0 release candidate is available for download
On Fri, Jan 13, 2012 at 9:47 PM, lars hofhansl <[EMAIL PROTECTED]> wrote:
> My $0.02:
>
> Let's get 0.92.0 out the door (these all are problems that exist in 0.90 as well).
> Then do a 0.92.1 relatively soon.
>
Are you +1'ing RC3 Lars (smile)?
St.Ack
I have been following all the jiras.Took it for a quick spin:
Created some table, inserted some data, set some delete markers, tested TTL with MIN_VERSIONS, disabled/droped some tables, etc.
Also visually inspected the content of the tar file.
All looks good.
I noticed one small strange thing. Upon first startup ROOT and META are opened, then closed, then opened again.
This is repeatable if the HBase data directory is removed between restarts.
Sat Jan 14 13:59:23 PST 2012 Initializing region .META.,,1.1028785192 COMPLETE (since 2sec ago) Region opened successfully (since 2sec ago)
Sat Jan 14 13:59:22 PST 2012 Initializing region -ROOT-,,0.70236052 COMPLETE (since 3sec ago) Region opened successfully (since 3sec ago)
Sat Jan 14 13:59:16 PST 2012 Closing region .META.,,1.1028785192 COMPLETE (since 10sec ago) Closed (since 10sec ago)
Sat Jan 14 13:59:16 PST 2012 Closing region -ROOT-,,0.70236052 COMPLETE (since 10sec ago) Closed (since 10sec ago)
Sat Jan 14 13:59:16 PST 2012 Initializing region .META.,,1.1028785192 COMPLETE (since 10sec ago) Region opened successfully (since 10sec ago)
Sat Jan 14 13:59:16 PST 2012 Initializing region -ROOT-,,0.70236052 COMPLETE (since 10sec ago) Region opened successfully (since 10sec ago)
Sat Jan 14 13:59:15 PST 2012 Master startup COMPLETE (since 2sec ago) Initialization successful (since 2sec ago)
-- Lars
----- Original Message -----
From: Stack <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]; lars hofhansl <[EMAIL PROTECTED]>
Cc:
Sent: Friday, January 13, 2012 10:16 PM
Subject: Re: ANN: The fourth hbase 0.92.0 release candidate is available for download
On Fri, Jan 13, 2012 at 9:47 PM, lars hofhansl <[EMAIL PROTECTED]> wrote:
> My $0.02:
>
> Let's get 0.92.0 out the door (these all are problems that exist in 0.90 as well).
> Then do a 0.92.1 relatively soon.
>
Are you +1'ing RC3 Lars (smile)?
St.Ack
+
lars hofhansl 2012-01-14, 22:09
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadStack 2012-01-14, 06:15
On Fri, Jan 13, 2012 at 9:35 PM, Ted Yu <[EMAIL PROTECTED]> wrote:
> The following went in after last Sunday:
>
None are blockers.
> HBASE-5196 Failure in region split after PONR could cause region hole
The above should be rare in the field. From Jimmy, "There are only 3
threads to do the clean up. If there are lots of (most in the cluster)
region servers died, the shutdown handler may stuck in log splitting
for quite sometime. During this period, if the master died somehow, it
won't be able to finish the clean up."
> HBASE-5121 MajorCompaction may affect scan's correctness
This is the most severe of the issues committed since last Sunday but
not enough to sink the RC IMO.
> HBASE-5152 Region is on service before completing initialization when
> doing rollback of split, it will affect read correctness
Not severe enough of an issue to sink the RC IMO.
St.Ack
> The following went in after last Sunday:
>
None are blockers.
> HBASE-5196 Failure in region split after PONR could cause region hole
The above should be rare in the field. From Jimmy, "There are only 3
threads to do the clean up. If there are lots of (most in the cluster)
region servers died, the shutdown handler may stuck in log splitting
for quite sometime. During this period, if the master died somehow, it
won't be able to finish the clean up."
> HBASE-5121 MajorCompaction may affect scan's correctness
This is the most severe of the issues committed since last Sunday but
not enough to sink the RC IMO.
> HBASE-5152 Region is on service before completing initialization when
> doing rollback of split, it will affect read correctness
Not severe enough of an issue to sink the RC IMO.
St.Ack
+
Stack 2012-01-14, 06:15
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadTed Yu 2012-01-14, 11:15
Without HBASE-5121, user wouldn't notice incorrect scan result under
certain scenario.
Wouldn't that be a blocker ?
Cheers
On Fri, Jan 13, 2012 at 10:15 PM, Stack <[EMAIL PROTECTED]> wrote:
> On Fri, Jan 13, 2012 at 9:35 PM, Ted Yu <[EMAIL PROTECTED]> wrote:
> > The following went in after last Sunday:
> >
>
> None are blockers.
>
>
> > HBASE-5196 Failure in region split after PONR could cause region hole
>
>
> The above should be rare in the field. From Jimmy, "There are only 3
> threads to do the clean up. If there are lots of (most in the cluster)
> region servers died, the shutdown handler may stuck in log splitting
> for quite sometime. During this period, if the master died somehow, it
> won't be able to finish the clean up."
>
>
> > HBASE-5121 MajorCompaction may affect scan's correctness
>
>
> This is the most severe of the issues committed since last Sunday but
> not enough to sink the RC IMO.
>
>
> > HBASE-5152 Region is on service before completing initialization when
> > doing rollback of split, it will affect read correctness
>
> Not severe enough of an issue to sink the RC IMO.
>
> St.Ack
>
certain scenario.
Wouldn't that be a blocker ?
Cheers
On Fri, Jan 13, 2012 at 10:15 PM, Stack <[EMAIL PROTECTED]> wrote:
> On Fri, Jan 13, 2012 at 9:35 PM, Ted Yu <[EMAIL PROTECTED]> wrote:
> > The following went in after last Sunday:
> >
>
> None are blockers.
>
>
> > HBASE-5196 Failure in region split after PONR could cause region hole
>
>
> The above should be rare in the field. From Jimmy, "There are only 3
> threads to do the clean up. If there are lots of (most in the cluster)
> region servers died, the shutdown handler may stuck in log splitting
> for quite sometime. During this period, if the master died somehow, it
> won't be able to finish the clean up."
>
>
> > HBASE-5121 MajorCompaction may affect scan's correctness
>
>
> This is the most severe of the issues committed since last Sunday but
> not enough to sink the RC IMO.
>
>
> > HBASE-5152 Region is on service before completing initialization when
> > doing rollback of split, it will affect read correctness
>
> Not severe enough of an issue to sink the RC IMO.
>
> St.Ack
>
+
Ted Yu 2012-01-14, 11:15
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadStack 2012-01-15, 02:42
On Sat, Jan 14, 2012 at 3:15 AM, Ted Yu <[EMAIL PROTECTED]> wrote:
> Without HBASE-5121, user wouldn't notice incorrect scan result under
> certain scenario.
> Wouldn't that be a blocker ?
>
It could be.
Lets make it a blocker on 0.92.1 and let 0.92.0 out; there's a years
worth of work in 0.92.0 and it needs to see the light of day. I can
put up a candidate 0.92.1 a week or two after the 0.92.0 release. At
our shop we're going to deploy 0.92.0 to production and as part of
that process will likely find a few other bugs that can should go into
a 0.92.1.
St.Ack
> Without HBASE-5121, user wouldn't notice incorrect scan result under
> certain scenario.
> Wouldn't that be a blocker ?
>
It could be.
Lets make it a blocker on 0.92.1 and let 0.92.0 out; there's a years
worth of work in 0.92.0 and it needs to see the light of day. I can
put up a candidate 0.92.1 a week or two after the 0.92.0 release. At
our shop we're going to deploy 0.92.0 to production and as part of
that process will likely find a few other bugs that can should go into
a 0.92.1.
St.Ack
+
Stack 2012-01-15, 02:42
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadyuzhihong@... 2012-01-15, 03:05
Sounds good.
We just need to adjust the Fix Version for a few Jiras.
Cheers
On Jan 14, 2012, at 6:42 PM, Stack <[EMAIL PROTECTED]> wrote:
> On Sat, Jan 14, 2012 at 3:15 AM, Ted Yu <[EMAIL PROTECTED]> wrote:
>> Without HBASE-5121, user wouldn't notice incorrect scan result under
>> certain scenario.
>> Wouldn't that be a blocker ?
>>
>
> It could be.
>
> Lets make it a blocker on 0.92.1 and let 0.92.0 out; there's a years
> worth of work in 0.92.0 and it needs to see the light of day. I can
> put up a candidate 0.92.1 a week or two after the 0.92.0 release. At
> our shop we're going to deploy 0.92.0 to production and as part of
> that process will likely find a few other bugs that can should go into
> a 0.92.1.
>
> St.Ack
We just need to adjust the Fix Version for a few Jiras.
Cheers
On Jan 14, 2012, at 6:42 PM, Stack <[EMAIL PROTECTED]> wrote:
> On Sat, Jan 14, 2012 at 3:15 AM, Ted Yu <[EMAIL PROTECTED]> wrote:
>> Without HBASE-5121, user wouldn't notice incorrect scan result under
>> certain scenario.
>> Wouldn't that be a blocker ?
>>
>
> It could be.
>
> Lets make it a blocker on 0.92.1 and let 0.92.0 out; there's a years
> worth of work in 0.92.0 and it needs to see the light of day. I can
> put up a candidate 0.92.1 a week or two after the 0.92.0 release. At
> our shop we're going to deploy 0.92.0 to production and as part of
> that process will likely find a few other bugs that can should go into
> a 0.92.1.
>
> St.Ack
+
yuzhihong@... 2012-01-15, 03:05
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadStack 2012-01-15, 03:42
On Sat, Jan 14, 2012 at 7:05 PM, <[EMAIL PROTECTED]> wrote:
> Sounds good.
>
> We just need to adjust the Fix Version for a few Jiras.
>
I can do that np.
St.Ack
> Sounds good.
>
> We just need to adjust the Fix Version for a few Jiras.
>
I can do that np.
St.Ack
+
Stack 2012-01-15, 03:42
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadAndrew Purtell 2012-01-14, 17:45
+1
I filed a bug yesterday that is a correctness problem for security but we ought to tag that module as alpha quality anyway like we did with Stargate in the beginning. Like that earlier case I think it should have that designation until there is more than one successful deploy of it, it was a big port of changes from an internal code base to a divergent upstream. The fix for the issue I mention is ready to go in for 0.92.1. I may find others. We don't use 0.92 in production (yet) but since over the holidays I have been occasionally running the RCs up on EC2.
Best regards,
- Andy
On Jan 13, 2012, at 9:23 PM, Stack <[EMAIL PROTECTED]> wrote:
> On Fri, Jan 13, 2012 at 5:27 PM, Ted Yu <[EMAIL PROTECTED]> wrote:
>> Maybe we can declare Jenkins build #243 which is ongoing as 0.92.0 ?
>>
>
> Thats not how it works. RC3 is what we are voting on. +1 or -1 it.
> If -1, it should be a good reason.
>
> Regards the issues committed since RC3 was put up last Sunday, none
> are marked critical or even blocker.
>
> Is this the set (I searched all resolved since last Sunday)?
>
> HBASE-5143 HBASE-3274 Fix config typo in pluggable load balancer factory
> HBASE-5163 TestLogRolling#testLogRollOnDatanodeDeath fails sometimes
> on Jenkins or hadoop QA ("The directory is already locked.")
> HBASE-5088 A concurrency issue on SoftValueSortedMap
> HBASE-5137 MasterFileSystem.splitLog() should abort even if
> waitOnSafeMode() throws IOException
> HBASE-5041 Major compaction on non existing table does not throw error
>
> None of the above merit sinking the release in my opinion (The first
> is minor, the second an occasional test failure, the third and fourth
> something we've had for ages and rare is their incidence and the fifth
> is an improvement).
>
> St.Ack
I filed a bug yesterday that is a correctness problem for security but we ought to tag that module as alpha quality anyway like we did with Stargate in the beginning. Like that earlier case I think it should have that designation until there is more than one successful deploy of it, it was a big port of changes from an internal code base to a divergent upstream. The fix for the issue I mention is ready to go in for 0.92.1. I may find others. We don't use 0.92 in production (yet) but since over the holidays I have been occasionally running the RCs up on EC2.
Best regards,
- Andy
On Jan 13, 2012, at 9:23 PM, Stack <[EMAIL PROTECTED]> wrote:
> On Fri, Jan 13, 2012 at 5:27 PM, Ted Yu <[EMAIL PROTECTED]> wrote:
>> Maybe we can declare Jenkins build #243 which is ongoing as 0.92.0 ?
>>
>
> Thats not how it works. RC3 is what we are voting on. +1 or -1 it.
> If -1, it should be a good reason.
>
> Regards the issues committed since RC3 was put up last Sunday, none
> are marked critical or even blocker.
>
> Is this the set (I searched all resolved since last Sunday)?
>
> HBASE-5143 HBASE-3274 Fix config typo in pluggable load balancer factory
> HBASE-5163 TestLogRolling#testLogRollOnDatanodeDeath fails sometimes
> on Jenkins or hadoop QA ("The directory is already locked.")
> HBASE-5088 A concurrency issue on SoftValueSortedMap
> HBASE-5137 MasterFileSystem.splitLog() should abort even if
> waitOnSafeMode() throws IOException
> HBASE-5041 Major compaction on non existing table does not throw error
>
> None of the above merit sinking the release in my opinion (The first
> is minor, the second an occasional test failure, the third and fourth
> something we've had for ages and rare is their incidence and the fifth
> is an improvement).
>
> St.Ack
+
Andrew Purtell 2012-01-14, 17:45
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadJonathan Hsieh 2012-01-14, 18:11
+1.
This has been in RC an state for quite a while now and has been looping
some of the Bigtop system tests for a long time. I'm going to echo Lars,
Andrew, etc about getting it out and then getting a 0.92.1 out soon after.
The bugs Jimmy and I have found and been diagnosing don't seem to be
regressions from 0.90.x.
Jon.
On Sun, Jan 8, 2012 at 1:39 PM, Stack <[EMAIL PROTECTED]> wrote:
> The fourth hbase 0.92.0 release candidate is available for download:
>
> http://people.apache.org/~stack/hbase-0.92.0-candidate-3/
>
> I've posted a secure and an insecure tarball built by jenkins.
>
> This RC bundles hadoop 1.0.0 and zookeeper 3.4.2.
>
> As said previous, HBase 0.92.0 includes a wagon-load of new features
> including coprocessors, security, a new (self-migrating) file format,
> distributed log splitting, etc. There's been > 660 fixes since 0.90.
> See the list here: http://su.pr/1VZzl5
>
> See the hbase manual for the low-down on what hadoop version this
> release will run on: http://hbase.apache.org/book.html#hadoop
>
> There is no migration necessary. A shutdown and restart after putting
> in place the new HBase should be all thats involved (A rolling restart
> from 0.90.x to 0.92.0 will not work). That said, once you have moved
> to 0.92.x, there is no going back to 0.90.x after the transition has
> been made.
>
> Should we release this candidate as hbase 0.92.0? Take it for a spin.
> Check out the doc. Vote +1/-1 by Martin Luther King Day, January 16th,
> 2012.
>
> Yours,
> The HBasistas
>
> P.S. This candidate is looking pretty good. If you are thinking of using
> 0.92.0
> any time soon, we would suggest that you dig in now and try it. Thanks.
>
--
// Jonathan Hsieh (shay)
// Software Engineer, Cloudera
// [EMAIL PROTECTED]
This has been in RC an state for quite a while now and has been looping
some of the Bigtop system tests for a long time. I'm going to echo Lars,
Andrew, etc about getting it out and then getting a 0.92.1 out soon after.
The bugs Jimmy and I have found and been diagnosing don't seem to be
regressions from 0.90.x.
Jon.
On Sun, Jan 8, 2012 at 1:39 PM, Stack <[EMAIL PROTECTED]> wrote:
> The fourth hbase 0.92.0 release candidate is available for download:
>
> http://people.apache.org/~stack/hbase-0.92.0-candidate-3/
>
> I've posted a secure and an insecure tarball built by jenkins.
>
> This RC bundles hadoop 1.0.0 and zookeeper 3.4.2.
>
> As said previous, HBase 0.92.0 includes a wagon-load of new features
> including coprocessors, security, a new (self-migrating) file format,
> distributed log splitting, etc. There's been > 660 fixes since 0.90.
> See the list here: http://su.pr/1VZzl5
>
> See the hbase manual for the low-down on what hadoop version this
> release will run on: http://hbase.apache.org/book.html#hadoop
>
> There is no migration necessary. A shutdown and restart after putting
> in place the new HBase should be all thats involved (A rolling restart
> from 0.90.x to 0.92.0 will not work). That said, once you have moved
> to 0.92.x, there is no going back to 0.90.x after the transition has
> been made.
>
> Should we release this candidate as hbase 0.92.0? Take it for a spin.
> Check out the doc. Vote +1/-1 by Martin Luther King Day, January 16th,
> 2012.
>
> Yours,
> The HBasistas
>
> P.S. This candidate is looking pretty good. If you are thinking of using
> 0.92.0
> any time soon, we would suggest that you dig in now and try it. Thanks.
>
--
// Jonathan Hsieh (shay)
// Software Engineer, Cloudera
// [EMAIL PROTECTED]
+
Jonathan Hsieh 2012-01-14, 18:11
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadRoman Shaposhnik 2012-01-14, 23:59
On Sun, Jan 8, 2012 at 1:39 PM, Stack <[EMAIL PROTECTED]> wrote:
> The fourth hbase 0.92.0 release candidate is available for download:
>
> http://people.apache.org/~stack/hbase-0.92.0-candidate-3/
+1. Unfortunately Amazon has decided to swallow Bigtop's Jenkin VM
and EBS that was backing it so, I can't really post a link with the test
results. But all the usual suspects were passing.
Thanks,
Roman.
> The fourth hbase 0.92.0 release candidate is available for download:
>
> http://people.apache.org/~stack/hbase-0.92.0-candidate-3/
+1. Unfortunately Amazon has decided to swallow Bigtop's Jenkin VM
and EBS that was backing it so, I can't really post a link with the test
results. But all the usual suspects were passing.
Thanks,
Roman.
+
Roman Shaposhnik 2012-01-14, 23:59
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadLars George 2012-01-15, 13:31
I checked the MD5 and SHA-1 hashes, which are all OK, I did get an error though on my MBA (see below). Assuming this is only me I am also +1.
Here is the test failure:
...
Running org.apache.hadoop.hbase.zookeeper.TestZooKeeperACL
2012-01-15 01:52:46.002 java[49123:1903] Unable to load realm info from SCDynamicStore
2012-01-15 01:52:48.715 java[49123:1e807] Unable to load realm info from SCDynamicStore
Tests run: 5, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 44.246 sec
Running org.apache.hadoop.hbase.zookeeper.TestZooKeeperMainServerArg
Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.376 sec
Running org.apache.hadoop.hbase.zookeeper.TestZooKeeperNodeTracker
Tests run: 2, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 1.129 sec
Results :
Failed tests: testCyclicReplication(org.apache.hadoop.hbase.replication.TestMasterReplication): Waited too much time for put replication
Tests in error:
testSimplePutDelete(org.apache.hadoop.hbase.replication.TestMasterReplication): Cluster already running at /Users/larsgeorge/Downloads/hbase-0.92.0-rc3/hbase-0.92.0/target/test-data/7956128b-c9e3-4473-b3a2-c2faf140fb99/dfscluster_80ddefc2-e1be-477f-b761-df1ed883fc31
Tests run: 1037, Failures: 1, Errors: 1, Skipped: 8
[INFO] ------------------------------------------------------------------------
[INFO] BUILD FAILURE
[INFO] ------------------------------------------------------------------------
[INFO] Total time: 2:57:25.455s
[INFO] Finished at: Sun Jan 15 01:53:33 CET 2012
[INFO] Final Memory: 108M/213M
[INFO] ------------------------------------------------------------------------
[WARNING] The requested profile "runAllTests" could not be activated because it does not exist.
[ERROR] Failed to execute goal org.apache.maven.plugins:maven-surefire-plugin:2.10:test (default-test) on project hbase: Failure or timeout -> [Help 1]
[ERROR]
[ERROR] To see the full stack trace of the errors, re-run Maven with the -e switch.
[ERROR] Re-run Maven using the -X switch to enable full debug logging.
[ERROR]
[ERROR] For more information about the errors and possible solutions, please read the following articles:
[ERROR] [Help 1] http://cwiki.apache.org/confluence/display/MAVEN/MojoExecutionException
Did I run this wrong? What about the "runAllTests" warning above?
Thanks,
Lars
On Jan 8, 2012, at 10:39 PM, Stack wrote:
> The fourth hbase 0.92.0 release candidate is available for download:
>
> http://people.apache.org/~stack/hbase-0.92.0-candidate-3/
>
> I've posted a secure and an insecure tarball built by jenkins.
>
> This RC bundles hadoop 1.0.0 and zookeeper 3.4.2.
>
> As said previous, HBase 0.92.0 includes a wagon-load of new features
> including coprocessors, security, a new (self-migrating) file format,
> distributed log splitting, etc. There's been > 660 fixes since 0.90.
> See the list here: http://su.pr/1VZzl5
>
> See the hbase manual for the low-down on what hadoop version this
> release will run on: http://hbase.apache.org/book.html#hadoop
>
> There is no migration necessary. A shutdown and restart after putting
> in place the new HBase should be all thats involved (A rolling restart
> from 0.90.x to 0.92.0 will not work). That said, once you have moved
> to 0.92.x, there is no going back to 0.90.x after the transition has
> been made.
>
> Should we release this candidate as hbase 0.92.0? Take it for a spin.
> Check out the doc. Vote +1/-1 by Martin Luther King Day, January 16th, 2012.
>
> Yours,
> The HBasistas
>
> P.S. This candidate is looking pretty good. If you are thinking of using 0.92.0
> any time soon, we would suggest that you dig in now and try it. Thanks.
Here is the test failure:
...
Running org.apache.hadoop.hbase.zookeeper.TestZooKeeperACL
2012-01-15 01:52:46.002 java[49123:1903] Unable to load realm info from SCDynamicStore
2012-01-15 01:52:48.715 java[49123:1e807] Unable to load realm info from SCDynamicStore
Tests run: 5, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 44.246 sec
Running org.apache.hadoop.hbase.zookeeper.TestZooKeeperMainServerArg
Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.376 sec
Running org.apache.hadoop.hbase.zookeeper.TestZooKeeperNodeTracker
Tests run: 2, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 1.129 sec
Results :
Failed tests: testCyclicReplication(org.apache.hadoop.hbase.replication.TestMasterReplication): Waited too much time for put replication
Tests in error:
testSimplePutDelete(org.apache.hadoop.hbase.replication.TestMasterReplication): Cluster already running at /Users/larsgeorge/Downloads/hbase-0.92.0-rc3/hbase-0.92.0/target/test-data/7956128b-c9e3-4473-b3a2-c2faf140fb99/dfscluster_80ddefc2-e1be-477f-b761-df1ed883fc31
Tests run: 1037, Failures: 1, Errors: 1, Skipped: 8
[INFO] ------------------------------------------------------------------------
[INFO] BUILD FAILURE
[INFO] ------------------------------------------------------------------------
[INFO] Total time: 2:57:25.455s
[INFO] Finished at: Sun Jan 15 01:53:33 CET 2012
[INFO] Final Memory: 108M/213M
[INFO] ------------------------------------------------------------------------
[WARNING] The requested profile "runAllTests" could not be activated because it does not exist.
[ERROR] Failed to execute goal org.apache.maven.plugins:maven-surefire-plugin:2.10:test (default-test) on project hbase: Failure or timeout -> [Help 1]
[ERROR]
[ERROR] To see the full stack trace of the errors, re-run Maven with the -e switch.
[ERROR] Re-run Maven using the -X switch to enable full debug logging.
[ERROR]
[ERROR] For more information about the errors and possible solutions, please read the following articles:
[ERROR] [Help 1] http://cwiki.apache.org/confluence/display/MAVEN/MojoExecutionException
Did I run this wrong? What about the "runAllTests" warning above?
Thanks,
Lars
On Jan 8, 2012, at 10:39 PM, Stack wrote:
> The fourth hbase 0.92.0 release candidate is available for download:
>
> http://people.apache.org/~stack/hbase-0.92.0-candidate-3/
>
> I've posted a secure and an insecure tarball built by jenkins.
>
> This RC bundles hadoop 1.0.0 and zookeeper 3.4.2.
>
> As said previous, HBase 0.92.0 includes a wagon-load of new features
> including coprocessors, security, a new (self-migrating) file format,
> distributed log splitting, etc. There's been > 660 fixes since 0.90.
> See the list here: http://su.pr/1VZzl5
>
> See the hbase manual for the low-down on what hadoop version this
> release will run on: http://hbase.apache.org/book.html#hadoop
>
> There is no migration necessary. A shutdown and restart after putting
> in place the new HBase should be all thats involved (A rolling restart
> from 0.90.x to 0.92.0 will not work). That said, once you have moved
> to 0.92.x, there is no going back to 0.90.x after the transition has
> been made.
>
> Should we release this candidate as hbase 0.92.0? Take it for a spin.
> Check out the doc. Vote +1/-1 by Martin Luther King Day, January 16th, 2012.
>
> Yours,
> The HBasistas
>
> P.S. This candidate is looking pretty good. If you are thinking of using 0.92.0
> any time soon, we would suggest that you dig in now and try it. Thanks.
+
Lars George 2012-01-15, 13:31
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadTed Yu 2012-01-15, 15:13
Lars:
runAllTests profile is only in TRUNK.
'mvn clean test' should be the command for 0.92
Cheers
On Sun, Jan 15, 2012 at 5:31 AM, Lars George <[EMAIL PROTECTED]> wrote:
> I checked the MD5 and SHA-1 hashes, which are all OK, I did get an error
> though on my MBA (see below). Assuming this is only me I am also +1.
>
> Here is the test failure:
>
> ...
> Running org.apache.hadoop.hbase.zookeeper.TestZooKeeperACL
> 2012-01-15 01:52:46.002 java[49123:1903] Unable to load realm info from
> SCDynamicStore
> 2012-01-15 01:52:48.715 java[49123:1e807] Unable to load realm info from
> SCDynamicStore
> Tests run: 5, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 44.246 sec
> Running org.apache.hadoop.hbase.zookeeper.TestZooKeeperMainServerArg
> Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.376 sec
> Running org.apache.hadoop.hbase.zookeeper.TestZooKeeperNodeTracker
> Tests run: 2, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 1.129 sec
>
> Results :
>
> Failed tests:
> testCyclicReplication(org.apache.hadoop.hbase.replication.TestMasterReplication):
> Waited too much time for put replication
>
> Tests in error:
> testSimplePutDelete(org.apache.hadoop.hbase.replication.TestMasterReplication):
> Cluster already running at
> /Users/larsgeorge/Downloads/hbase-0.92.0-rc3/hbase-0.92.0/target/test-data/7956128b-c9e3-4473-b3a2-c2faf140fb99/dfscluster_80ddefc2-e1be-477f-b761-df1ed883fc31
>
> Tests run: 1037, Failures: 1, Errors: 1, Skipped: 8
>
> [INFO]
> ------------------------------------------------------------------------
> [INFO] BUILD FAILURE
> [INFO]
> ------------------------------------------------------------------------
> [INFO] Total time: 2:57:25.455s
> [INFO] Finished at: Sun Jan 15 01:53:33 CET 2012
> [INFO] Final Memory: 108M/213M
> [INFO]
> ------------------------------------------------------------------------
> [WARNING] The requested profile "runAllTests" could not be activated
> because it does not exist.
> [ERROR] Failed to execute goal
> org.apache.maven.plugins:maven-surefire-plugin:2.10:test (default-test) on
> project hbase: Failure or timeout -> [Help 1]
> [ERROR]
> [ERROR] To see the full stack trace of the errors, re-run Maven with the
> -e switch.
> [ERROR] Re-run Maven using the -X switch to enable full debug logging.
> [ERROR]
> [ERROR] For more information about the errors and possible solutions,
> please read the following articles:
> [ERROR] [Help 1]
> http://cwiki.apache.org/confluence/display/MAVEN/MojoExecutionException
>
> Did I run this wrong? What about the "runAllTests" warning above?
>
> Thanks,
> Lars
>
> On Jan 8, 2012, at 10:39 PM, Stack wrote:
>
> > The fourth hbase 0.92.0 release candidate is available for download:
> >
> > http://people.apache.org/~stack/hbase-0.92.0-candidate-3/
> >
> > I've posted a secure and an insecure tarball built by jenkins.
> >
> > This RC bundles hadoop 1.0.0 and zookeeper 3.4.2.
> >
> > As said previous, HBase 0.92.0 includes a wagon-load of new features
> > including coprocessors, security, a new (self-migrating) file format,
> > distributed log splitting, etc. There's been > 660 fixes since 0.90.
> > See the list here: http://su.pr/1VZzl5
> >
> > See the hbase manual for the low-down on what hadoop version this
> > release will run on: http://hbase.apache.org/book.html#hadoop
> >
> > There is no migration necessary. A shutdown and restart after putting
> > in place the new HBase should be all thats involved (A rolling restart
> > from 0.90.x to 0.92.0 will not work). That said, once you have moved
> > to 0.92.x, there is no going back to 0.90.x after the transition has
> > been made.
> >
> > Should we release this candidate as hbase 0.92.0? Take it for a spin.
> > Check out the doc. Vote +1/-1 by Martin Luther King Day, January 16th,
> 2012.
> >
> > Yours,
> > The HBasistas
> >
> > P.S. This candidate is looking pretty good. If you are thinking of
> using 0.92.0
> > any time soon, we would suggest that you dig in now and try it. Thanks.
runAllTests profile is only in TRUNK.
'mvn clean test' should be the command for 0.92
Cheers
On Sun, Jan 15, 2012 at 5:31 AM, Lars George <[EMAIL PROTECTED]> wrote:
> I checked the MD5 and SHA-1 hashes, which are all OK, I did get an error
> though on my MBA (see below). Assuming this is only me I am also +1.
>
> Here is the test failure:
>
> ...
> Running org.apache.hadoop.hbase.zookeeper.TestZooKeeperACL
> 2012-01-15 01:52:46.002 java[49123:1903] Unable to load realm info from
> SCDynamicStore
> 2012-01-15 01:52:48.715 java[49123:1e807] Unable to load realm info from
> SCDynamicStore
> Tests run: 5, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 44.246 sec
> Running org.apache.hadoop.hbase.zookeeper.TestZooKeeperMainServerArg
> Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.376 sec
> Running org.apache.hadoop.hbase.zookeeper.TestZooKeeperNodeTracker
> Tests run: 2, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 1.129 sec
>
> Results :
>
> Failed tests:
> testCyclicReplication(org.apache.hadoop.hbase.replication.TestMasterReplication):
> Waited too much time for put replication
>
> Tests in error:
> testSimplePutDelete(org.apache.hadoop.hbase.replication.TestMasterReplication):
> Cluster already running at
> /Users/larsgeorge/Downloads/hbase-0.92.0-rc3/hbase-0.92.0/target/test-data/7956128b-c9e3-4473-b3a2-c2faf140fb99/dfscluster_80ddefc2-e1be-477f-b761-df1ed883fc31
>
> Tests run: 1037, Failures: 1, Errors: 1, Skipped: 8
>
> [INFO]
> ------------------------------------------------------------------------
> [INFO] BUILD FAILURE
> [INFO]
> ------------------------------------------------------------------------
> [INFO] Total time: 2:57:25.455s
> [INFO] Finished at: Sun Jan 15 01:53:33 CET 2012
> [INFO] Final Memory: 108M/213M
> [INFO]
> ------------------------------------------------------------------------
> [WARNING] The requested profile "runAllTests" could not be activated
> because it does not exist.
> [ERROR] Failed to execute goal
> org.apache.maven.plugins:maven-surefire-plugin:2.10:test (default-test) on
> project hbase: Failure or timeout -> [Help 1]
> [ERROR]
> [ERROR] To see the full stack trace of the errors, re-run Maven with the
> -e switch.
> [ERROR] Re-run Maven using the -X switch to enable full debug logging.
> [ERROR]
> [ERROR] For more information about the errors and possible solutions,
> please read the following articles:
> [ERROR] [Help 1]
> http://cwiki.apache.org/confluence/display/MAVEN/MojoExecutionException
>
> Did I run this wrong? What about the "runAllTests" warning above?
>
> Thanks,
> Lars
>
> On Jan 8, 2012, at 10:39 PM, Stack wrote:
>
> > The fourth hbase 0.92.0 release candidate is available for download:
> >
> > http://people.apache.org/~stack/hbase-0.92.0-candidate-3/
> >
> > I've posted a secure and an insecure tarball built by jenkins.
> >
> > This RC bundles hadoop 1.0.0 and zookeeper 3.4.2.
> >
> > As said previous, HBase 0.92.0 includes a wagon-load of new features
> > including coprocessors, security, a new (self-migrating) file format,
> > distributed log splitting, etc. There's been > 660 fixes since 0.90.
> > See the list here: http://su.pr/1VZzl5
> >
> > See the hbase manual for the low-down on what hadoop version this
> > release will run on: http://hbase.apache.org/book.html#hadoop
> >
> > There is no migration necessary. A shutdown and restart after putting
> > in place the new HBase should be all thats involved (A rolling restart
> > from 0.90.x to 0.92.0 will not work). That said, once you have moved
> > to 0.92.x, there is no going back to 0.90.x after the transition has
> > been made.
> >
> > Should we release this candidate as hbase 0.92.0? Take it for a spin.
> > Check out the doc. Vote +1/-1 by Martin Luther King Day, January 16th,
> 2012.
> >
> > Yours,
> > The HBasistas
> >
> > P.S. This candidate is looking pretty good. If you are thinking of
> using 0.92.0
> > any time soon, we would suggest that you dig in now and try it. Thanks.
+
Ted Yu 2012-01-15, 15:13
-
Re: ANN: The fourth hbase 0.92.0 release candidate is available for downloadLars George 2012-01-16, 03:01
Ah, course. Well, it ran the test ignoring the profile anyways, and failed. But as I said, that is the first time I ran it on my MacBook Air and not sure if that needs tweaking, Let me run it again, just to confirm.
On Jan 15, 2012, at 10:13 AM, Ted Yu wrote:
> Lars:
> runAllTests profile is only in TRUNK.
> 'mvn clean test' should be the command for 0.92
>
> Cheers
>
> On Sun, Jan 15, 2012 at 5:31 AM, Lars George <[EMAIL PROTECTED]> wrote:
>
>> I checked the MD5 and SHA-1 hashes, which are all OK, I did get an error
>> though on my MBA (see below). Assuming this is only me I am also +1.
>>
>> Here is the test failure:
>>
>> ...
>> Running org.apache.hadoop.hbase.zookeeper.TestZooKeeperACL
>> 2012-01-15 01:52:46.002 java[49123:1903] Unable to load realm info from
>> SCDynamicStore
>> 2012-01-15 01:52:48.715 java[49123:1e807] Unable to load realm info from
>> SCDynamicStore
>> Tests run: 5, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 44.246 sec
>> Running org.apache.hadoop.hbase.zookeeper.TestZooKeeperMainServerArg
>> Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.376 sec
>> Running org.apache.hadoop.hbase.zookeeper.TestZooKeeperNodeTracker
>> Tests run: 2, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 1.129 sec
>>
>> Results :
>>
>> Failed tests:
>> testCyclicReplication(org.apache.hadoop.hbase.replication.TestMasterReplication):
>> Waited too much time for put replication
>>
>> Tests in error:
>> testSimplePutDelete(org.apache.hadoop.hbase.replication.TestMasterReplication):
>> Cluster already running at
>> /Users/larsgeorge/Downloads/hbase-0.92.0-rc3/hbase-0.92.0/target/test-data/7956128b-c9e3-4473-b3a2-c2faf140fb99/dfscluster_80ddefc2-e1be-477f-b761-df1ed883fc31
>>
>> Tests run: 1037, Failures: 1, Errors: 1, Skipped: 8
>>
>> [INFO]
>> ------------------------------------------------------------------------
>> [INFO] BUILD FAILURE
>> [INFO]
>> ------------------------------------------------------------------------
>> [INFO] Total time: 2:57:25.455s
>> [INFO] Finished at: Sun Jan 15 01:53:33 CET 2012
>> [INFO] Final Memory: 108M/213M
>> [INFO]
>> ------------------------------------------------------------------------
>> [WARNING] The requested profile "runAllTests" could not be activated
>> because it does not exist.
>> [ERROR] Failed to execute goal
>> org.apache.maven.plugins:maven-surefire-plugin:2.10:test (default-test) on
>> project hbase: Failure or timeout -> [Help 1]
>> [ERROR]
>> [ERROR] To see the full stack trace of the errors, re-run Maven with the
>> -e switch.
>> [ERROR] Re-run Maven using the -X switch to enable full debug logging.
>> [ERROR]
>> [ERROR] For more information about the errors and possible solutions,
>> please read the following articles:
>> [ERROR] [Help 1]
>> http://cwiki.apache.org/confluence/display/MAVEN/MojoExecutionException
>>
>> Did I run this wrong? What about the "runAllTests" warning above?
>>
>> Thanks,
>> Lars
>>
>> On Jan 8, 2012, at 10:39 PM, Stack wrote:
>>
>>> The fourth hbase 0.92.0 release candidate is available for download:
>>>
>>> http://people.apache.org/~stack/hbase-0.92.0-candidate-3/
>>>
>>> I've posted a secure and an insecure tarball built by jenkins.
>>>
>>> This RC bundles hadoop 1.0.0 and zookeeper 3.4.2.
>>>
>>> As said previous, HBase 0.92.0 includes a wagon-load of new features
>>> including coprocessors, security, a new (self-migrating) file format,
>>> distributed log splitting, etc. There's been > 660 fixes since 0.90.
>>> See the list here: http://su.pr/1VZzl5
>>>
>>> See the hbase manual for the low-down on what hadoop version this
>>> release will run on: http://hbase.apache.org/book.html#hadoop
>>>
>>> There is no migration necessary. A shutdown and restart after putting
>>> in place the new HBase should be all thats involved (A rolling restart
>>> from 0.90.x to 0.92.0 will not work). That said, once you have moved
>>> to 0.92.x, there is no going back to 0.90.x after the transition has
>>> been made.
On Jan 15, 2012, at 10:13 AM, Ted Yu wrote:
> Lars:
> runAllTests profile is only in TRUNK.
> 'mvn clean test' should be the command for 0.92
>
> Cheers
>
> On Sun, Jan 15, 2012 at 5:31 AM, Lars George <[EMAIL PROTECTED]> wrote:
>
>> I checked the MD5 and SHA-1 hashes, which are all OK, I did get an error
>> though on my MBA (see below). Assuming this is only me I am also +1.
>>
>> Here is the test failure:
>>
>> ...
>> Running org.apache.hadoop.hbase.zookeeper.TestZooKeeperACL
>> 2012-01-15 01:52:46.002 java[49123:1903] Unable to load realm info from
>> SCDynamicStore
>> 2012-01-15 01:52:48.715 java[49123:1e807] Unable to load realm info from
>> SCDynamicStore
>> Tests run: 5, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 44.246 sec
>> Running org.apache.hadoop.hbase.zookeeper.TestZooKeeperMainServerArg
>> Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.376 sec
>> Running org.apache.hadoop.hbase.zookeeper.TestZooKeeperNodeTracker
>> Tests run: 2, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 1.129 sec
>>
>> Results :
>>
>> Failed tests:
>> testCyclicReplication(org.apache.hadoop.hbase.replication.TestMasterReplication):
>> Waited too much time for put replication
>>
>> Tests in error:
>> testSimplePutDelete(org.apache.hadoop.hbase.replication.TestMasterReplication):
>> Cluster already running at
>> /Users/larsgeorge/Downloads/hbase-0.92.0-rc3/hbase-0.92.0/target/test-data/7956128b-c9e3-4473-b3a2-c2faf140fb99/dfscluster_80ddefc2-e1be-477f-b761-df1ed883fc31
>>
>> Tests run: 1037, Failures: 1, Errors: 1, Skipped: 8
>>
>> [INFO]
>> ------------------------------------------------------------------------
>> [INFO] BUILD FAILURE
>> [INFO]
>> ------------------------------------------------------------------------
>> [INFO] Total time: 2:57:25.455s
>> [INFO] Finished at: Sun Jan 15 01:53:33 CET 2012
>> [INFO] Final Memory: 108M/213M
>> [INFO]
>> ------------------------------------------------------------------------
>> [WARNING] The requested profile "runAllTests" could not be activated
>> because it does not exist.
>> [ERROR] Failed to execute goal
>> org.apache.maven.plugins:maven-surefire-plugin:2.10:test (default-test) on
>> project hbase: Failure or timeout -> [Help 1]
>> [ERROR]
>> [ERROR] To see the full stack trace of the errors, re-run Maven with the
>> -e switch.
>> [ERROR] Re-run Maven using the -X switch to enable full debug logging.
>> [ERROR]
>> [ERROR] For more information about the errors and possible solutions,
>> please read the following articles:
>> [ERROR] [Help 1]
>> http://cwiki.apache.org/confluence/display/MAVEN/MojoExecutionException
>>
>> Did I run this wrong? What about the "runAllTests" warning above?
>>
>> Thanks,
>> Lars
>>
>> On Jan 8, 2012, at 10:39 PM, Stack wrote:
>>
>>> The fourth hbase 0.92.0 release candidate is available for download:
>>>
>>> http://people.apache.org/~stack/hbase-0.92.0-candidate-3/
>>>
>>> I've posted a secure and an insecure tarball built by jenkins.
>>>
>>> This RC bundles hadoop 1.0.0 and zookeeper 3.4.2.
>>>
>>> As said previous, HBase 0.92.0 includes a wagon-load of new features
>>> including coprocessors, security, a new (self-migrating) file format,
>>> distributed log splitting, etc. There's been > 660 fixes since 0.90.
>>> See the list here: http://su.pr/1VZzl5
>>>
>>> See the hbase manual for the low-down on what hadoop version this
>>> release will run on: http://hbase.apache.org/book.html#hadoop
>>>
>>> There is no migration necessary. A shutdown and restart after putting
>>> in place the new HBase should be all thats involved (A rolling restart
>>> from 0.90.x to 0.92.0 will not work). That said, once you have moved
>>> to 0.92.x, there is no going back to 0.90.x after the transition has
>>> been made.
+
Lars George 2012-01-16, 03:01
