You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@bigtop.apache.org by Roman Shaposhnik <rv...@apache.org> on 2014/09/15 04:00:17 UTC
Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Hi!
I just wanted to remind everybody that the bits
currently published over here:
http://bigtop01.cloudera.org:8080/view/Releases/job/Bigtop-0.8.0/
are as close as it gets to Bigtop 0.8.0.
We can't make a good release, without all
of the help from everybody who can invest
some amount of time in actually testing the
thing.
This coming week, Cos is likely to produce an
official RC0 set of artifacts, but packages are
already there -- lets test 'em!
Thanks,
Roman.
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Sean Mackrory <ma...@gmail.com>.
>> So, what I hear is that the stack in a decent condition and we can do the
>> formal source code release then?
+1 from me. I wish I had more time to test, but I what I have tested is
looking good.
On Fri, Sep 26, 2014 at 12:11 PM, Konstantin Boudnik <co...@apache.org> wrote:
> So, what I hear is that the stack in a decent condition and we can do the
> formal source code release then?
>
> --
> Take care,
> Konstantin (Cos) Boudnik
> 2CAC 8312 4870 D885 8616 6115 220F 6980 1F27 E622
>
> Disclaimer: Opinions expressed in this email are those of the author, and
> do not necessarily represent the views of any company the author might be
> affiliated with at the moment of writing.
>
> On Fri, Sep 26, 2014 at 11:12 AM, Sean Mackrory <ma...@gmail.com>
> wrote:
>
> > My time on it has been scattered so I don't have a very good record of
> > everything I've verified, but I've been playing with the bits since they
> > were first posted and haven't run across any issues. I've played with at
> > least HBase, ZooKeeper, YARN / MR and HDFS. I believe I messed with Sqoop
> > and Hive a bit too.
> >
> > On Wed, Sep 24, 2014 at 9:15 PM, Konstantin Boudnik <co...@apache.org>
> > wrote:
> >
> > > Thanks for testing the release candidate folks!
> > >
> > > And sorry for the delays guys - this is officially the most crazy two
> > > weeks in
> > > my life (and still not completely over). I have done some testing using
> > > 0.8.0
> > > bits - HDFS ops, YARN w/ MR, some standalone Spark - all looks good.
> > >
> > > I will try to update the Changes and cut the officially signed bits
> over
> > > the
> > > weekend.
> > >
> > > Cos
> > >
> > > On Wed, Sep 24, 2014 at 07:59PM, Roman Shaposhnik wrote:
> > > > Thanks for keeping us up-to-date on your progress. I plan to do some
> > > > additional testing over the weekend as well.
> > > >
> > > > Cos is also working on cutting formal release bits.
> > > >
> > > > Thanks,
> > > > Roman.
> > > >
> > > > On Wed, Sep 24, 2014 at 5:50 PM, Jean-Marc Spaggiari
> > > > <je...@spaggiari.org> wrote:
> > > > > Build 0.8 again today on CentOs without any issue. Will run
> > > applications
> > > > > testing tomorrow.
> > > > >
> > > > > 2014-09-20 20:38 GMT-04:00 Roman Shaposhnik <roman@shaposhnik.org
> >:
> > > > >>
> > > > >> First of all -- thanks a million for all the testing so far!
> > > > >>
> > > > >> Also, I seem to be able to use Mahout from RC0.8 just
> > > > >> fine. Lets get to the bottom of this.
> > > > >>
> > > > >> Thanks,
> > > > >> Roman.
> > > > >>
> > > > >> On Sat, Sep 20, 2014 at 5:57 AM, jay vyas <
> > > jayunit100.apache@gmail.com>
> > > > >> wrote:
> > > > >> > hold the phone on this...............
> > > > >> >
> > > > >> > I just spun up an rc08 instance, and it looks like some of the
> > > mahout
> > > > >> > jobs
> > > > >> > actually failed .... because the mahout version is for hadoop 1x
> > is
> > > > >> > being
> > > > >> > built (rather than 2x) . JIRA created.
> > > > >> >
> > > > >> > https://issues.apache.org/jira/browse/BIGTOP-1453
> > > > >> >
> > > > >
> > > > >
> > >
> >
>
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Sean Mackrory <ma...@gmail.com>.
>> So, what I hear is that the stack in a decent condition and we can do the
>> formal source code release then?
+1 from me. I wish I had more time to test, but I what I have tested is
looking good.
On Fri, Sep 26, 2014 at 12:11 PM, Konstantin Boudnik <co...@apache.org> wrote:
> So, what I hear is that the stack in a decent condition and we can do the
> formal source code release then?
>
> --
> Take care,
> Konstantin (Cos) Boudnik
> 2CAC 8312 4870 D885 8616 6115 220F 6980 1F27 E622
>
> Disclaimer: Opinions expressed in this email are those of the author, and
> do not necessarily represent the views of any company the author might be
> affiliated with at the moment of writing.
>
> On Fri, Sep 26, 2014 at 11:12 AM, Sean Mackrory <ma...@gmail.com>
> wrote:
>
> > My time on it has been scattered so I don't have a very good record of
> > everything I've verified, but I've been playing with the bits since they
> > were first posted and haven't run across any issues. I've played with at
> > least HBase, ZooKeeper, YARN / MR and HDFS. I believe I messed with Sqoop
> > and Hive a bit too.
> >
> > On Wed, Sep 24, 2014 at 9:15 PM, Konstantin Boudnik <co...@apache.org>
> > wrote:
> >
> > > Thanks for testing the release candidate folks!
> > >
> > > And sorry for the delays guys - this is officially the most crazy two
> > > weeks in
> > > my life (and still not completely over). I have done some testing using
> > > 0.8.0
> > > bits - HDFS ops, YARN w/ MR, some standalone Spark - all looks good.
> > >
> > > I will try to update the Changes and cut the officially signed bits
> over
> > > the
> > > weekend.
> > >
> > > Cos
> > >
> > > On Wed, Sep 24, 2014 at 07:59PM, Roman Shaposhnik wrote:
> > > > Thanks for keeping us up-to-date on your progress. I plan to do some
> > > > additional testing over the weekend as well.
> > > >
> > > > Cos is also working on cutting formal release bits.
> > > >
> > > > Thanks,
> > > > Roman.
> > > >
> > > > On Wed, Sep 24, 2014 at 5:50 PM, Jean-Marc Spaggiari
> > > > <je...@spaggiari.org> wrote:
> > > > > Build 0.8 again today on CentOs without any issue. Will run
> > > applications
> > > > > testing tomorrow.
> > > > >
> > > > > 2014-09-20 20:38 GMT-04:00 Roman Shaposhnik <roman@shaposhnik.org
> >:
> > > > >>
> > > > >> First of all -- thanks a million for all the testing so far!
> > > > >>
> > > > >> Also, I seem to be able to use Mahout from RC0.8 just
> > > > >> fine. Lets get to the bottom of this.
> > > > >>
> > > > >> Thanks,
> > > > >> Roman.
> > > > >>
> > > > >> On Sat, Sep 20, 2014 at 5:57 AM, jay vyas <
> > > jayunit100.apache@gmail.com>
> > > > >> wrote:
> > > > >> > hold the phone on this...............
> > > > >> >
> > > > >> > I just spun up an rc08 instance, and it looks like some of the
> > > mahout
> > > > >> > jobs
> > > > >> > actually failed .... because the mahout version is for hadoop 1x
> > is
> > > > >> > being
> > > > >> > built (rather than 2x) . JIRA created.
> > > > >> >
> > > > >> > https://issues.apache.org/jira/browse/BIGTOP-1453
> > > > >> >
> > > > >
> > > > >
> > >
> >
>
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Konstantin Boudnik <co...@apache.org>.
Could you please open a JIRA and document the workaround, so I can include
it into 0.8.0 source release that I am planning on doing over the weekend?
Thanks a lot!
--
Take care,
Konstantin (Cos) Boudnik
2CAC 8312 4870 D885 8616 6115 220F 6980 1F27 E622
Disclaimer: Opinions expressed in this email are those of the author, and
do not necessarily represent the views of any company the author might be
affiliated with at the moment of writing.
On Fri, Sep 26, 2014 at 1:47 PM, Jean-Marc Spaggiari <
jean-marc@spaggiari.org> wrote:
> Oh, yes. There is a workarround for sure. It just needs to be added in the
> HBase classpath. Nothing complicated.
>
> 2014-09-26 16:28 GMT-04:00 Konstantin Boudnik <co...@apache.org>:
>
> I let Andrew or anyone else with more Phoenix experience to comment on
>> this, but from my standpoint - if there's a configuration workaround I am
>> willing to document this in RELEASE NOTES and go with release. If the fix
>> requires changes in the packages - I guess we'll have to do it.
>>
>> --
>> Take care,
>> Konstantin (Cos) Boudnik
>> 2CAC 8312 4870 D885 8616 6115 220F 6980 1F27 E622
>>
>> Disclaimer: Opinions expressed in this email are those of the author, and
>> do not necessarily represent the views of any company the author might be
>> affiliated with at the moment of writing.
>>
>> On Fri, Sep 26, 2014 at 12:50 PM, Jean-Marc Spaggiari <
>> jean-marc@spaggiari.org> wrote:
>>
>>> One thing. Just build 0.8.0. Went weel, but when running Phoenix on top
>>> of HBase got this:
>>> java.io.IOException: No jar path specified for
>>> org.apache.phoenix.coprocessor.MetaDataRegionObserver
>>> (Many of those)
>>>
>>> I think we are still missing a jar somewhere in the claspath for it.
>>>
>>> JM
>>>
>>> 2014-09-26 15:11 GMT-04:00 Konstantin Boudnik <co...@apache.org>:
>>>
>>> So, what I hear is that the stack in a decent condition and we can do
>>>> the formal source code release then?
>>>>
>>>> --
>>>> Take care,
>>>> Konstantin (Cos) Boudnik
>>>> 2CAC 8312 4870 D885 8616 6115 220F 6980 1F27 E622
>>>>
>>>> Disclaimer: Opinions expressed in this email are those of the author,
>>>> and do not necessarily represent the views of any company the author might
>>>> be affiliated with at the moment of writing.
>>>>
>>>> On Fri, Sep 26, 2014 at 11:12 AM, Sean Mackrory <ma...@gmail.com>
>>>> wrote:
>>>>
>>>>> My time on it has been scattered so I don't have a very good record of
>>>>> everything I've verified, but I've been playing with the bits since
>>>>> they
>>>>> were first posted and haven't run across any issues. I've played with
>>>>> at
>>>>> least HBase, ZooKeeper, YARN / MR and HDFS. I believe I messed with
>>>>> Sqoop
>>>>> and Hive a bit too.
>>>>>
>>>>> On Wed, Sep 24, 2014 at 9:15 PM, Konstantin Boudnik <co...@apache.org>
>>>>> wrote:
>>>>>
>>>>> > Thanks for testing the release candidate folks!
>>>>> >
>>>>> > And sorry for the delays guys - this is officially the most crazy two
>>>>> > weeks in
>>>>> > my life (and still not completely over). I have done some testing
>>>>> using
>>>>> > 0.8.0
>>>>> > bits - HDFS ops, YARN w/ MR, some standalone Spark - all looks good.
>>>>> >
>>>>> > I will try to update the Changes and cut the officially signed bits
>>>>> over
>>>>> > the
>>>>> > weekend.
>>>>> >
>>>>> > Cos
>>>>> >
>>>>> > On Wed, Sep 24, 2014 at 07:59PM, Roman Shaposhnik wrote:
>>>>> > > Thanks for keeping us up-to-date on your progress. I plan to do
>>>>> some
>>>>> > > additional testing over the weekend as well.
>>>>> > >
>>>>> > > Cos is also working on cutting formal release bits.
>>>>> > >
>>>>> > > Thanks,
>>>>> > > Roman.
>>>>> > >
>>>>> > > On Wed, Sep 24, 2014 at 5:50 PM, Jean-Marc Spaggiari
>>>>> > > <je...@spaggiari.org> wrote:
>>>>> > > > Build 0.8 again today on CentOs without any issue. Will run
>>>>> > applications
>>>>> > > > testing tomorrow.
>>>>> > > >
>>>>> > > > 2014-09-20 20:38 GMT-04:00 Roman Shaposhnik <
>>>>> roman@shaposhnik.org>:
>>>>> > > >>
>>>>> > > >> First of all -- thanks a million for all the testing so far!
>>>>> > > >>
>>>>> > > >> Also, I seem to be able to use Mahout from RC0.8 just
>>>>> > > >> fine. Lets get to the bottom of this.
>>>>> > > >>
>>>>> > > >> Thanks,
>>>>> > > >> Roman.
>>>>> > > >>
>>>>> > > >> On Sat, Sep 20, 2014 at 5:57 AM, jay vyas <
>>>>> > jayunit100.apache@gmail.com>
>>>>> > > >> wrote:
>>>>> > > >> > hold the phone on this...............
>>>>> > > >> >
>>>>> > > >> > I just spun up an rc08 instance, and it looks like some of the
>>>>> > mahout
>>>>> > > >> > jobs
>>>>> > > >> > actually failed .... because the mahout version is for hadoop
>>>>> 1x is
>>>>> > > >> > being
>>>>> > > >> > built (rather than 2x) . JIRA created.
>>>>> > > >> >
>>>>> > > >> > https://issues.apache.org/jira/browse/BIGTOP-1453
>>>>> > > >> >
>>>>> > > >
>>>>> > > >
>>>>> >
>>>>>
>>>>
>>>>
>>>
>>
>
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Konstantin Boudnik <co...@apache.org>.
Could you please open a JIRA and document the workaround, so I can include
it into 0.8.0 source release that I am planning on doing over the weekend?
Thanks a lot!
--
Take care,
Konstantin (Cos) Boudnik
2CAC 8312 4870 D885 8616 6115 220F 6980 1F27 E622
Disclaimer: Opinions expressed in this email are those of the author, and
do not necessarily represent the views of any company the author might be
affiliated with at the moment of writing.
On Fri, Sep 26, 2014 at 1:47 PM, Jean-Marc Spaggiari <
jean-marc@spaggiari.org> wrote:
> Oh, yes. There is a workarround for sure. It just needs to be added in the
> HBase classpath. Nothing complicated.
>
> 2014-09-26 16:28 GMT-04:00 Konstantin Boudnik <co...@apache.org>:
>
> I let Andrew or anyone else with more Phoenix experience to comment on
>> this, but from my standpoint - if there's a configuration workaround I am
>> willing to document this in RELEASE NOTES and go with release. If the fix
>> requires changes in the packages - I guess we'll have to do it.
>>
>> --
>> Take care,
>> Konstantin (Cos) Boudnik
>> 2CAC 8312 4870 D885 8616 6115 220F 6980 1F27 E622
>>
>> Disclaimer: Opinions expressed in this email are those of the author, and
>> do not necessarily represent the views of any company the author might be
>> affiliated with at the moment of writing.
>>
>> On Fri, Sep 26, 2014 at 12:50 PM, Jean-Marc Spaggiari <
>> jean-marc@spaggiari.org> wrote:
>>
>>> One thing. Just build 0.8.0. Went weel, but when running Phoenix on top
>>> of HBase got this:
>>> java.io.IOException: No jar path specified for
>>> org.apache.phoenix.coprocessor.MetaDataRegionObserver
>>> (Many of those)
>>>
>>> I think we are still missing a jar somewhere in the claspath for it.
>>>
>>> JM
>>>
>>> 2014-09-26 15:11 GMT-04:00 Konstantin Boudnik <co...@apache.org>:
>>>
>>> So, what I hear is that the stack in a decent condition and we can do
>>>> the formal source code release then?
>>>>
>>>> --
>>>> Take care,
>>>> Konstantin (Cos) Boudnik
>>>> 2CAC 8312 4870 D885 8616 6115 220F 6980 1F27 E622
>>>>
>>>> Disclaimer: Opinions expressed in this email are those of the author,
>>>> and do not necessarily represent the views of any company the author might
>>>> be affiliated with at the moment of writing.
>>>>
>>>> On Fri, Sep 26, 2014 at 11:12 AM, Sean Mackrory <ma...@gmail.com>
>>>> wrote:
>>>>
>>>>> My time on it has been scattered so I don't have a very good record of
>>>>> everything I've verified, but I've been playing with the bits since
>>>>> they
>>>>> were first posted and haven't run across any issues. I've played with
>>>>> at
>>>>> least HBase, ZooKeeper, YARN / MR and HDFS. I believe I messed with
>>>>> Sqoop
>>>>> and Hive a bit too.
>>>>>
>>>>> On Wed, Sep 24, 2014 at 9:15 PM, Konstantin Boudnik <co...@apache.org>
>>>>> wrote:
>>>>>
>>>>> > Thanks for testing the release candidate folks!
>>>>> >
>>>>> > And sorry for the delays guys - this is officially the most crazy two
>>>>> > weeks in
>>>>> > my life (and still not completely over). I have done some testing
>>>>> using
>>>>> > 0.8.0
>>>>> > bits - HDFS ops, YARN w/ MR, some standalone Spark - all looks good.
>>>>> >
>>>>> > I will try to update the Changes and cut the officially signed bits
>>>>> over
>>>>> > the
>>>>> > weekend.
>>>>> >
>>>>> > Cos
>>>>> >
>>>>> > On Wed, Sep 24, 2014 at 07:59PM, Roman Shaposhnik wrote:
>>>>> > > Thanks for keeping us up-to-date on your progress. I plan to do
>>>>> some
>>>>> > > additional testing over the weekend as well.
>>>>> > >
>>>>> > > Cos is also working on cutting formal release bits.
>>>>> > >
>>>>> > > Thanks,
>>>>> > > Roman.
>>>>> > >
>>>>> > > On Wed, Sep 24, 2014 at 5:50 PM, Jean-Marc Spaggiari
>>>>> > > <je...@spaggiari.org> wrote:
>>>>> > > > Build 0.8 again today on CentOs without any issue. Will run
>>>>> > applications
>>>>> > > > testing tomorrow.
>>>>> > > >
>>>>> > > > 2014-09-20 20:38 GMT-04:00 Roman Shaposhnik <
>>>>> roman@shaposhnik.org>:
>>>>> > > >>
>>>>> > > >> First of all -- thanks a million for all the testing so far!
>>>>> > > >>
>>>>> > > >> Also, I seem to be able to use Mahout from RC0.8 just
>>>>> > > >> fine. Lets get to the bottom of this.
>>>>> > > >>
>>>>> > > >> Thanks,
>>>>> > > >> Roman.
>>>>> > > >>
>>>>> > > >> On Sat, Sep 20, 2014 at 5:57 AM, jay vyas <
>>>>> > jayunit100.apache@gmail.com>
>>>>> > > >> wrote:
>>>>> > > >> > hold the phone on this...............
>>>>> > > >> >
>>>>> > > >> > I just spun up an rc08 instance, and it looks like some of the
>>>>> > mahout
>>>>> > > >> > jobs
>>>>> > > >> > actually failed .... because the mahout version is for hadoop
>>>>> 1x is
>>>>> > > >> > being
>>>>> > > >> > built (rather than 2x) . JIRA created.
>>>>> > > >> >
>>>>> > > >> > https://issues.apache.org/jira/browse/BIGTOP-1453
>>>>> > > >> >
>>>>> > > >
>>>>> > > >
>>>>> >
>>>>>
>>>>
>>>>
>>>
>>
>
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Jean-Marc Spaggiari <je...@spaggiari.org>.
Oh, yes. There is a workarround for sure. It just needs to be added in the
HBase classpath. Nothing complicated.
2014-09-26 16:28 GMT-04:00 Konstantin Boudnik <co...@apache.org>:
> I let Andrew or anyone else with more Phoenix experience to comment on
> this, but from my standpoint - if there's a configuration workaround I am
> willing to document this in RELEASE NOTES and go with release. If the fix
> requires changes in the packages - I guess we'll have to do it.
>
> --
> Take care,
> Konstantin (Cos) Boudnik
> 2CAC 8312 4870 D885 8616 6115 220F 6980 1F27 E622
>
> Disclaimer: Opinions expressed in this email are those of the author, and
> do not necessarily represent the views of any company the author might be
> affiliated with at the moment of writing.
>
> On Fri, Sep 26, 2014 at 12:50 PM, Jean-Marc Spaggiari <
> jean-marc@spaggiari.org> wrote:
>
>> One thing. Just build 0.8.0. Went weel, but when running Phoenix on top
>> of HBase got this:
>> java.io.IOException: No jar path specified for
>> org.apache.phoenix.coprocessor.MetaDataRegionObserver
>> (Many of those)
>>
>> I think we are still missing a jar somewhere in the claspath for it.
>>
>> JM
>>
>> 2014-09-26 15:11 GMT-04:00 Konstantin Boudnik <co...@apache.org>:
>>
>> So, what I hear is that the stack in a decent condition and we can do the
>>> formal source code release then?
>>>
>>> --
>>> Take care,
>>> Konstantin (Cos) Boudnik
>>> 2CAC 8312 4870 D885 8616 6115 220F 6980 1F27 E622
>>>
>>> Disclaimer: Opinions expressed in this email are those of the author,
>>> and do not necessarily represent the views of any company the author might
>>> be affiliated with at the moment of writing.
>>>
>>> On Fri, Sep 26, 2014 at 11:12 AM, Sean Mackrory <ma...@gmail.com>
>>> wrote:
>>>
>>>> My time on it has been scattered so I don't have a very good record of
>>>> everything I've verified, but I've been playing with the bits since they
>>>> were first posted and haven't run across any issues. I've played with at
>>>> least HBase, ZooKeeper, YARN / MR and HDFS. I believe I messed with
>>>> Sqoop
>>>> and Hive a bit too.
>>>>
>>>> On Wed, Sep 24, 2014 at 9:15 PM, Konstantin Boudnik <co...@apache.org>
>>>> wrote:
>>>>
>>>> > Thanks for testing the release candidate folks!
>>>> >
>>>> > And sorry for the delays guys - this is officially the most crazy two
>>>> > weeks in
>>>> > my life (and still not completely over). I have done some testing
>>>> using
>>>> > 0.8.0
>>>> > bits - HDFS ops, YARN w/ MR, some standalone Spark - all looks good.
>>>> >
>>>> > I will try to update the Changes and cut the officially signed bits
>>>> over
>>>> > the
>>>> > weekend.
>>>> >
>>>> > Cos
>>>> >
>>>> > On Wed, Sep 24, 2014 at 07:59PM, Roman Shaposhnik wrote:
>>>> > > Thanks for keeping us up-to-date on your progress. I plan to do some
>>>> > > additional testing over the weekend as well.
>>>> > >
>>>> > > Cos is also working on cutting formal release bits.
>>>> > >
>>>> > > Thanks,
>>>> > > Roman.
>>>> > >
>>>> > > On Wed, Sep 24, 2014 at 5:50 PM, Jean-Marc Spaggiari
>>>> > > <je...@spaggiari.org> wrote:
>>>> > > > Build 0.8 again today on CentOs without any issue. Will run
>>>> > applications
>>>> > > > testing tomorrow.
>>>> > > >
>>>> > > > 2014-09-20 20:38 GMT-04:00 Roman Shaposhnik <roman@shaposhnik.org
>>>> >:
>>>> > > >>
>>>> > > >> First of all -- thanks a million for all the testing so far!
>>>> > > >>
>>>> > > >> Also, I seem to be able to use Mahout from RC0.8 just
>>>> > > >> fine. Lets get to the bottom of this.
>>>> > > >>
>>>> > > >> Thanks,
>>>> > > >> Roman.
>>>> > > >>
>>>> > > >> On Sat, Sep 20, 2014 at 5:57 AM, jay vyas <
>>>> > jayunit100.apache@gmail.com>
>>>> > > >> wrote:
>>>> > > >> > hold the phone on this...............
>>>> > > >> >
>>>> > > >> > I just spun up an rc08 instance, and it looks like some of the
>>>> > mahout
>>>> > > >> > jobs
>>>> > > >> > actually failed .... because the mahout version is for hadoop
>>>> 1x is
>>>> > > >> > being
>>>> > > >> > built (rather than 2x) . JIRA created.
>>>> > > >> >
>>>> > > >> > https://issues.apache.org/jira/browse/BIGTOP-1453
>>>> > > >> >
>>>> > > >
>>>> > > >
>>>> >
>>>>
>>>
>>>
>>
>
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Jean-Marc Spaggiari <je...@spaggiari.org>.
Oh, yes. There is a workarround for sure. It just needs to be added in the
HBase classpath. Nothing complicated.
2014-09-26 16:28 GMT-04:00 Konstantin Boudnik <co...@apache.org>:
> I let Andrew or anyone else with more Phoenix experience to comment on
> this, but from my standpoint - if there's a configuration workaround I am
> willing to document this in RELEASE NOTES and go with release. If the fix
> requires changes in the packages - I guess we'll have to do it.
>
> --
> Take care,
> Konstantin (Cos) Boudnik
> 2CAC 8312 4870 D885 8616 6115 220F 6980 1F27 E622
>
> Disclaimer: Opinions expressed in this email are those of the author, and
> do not necessarily represent the views of any company the author might be
> affiliated with at the moment of writing.
>
> On Fri, Sep 26, 2014 at 12:50 PM, Jean-Marc Spaggiari <
> jean-marc@spaggiari.org> wrote:
>
>> One thing. Just build 0.8.0. Went weel, but when running Phoenix on top
>> of HBase got this:
>> java.io.IOException: No jar path specified for
>> org.apache.phoenix.coprocessor.MetaDataRegionObserver
>> (Many of those)
>>
>> I think we are still missing a jar somewhere in the claspath for it.
>>
>> JM
>>
>> 2014-09-26 15:11 GMT-04:00 Konstantin Boudnik <co...@apache.org>:
>>
>> So, what I hear is that the stack in a decent condition and we can do the
>>> formal source code release then?
>>>
>>> --
>>> Take care,
>>> Konstantin (Cos) Boudnik
>>> 2CAC 8312 4870 D885 8616 6115 220F 6980 1F27 E622
>>>
>>> Disclaimer: Opinions expressed in this email are those of the author,
>>> and do not necessarily represent the views of any company the author might
>>> be affiliated with at the moment of writing.
>>>
>>> On Fri, Sep 26, 2014 at 11:12 AM, Sean Mackrory <ma...@gmail.com>
>>> wrote:
>>>
>>>> My time on it has been scattered so I don't have a very good record of
>>>> everything I've verified, but I've been playing with the bits since they
>>>> were first posted and haven't run across any issues. I've played with at
>>>> least HBase, ZooKeeper, YARN / MR and HDFS. I believe I messed with
>>>> Sqoop
>>>> and Hive a bit too.
>>>>
>>>> On Wed, Sep 24, 2014 at 9:15 PM, Konstantin Boudnik <co...@apache.org>
>>>> wrote:
>>>>
>>>> > Thanks for testing the release candidate folks!
>>>> >
>>>> > And sorry for the delays guys - this is officially the most crazy two
>>>> > weeks in
>>>> > my life (and still not completely over). I have done some testing
>>>> using
>>>> > 0.8.0
>>>> > bits - HDFS ops, YARN w/ MR, some standalone Spark - all looks good.
>>>> >
>>>> > I will try to update the Changes and cut the officially signed bits
>>>> over
>>>> > the
>>>> > weekend.
>>>> >
>>>> > Cos
>>>> >
>>>> > On Wed, Sep 24, 2014 at 07:59PM, Roman Shaposhnik wrote:
>>>> > > Thanks for keeping us up-to-date on your progress. I plan to do some
>>>> > > additional testing over the weekend as well.
>>>> > >
>>>> > > Cos is also working on cutting formal release bits.
>>>> > >
>>>> > > Thanks,
>>>> > > Roman.
>>>> > >
>>>> > > On Wed, Sep 24, 2014 at 5:50 PM, Jean-Marc Spaggiari
>>>> > > <je...@spaggiari.org> wrote:
>>>> > > > Build 0.8 again today on CentOs without any issue. Will run
>>>> > applications
>>>> > > > testing tomorrow.
>>>> > > >
>>>> > > > 2014-09-20 20:38 GMT-04:00 Roman Shaposhnik <roman@shaposhnik.org
>>>> >:
>>>> > > >>
>>>> > > >> First of all -- thanks a million for all the testing so far!
>>>> > > >>
>>>> > > >> Also, I seem to be able to use Mahout from RC0.8 just
>>>> > > >> fine. Lets get to the bottom of this.
>>>> > > >>
>>>> > > >> Thanks,
>>>> > > >> Roman.
>>>> > > >>
>>>> > > >> On Sat, Sep 20, 2014 at 5:57 AM, jay vyas <
>>>> > jayunit100.apache@gmail.com>
>>>> > > >> wrote:
>>>> > > >> > hold the phone on this...............
>>>> > > >> >
>>>> > > >> > I just spun up an rc08 instance, and it looks like some of the
>>>> > mahout
>>>> > > >> > jobs
>>>> > > >> > actually failed .... because the mahout version is for hadoop
>>>> 1x is
>>>> > > >> > being
>>>> > > >> > built (rather than 2x) . JIRA created.
>>>> > > >> >
>>>> > > >> > https://issues.apache.org/jira/browse/BIGTOP-1453
>>>> > > >> >
>>>> > > >
>>>> > > >
>>>> >
>>>>
>>>
>>>
>>
>
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Konstantin Boudnik <co...@apache.org>.
I let Andrew or anyone else with more Phoenix experience to comment on
this, but from my standpoint - if there's a configuration workaround I am
willing to document this in RELEASE NOTES and go with release. If the fix
requires changes in the packages - I guess we'll have to do it.
--
Take care,
Konstantin (Cos) Boudnik
2CAC 8312 4870 D885 8616 6115 220F 6980 1F27 E622
Disclaimer: Opinions expressed in this email are those of the author, and
do not necessarily represent the views of any company the author might be
affiliated with at the moment of writing.
On Fri, Sep 26, 2014 at 12:50 PM, Jean-Marc Spaggiari <
jean-marc@spaggiari.org> wrote:
> One thing. Just build 0.8.0. Went weel, but when running Phoenix on top of
> HBase got this:
> java.io.IOException: No jar path specified for
> org.apache.phoenix.coprocessor.MetaDataRegionObserver
> (Many of those)
>
> I think we are still missing a jar somewhere in the claspath for it.
>
> JM
>
> 2014-09-26 15:11 GMT-04:00 Konstantin Boudnik <co...@apache.org>:
>
> So, what I hear is that the stack in a decent condition and we can do the
>> formal source code release then?
>>
>> --
>> Take care,
>> Konstantin (Cos) Boudnik
>> 2CAC 8312 4870 D885 8616 6115 220F 6980 1F27 E622
>>
>> Disclaimer: Opinions expressed in this email are those of the author, and
>> do not necessarily represent the views of any company the author might be
>> affiliated with at the moment of writing.
>>
>> On Fri, Sep 26, 2014 at 11:12 AM, Sean Mackrory <ma...@gmail.com>
>> wrote:
>>
>>> My time on it has been scattered so I don't have a very good record of
>>> everything I've verified, but I've been playing with the bits since they
>>> were first posted and haven't run across any issues. I've played with at
>>> least HBase, ZooKeeper, YARN / MR and HDFS. I believe I messed with Sqoop
>>> and Hive a bit too.
>>>
>>> On Wed, Sep 24, 2014 at 9:15 PM, Konstantin Boudnik <co...@apache.org>
>>> wrote:
>>>
>>> > Thanks for testing the release candidate folks!
>>> >
>>> > And sorry for the delays guys - this is officially the most crazy two
>>> > weeks in
>>> > my life (and still not completely over). I have done some testing using
>>> > 0.8.0
>>> > bits - HDFS ops, YARN w/ MR, some standalone Spark - all looks good.
>>> >
>>> > I will try to update the Changes and cut the officially signed bits
>>> over
>>> > the
>>> > weekend.
>>> >
>>> > Cos
>>> >
>>> > On Wed, Sep 24, 2014 at 07:59PM, Roman Shaposhnik wrote:
>>> > > Thanks for keeping us up-to-date on your progress. I plan to do some
>>> > > additional testing over the weekend as well.
>>> > >
>>> > > Cos is also working on cutting formal release bits.
>>> > >
>>> > > Thanks,
>>> > > Roman.
>>> > >
>>> > > On Wed, Sep 24, 2014 at 5:50 PM, Jean-Marc Spaggiari
>>> > > <je...@spaggiari.org> wrote:
>>> > > > Build 0.8 again today on CentOs without any issue. Will run
>>> > applications
>>> > > > testing tomorrow.
>>> > > >
>>> > > > 2014-09-20 20:38 GMT-04:00 Roman Shaposhnik <roman@shaposhnik.org
>>> >:
>>> > > >>
>>> > > >> First of all -- thanks a million for all the testing so far!
>>> > > >>
>>> > > >> Also, I seem to be able to use Mahout from RC0.8 just
>>> > > >> fine. Lets get to the bottom of this.
>>> > > >>
>>> > > >> Thanks,
>>> > > >> Roman.
>>> > > >>
>>> > > >> On Sat, Sep 20, 2014 at 5:57 AM, jay vyas <
>>> > jayunit100.apache@gmail.com>
>>> > > >> wrote:
>>> > > >> > hold the phone on this...............
>>> > > >> >
>>> > > >> > I just spun up an rc08 instance, and it looks like some of the
>>> > mahout
>>> > > >> > jobs
>>> > > >> > actually failed .... because the mahout version is for hadoop
>>> 1x is
>>> > > >> > being
>>> > > >> > built (rather than 2x) . JIRA created.
>>> > > >> >
>>> > > >> > https://issues.apache.org/jira/browse/BIGTOP-1453
>>> > > >> >
>>> > > >
>>> > > >
>>> >
>>>
>>
>>
>
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Konstantin Boudnik <co...@apache.org>.
I let Andrew or anyone else with more Phoenix experience to comment on
this, but from my standpoint - if there's a configuration workaround I am
willing to document this in RELEASE NOTES and go with release. If the fix
requires changes in the packages - I guess we'll have to do it.
--
Take care,
Konstantin (Cos) Boudnik
2CAC 8312 4870 D885 8616 6115 220F 6980 1F27 E622
Disclaimer: Opinions expressed in this email are those of the author, and
do not necessarily represent the views of any company the author might be
affiliated with at the moment of writing.
On Fri, Sep 26, 2014 at 12:50 PM, Jean-Marc Spaggiari <
jean-marc@spaggiari.org> wrote:
> One thing. Just build 0.8.0. Went weel, but when running Phoenix on top of
> HBase got this:
> java.io.IOException: No jar path specified for
> org.apache.phoenix.coprocessor.MetaDataRegionObserver
> (Many of those)
>
> I think we are still missing a jar somewhere in the claspath for it.
>
> JM
>
> 2014-09-26 15:11 GMT-04:00 Konstantin Boudnik <co...@apache.org>:
>
> So, what I hear is that the stack in a decent condition and we can do the
>> formal source code release then?
>>
>> --
>> Take care,
>> Konstantin (Cos) Boudnik
>> 2CAC 8312 4870 D885 8616 6115 220F 6980 1F27 E622
>>
>> Disclaimer: Opinions expressed in this email are those of the author, and
>> do not necessarily represent the views of any company the author might be
>> affiliated with at the moment of writing.
>>
>> On Fri, Sep 26, 2014 at 11:12 AM, Sean Mackrory <ma...@gmail.com>
>> wrote:
>>
>>> My time on it has been scattered so I don't have a very good record of
>>> everything I've verified, but I've been playing with the bits since they
>>> were first posted and haven't run across any issues. I've played with at
>>> least HBase, ZooKeeper, YARN / MR and HDFS. I believe I messed with Sqoop
>>> and Hive a bit too.
>>>
>>> On Wed, Sep 24, 2014 at 9:15 PM, Konstantin Boudnik <co...@apache.org>
>>> wrote:
>>>
>>> > Thanks for testing the release candidate folks!
>>> >
>>> > And sorry for the delays guys - this is officially the most crazy two
>>> > weeks in
>>> > my life (and still not completely over). I have done some testing using
>>> > 0.8.0
>>> > bits - HDFS ops, YARN w/ MR, some standalone Spark - all looks good.
>>> >
>>> > I will try to update the Changes and cut the officially signed bits
>>> over
>>> > the
>>> > weekend.
>>> >
>>> > Cos
>>> >
>>> > On Wed, Sep 24, 2014 at 07:59PM, Roman Shaposhnik wrote:
>>> > > Thanks for keeping us up-to-date on your progress. I plan to do some
>>> > > additional testing over the weekend as well.
>>> > >
>>> > > Cos is also working on cutting formal release bits.
>>> > >
>>> > > Thanks,
>>> > > Roman.
>>> > >
>>> > > On Wed, Sep 24, 2014 at 5:50 PM, Jean-Marc Spaggiari
>>> > > <je...@spaggiari.org> wrote:
>>> > > > Build 0.8 again today on CentOs without any issue. Will run
>>> > applications
>>> > > > testing tomorrow.
>>> > > >
>>> > > > 2014-09-20 20:38 GMT-04:00 Roman Shaposhnik <roman@shaposhnik.org
>>> >:
>>> > > >>
>>> > > >> First of all -- thanks a million for all the testing so far!
>>> > > >>
>>> > > >> Also, I seem to be able to use Mahout from RC0.8 just
>>> > > >> fine. Lets get to the bottom of this.
>>> > > >>
>>> > > >> Thanks,
>>> > > >> Roman.
>>> > > >>
>>> > > >> On Sat, Sep 20, 2014 at 5:57 AM, jay vyas <
>>> > jayunit100.apache@gmail.com>
>>> > > >> wrote:
>>> > > >> > hold the phone on this...............
>>> > > >> >
>>> > > >> > I just spun up an rc08 instance, and it looks like some of the
>>> > mahout
>>> > > >> > jobs
>>> > > >> > actually failed .... because the mahout version is for hadoop
>>> 1x is
>>> > > >> > being
>>> > > >> > built (rather than 2x) . JIRA created.
>>> > > >> >
>>> > > >> > https://issues.apache.org/jira/browse/BIGTOP-1453
>>> > > >> >
>>> > > >
>>> > > >
>>> >
>>>
>>
>>
>
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Jean-Marc Spaggiari <je...@spaggiari.org>.
One thing. Just build 0.8.0. Went weel, but when running Phoenix on top of
HBase got this:
java.io.IOException: No jar path specified for
org.apache.phoenix.coprocessor.MetaDataRegionObserver
(Many of those)
I think we are still missing a jar somewhere in the claspath for it.
JM
2014-09-26 15:11 GMT-04:00 Konstantin Boudnik <co...@apache.org>:
> So, what I hear is that the stack in a decent condition and we can do the
> formal source code release then?
>
> --
> Take care,
> Konstantin (Cos) Boudnik
> 2CAC 8312 4870 D885 8616 6115 220F 6980 1F27 E622
>
> Disclaimer: Opinions expressed in this email are those of the author, and
> do not necessarily represent the views of any company the author might be
> affiliated with at the moment of writing.
>
> On Fri, Sep 26, 2014 at 11:12 AM, Sean Mackrory <ma...@gmail.com>
> wrote:
>
>> My time on it has been scattered so I don't have a very good record of
>> everything I've verified, but I've been playing with the bits since they
>> were first posted and haven't run across any issues. I've played with at
>> least HBase, ZooKeeper, YARN / MR and HDFS. I believe I messed with Sqoop
>> and Hive a bit too.
>>
>> On Wed, Sep 24, 2014 at 9:15 PM, Konstantin Boudnik <co...@apache.org>
>> wrote:
>>
>> > Thanks for testing the release candidate folks!
>> >
>> > And sorry for the delays guys - this is officially the most crazy two
>> > weeks in
>> > my life (and still not completely over). I have done some testing using
>> > 0.8.0
>> > bits - HDFS ops, YARN w/ MR, some standalone Spark - all looks good.
>> >
>> > I will try to update the Changes and cut the officially signed bits over
>> > the
>> > weekend.
>> >
>> > Cos
>> >
>> > On Wed, Sep 24, 2014 at 07:59PM, Roman Shaposhnik wrote:
>> > > Thanks for keeping us up-to-date on your progress. I plan to do some
>> > > additional testing over the weekend as well.
>> > >
>> > > Cos is also working on cutting formal release bits.
>> > >
>> > > Thanks,
>> > > Roman.
>> > >
>> > > On Wed, Sep 24, 2014 at 5:50 PM, Jean-Marc Spaggiari
>> > > <je...@spaggiari.org> wrote:
>> > > > Build 0.8 again today on CentOs without any issue. Will run
>> > applications
>> > > > testing tomorrow.
>> > > >
>> > > > 2014-09-20 20:38 GMT-04:00 Roman Shaposhnik <ro...@shaposhnik.org>:
>> > > >>
>> > > >> First of all -- thanks a million for all the testing so far!
>> > > >>
>> > > >> Also, I seem to be able to use Mahout from RC0.8 just
>> > > >> fine. Lets get to the bottom of this.
>> > > >>
>> > > >> Thanks,
>> > > >> Roman.
>> > > >>
>> > > >> On Sat, Sep 20, 2014 at 5:57 AM, jay vyas <
>> > jayunit100.apache@gmail.com>
>> > > >> wrote:
>> > > >> > hold the phone on this...............
>> > > >> >
>> > > >> > I just spun up an rc08 instance, and it looks like some of the
>> > mahout
>> > > >> > jobs
>> > > >> > actually failed .... because the mahout version is for hadoop 1x
>> is
>> > > >> > being
>> > > >> > built (rather than 2x) . JIRA created.
>> > > >> >
>> > > >> > https://issues.apache.org/jira/browse/BIGTOP-1453
>> > > >> >
>> > > >
>> > > >
>> >
>>
>
>
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Jean-Marc Spaggiari <je...@spaggiari.org>.
One thing. Just build 0.8.0. Went weel, but when running Phoenix on top of
HBase got this:
java.io.IOException: No jar path specified for
org.apache.phoenix.coprocessor.MetaDataRegionObserver
(Many of those)
I think we are still missing a jar somewhere in the claspath for it.
JM
2014-09-26 15:11 GMT-04:00 Konstantin Boudnik <co...@apache.org>:
> So, what I hear is that the stack in a decent condition and we can do the
> formal source code release then?
>
> --
> Take care,
> Konstantin (Cos) Boudnik
> 2CAC 8312 4870 D885 8616 6115 220F 6980 1F27 E622
>
> Disclaimer: Opinions expressed in this email are those of the author, and
> do not necessarily represent the views of any company the author might be
> affiliated with at the moment of writing.
>
> On Fri, Sep 26, 2014 at 11:12 AM, Sean Mackrory <ma...@gmail.com>
> wrote:
>
>> My time on it has been scattered so I don't have a very good record of
>> everything I've verified, but I've been playing with the bits since they
>> were first posted and haven't run across any issues. I've played with at
>> least HBase, ZooKeeper, YARN / MR and HDFS. I believe I messed with Sqoop
>> and Hive a bit too.
>>
>> On Wed, Sep 24, 2014 at 9:15 PM, Konstantin Boudnik <co...@apache.org>
>> wrote:
>>
>> > Thanks for testing the release candidate folks!
>> >
>> > And sorry for the delays guys - this is officially the most crazy two
>> > weeks in
>> > my life (and still not completely over). I have done some testing using
>> > 0.8.0
>> > bits - HDFS ops, YARN w/ MR, some standalone Spark - all looks good.
>> >
>> > I will try to update the Changes and cut the officially signed bits over
>> > the
>> > weekend.
>> >
>> > Cos
>> >
>> > On Wed, Sep 24, 2014 at 07:59PM, Roman Shaposhnik wrote:
>> > > Thanks for keeping us up-to-date on your progress. I plan to do some
>> > > additional testing over the weekend as well.
>> > >
>> > > Cos is also working on cutting formal release bits.
>> > >
>> > > Thanks,
>> > > Roman.
>> > >
>> > > On Wed, Sep 24, 2014 at 5:50 PM, Jean-Marc Spaggiari
>> > > <je...@spaggiari.org> wrote:
>> > > > Build 0.8 again today on CentOs without any issue. Will run
>> > applications
>> > > > testing tomorrow.
>> > > >
>> > > > 2014-09-20 20:38 GMT-04:00 Roman Shaposhnik <ro...@shaposhnik.org>:
>> > > >>
>> > > >> First of all -- thanks a million for all the testing so far!
>> > > >>
>> > > >> Also, I seem to be able to use Mahout from RC0.8 just
>> > > >> fine. Lets get to the bottom of this.
>> > > >>
>> > > >> Thanks,
>> > > >> Roman.
>> > > >>
>> > > >> On Sat, Sep 20, 2014 at 5:57 AM, jay vyas <
>> > jayunit100.apache@gmail.com>
>> > > >> wrote:
>> > > >> > hold the phone on this...............
>> > > >> >
>> > > >> > I just spun up an rc08 instance, and it looks like some of the
>> > mahout
>> > > >> > jobs
>> > > >> > actually failed .... because the mahout version is for hadoop 1x
>> is
>> > > >> > being
>> > > >> > built (rather than 2x) . JIRA created.
>> > > >> >
>> > > >> > https://issues.apache.org/jira/browse/BIGTOP-1453
>> > > >> >
>> > > >
>> > > >
>> >
>>
>
>
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Konstantin Boudnik <co...@apache.org>.
So, what I hear is that the stack in a decent condition and we can do the
formal source code release then?
--
Take care,
Konstantin (Cos) Boudnik
2CAC 8312 4870 D885 8616 6115 220F 6980 1F27 E622
Disclaimer: Opinions expressed in this email are those of the author, and
do not necessarily represent the views of any company the author might be
affiliated with at the moment of writing.
On Fri, Sep 26, 2014 at 11:12 AM, Sean Mackrory <ma...@gmail.com>
wrote:
> My time on it has been scattered so I don't have a very good record of
> everything I've verified, but I've been playing with the bits since they
> were first posted and haven't run across any issues. I've played with at
> least HBase, ZooKeeper, YARN / MR and HDFS. I believe I messed with Sqoop
> and Hive a bit too.
>
> On Wed, Sep 24, 2014 at 9:15 PM, Konstantin Boudnik <co...@apache.org>
> wrote:
>
> > Thanks for testing the release candidate folks!
> >
> > And sorry for the delays guys - this is officially the most crazy two
> > weeks in
> > my life (and still not completely over). I have done some testing using
> > 0.8.0
> > bits - HDFS ops, YARN w/ MR, some standalone Spark - all looks good.
> >
> > I will try to update the Changes and cut the officially signed bits over
> > the
> > weekend.
> >
> > Cos
> >
> > On Wed, Sep 24, 2014 at 07:59PM, Roman Shaposhnik wrote:
> > > Thanks for keeping us up-to-date on your progress. I plan to do some
> > > additional testing over the weekend as well.
> > >
> > > Cos is also working on cutting formal release bits.
> > >
> > > Thanks,
> > > Roman.
> > >
> > > On Wed, Sep 24, 2014 at 5:50 PM, Jean-Marc Spaggiari
> > > <je...@spaggiari.org> wrote:
> > > > Build 0.8 again today on CentOs without any issue. Will run
> > applications
> > > > testing tomorrow.
> > > >
> > > > 2014-09-20 20:38 GMT-04:00 Roman Shaposhnik <ro...@shaposhnik.org>:
> > > >>
> > > >> First of all -- thanks a million for all the testing so far!
> > > >>
> > > >> Also, I seem to be able to use Mahout from RC0.8 just
> > > >> fine. Lets get to the bottom of this.
> > > >>
> > > >> Thanks,
> > > >> Roman.
> > > >>
> > > >> On Sat, Sep 20, 2014 at 5:57 AM, jay vyas <
> > jayunit100.apache@gmail.com>
> > > >> wrote:
> > > >> > hold the phone on this...............
> > > >> >
> > > >> > I just spun up an rc08 instance, and it looks like some of the
> > mahout
> > > >> > jobs
> > > >> > actually failed .... because the mahout version is for hadoop 1x
> is
> > > >> > being
> > > >> > built (rather than 2x) . JIRA created.
> > > >> >
> > > >> > https://issues.apache.org/jira/browse/BIGTOP-1453
> > > >> >
> > > >
> > > >
> >
>
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Konstantin Boudnik <co...@apache.org>.
So, what I hear is that the stack in a decent condition and we can do the
formal source code release then?
--
Take care,
Konstantin (Cos) Boudnik
2CAC 8312 4870 D885 8616 6115 220F 6980 1F27 E622
Disclaimer: Opinions expressed in this email are those of the author, and
do not necessarily represent the views of any company the author might be
affiliated with at the moment of writing.
On Fri, Sep 26, 2014 at 11:12 AM, Sean Mackrory <ma...@gmail.com>
wrote:
> My time on it has been scattered so I don't have a very good record of
> everything I've verified, but I've been playing with the bits since they
> were first posted and haven't run across any issues. I've played with at
> least HBase, ZooKeeper, YARN / MR and HDFS. I believe I messed with Sqoop
> and Hive a bit too.
>
> On Wed, Sep 24, 2014 at 9:15 PM, Konstantin Boudnik <co...@apache.org>
> wrote:
>
> > Thanks for testing the release candidate folks!
> >
> > And sorry for the delays guys - this is officially the most crazy two
> > weeks in
> > my life (and still not completely over). I have done some testing using
> > 0.8.0
> > bits - HDFS ops, YARN w/ MR, some standalone Spark - all looks good.
> >
> > I will try to update the Changes and cut the officially signed bits over
> > the
> > weekend.
> >
> > Cos
> >
> > On Wed, Sep 24, 2014 at 07:59PM, Roman Shaposhnik wrote:
> > > Thanks for keeping us up-to-date on your progress. I plan to do some
> > > additional testing over the weekend as well.
> > >
> > > Cos is also working on cutting formal release bits.
> > >
> > > Thanks,
> > > Roman.
> > >
> > > On Wed, Sep 24, 2014 at 5:50 PM, Jean-Marc Spaggiari
> > > <je...@spaggiari.org> wrote:
> > > > Build 0.8 again today on CentOs without any issue. Will run
> > applications
> > > > testing tomorrow.
> > > >
> > > > 2014-09-20 20:38 GMT-04:00 Roman Shaposhnik <ro...@shaposhnik.org>:
> > > >>
> > > >> First of all -- thanks a million for all the testing so far!
> > > >>
> > > >> Also, I seem to be able to use Mahout from RC0.8 just
> > > >> fine. Lets get to the bottom of this.
> > > >>
> > > >> Thanks,
> > > >> Roman.
> > > >>
> > > >> On Sat, Sep 20, 2014 at 5:57 AM, jay vyas <
> > jayunit100.apache@gmail.com>
> > > >> wrote:
> > > >> > hold the phone on this...............
> > > >> >
> > > >> > I just spun up an rc08 instance, and it looks like some of the
> > mahout
> > > >> > jobs
> > > >> > actually failed .... because the mahout version is for hadoop 1x
> is
> > > >> > being
> > > >> > built (rather than 2x) . JIRA created.
> > > >> >
> > > >> > https://issues.apache.org/jira/browse/BIGTOP-1453
> > > >> >
> > > >
> > > >
> >
>
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Sean Mackrory <ma...@gmail.com>.
My time on it has been scattered so I don't have a very good record of
everything I've verified, but I've been playing with the bits since they
were first posted and haven't run across any issues. I've played with at
least HBase, ZooKeeper, YARN / MR and HDFS. I believe I messed with Sqoop
and Hive a bit too.
On Wed, Sep 24, 2014 at 9:15 PM, Konstantin Boudnik <co...@apache.org> wrote:
> Thanks for testing the release candidate folks!
>
> And sorry for the delays guys - this is officially the most crazy two
> weeks in
> my life (and still not completely over). I have done some testing using
> 0.8.0
> bits - HDFS ops, YARN w/ MR, some standalone Spark - all looks good.
>
> I will try to update the Changes and cut the officially signed bits over
> the
> weekend.
>
> Cos
>
> On Wed, Sep 24, 2014 at 07:59PM, Roman Shaposhnik wrote:
> > Thanks for keeping us up-to-date on your progress. I plan to do some
> > additional testing over the weekend as well.
> >
> > Cos is also working on cutting formal release bits.
> >
> > Thanks,
> > Roman.
> >
> > On Wed, Sep 24, 2014 at 5:50 PM, Jean-Marc Spaggiari
> > <je...@spaggiari.org> wrote:
> > > Build 0.8 again today on CentOs without any issue. Will run
> applications
> > > testing tomorrow.
> > >
> > > 2014-09-20 20:38 GMT-04:00 Roman Shaposhnik <ro...@shaposhnik.org>:
> > >>
> > >> First of all -- thanks a million for all the testing so far!
> > >>
> > >> Also, I seem to be able to use Mahout from RC0.8 just
> > >> fine. Lets get to the bottom of this.
> > >>
> > >> Thanks,
> > >> Roman.
> > >>
> > >> On Sat, Sep 20, 2014 at 5:57 AM, jay vyas <
> jayunit100.apache@gmail.com>
> > >> wrote:
> > >> > hold the phone on this...............
> > >> >
> > >> > I just spun up an rc08 instance, and it looks like some of the
> mahout
> > >> > jobs
> > >> > actually failed .... because the mahout version is for hadoop 1x is
> > >> > being
> > >> > built (rather than 2x) . JIRA created.
> > >> >
> > >> > https://issues.apache.org/jira/browse/BIGTOP-1453
> > >> >
> > >
> > >
>
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Sean Mackrory <ma...@gmail.com>.
My time on it has been scattered so I don't have a very good record of
everything I've verified, but I've been playing with the bits since they
were first posted and haven't run across any issues. I've played with at
least HBase, ZooKeeper, YARN / MR and HDFS. I believe I messed with Sqoop
and Hive a bit too.
On Wed, Sep 24, 2014 at 9:15 PM, Konstantin Boudnik <co...@apache.org> wrote:
> Thanks for testing the release candidate folks!
>
> And sorry for the delays guys - this is officially the most crazy two
> weeks in
> my life (and still not completely over). I have done some testing using
> 0.8.0
> bits - HDFS ops, YARN w/ MR, some standalone Spark - all looks good.
>
> I will try to update the Changes and cut the officially signed bits over
> the
> weekend.
>
> Cos
>
> On Wed, Sep 24, 2014 at 07:59PM, Roman Shaposhnik wrote:
> > Thanks for keeping us up-to-date on your progress. I plan to do some
> > additional testing over the weekend as well.
> >
> > Cos is also working on cutting formal release bits.
> >
> > Thanks,
> > Roman.
> >
> > On Wed, Sep 24, 2014 at 5:50 PM, Jean-Marc Spaggiari
> > <je...@spaggiari.org> wrote:
> > > Build 0.8 again today on CentOs without any issue. Will run
> applications
> > > testing tomorrow.
> > >
> > > 2014-09-20 20:38 GMT-04:00 Roman Shaposhnik <ro...@shaposhnik.org>:
> > >>
> > >> First of all -- thanks a million for all the testing so far!
> > >>
> > >> Also, I seem to be able to use Mahout from RC0.8 just
> > >> fine. Lets get to the bottom of this.
> > >>
> > >> Thanks,
> > >> Roman.
> > >>
> > >> On Sat, Sep 20, 2014 at 5:57 AM, jay vyas <
> jayunit100.apache@gmail.com>
> > >> wrote:
> > >> > hold the phone on this...............
> > >> >
> > >> > I just spun up an rc08 instance, and it looks like some of the
> mahout
> > >> > jobs
> > >> > actually failed .... because the mahout version is for hadoop 1x is
> > >> > being
> > >> > built (rather than 2x) . JIRA created.
> > >> >
> > >> > https://issues.apache.org/jira/browse/BIGTOP-1453
> > >> >
> > >
> > >
>
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Konstantin Boudnik <co...@apache.org>.
Thanks for testing the release candidate folks!
And sorry for the delays guys - this is officially the most crazy two weeks in
my life (and still not completely over). I have done some testing using 0.8.0
bits - HDFS ops, YARN w/ MR, some standalone Spark - all looks good.
I will try to update the Changes and cut the officially signed bits over the
weekend.
Cos
On Wed, Sep 24, 2014 at 07:59PM, Roman Shaposhnik wrote:
> Thanks for keeping us up-to-date on your progress. I plan to do some
> additional testing over the weekend as well.
>
> Cos is also working on cutting formal release bits.
>
> Thanks,
> Roman.
>
> On Wed, Sep 24, 2014 at 5:50 PM, Jean-Marc Spaggiari
> <je...@spaggiari.org> wrote:
> > Build 0.8 again today on CentOs without any issue. Will run applications
> > testing tomorrow.
> >
> > 2014-09-20 20:38 GMT-04:00 Roman Shaposhnik <ro...@shaposhnik.org>:
> >>
> >> First of all -- thanks a million for all the testing so far!
> >>
> >> Also, I seem to be able to use Mahout from RC0.8 just
> >> fine. Lets get to the bottom of this.
> >>
> >> Thanks,
> >> Roman.
> >>
> >> On Sat, Sep 20, 2014 at 5:57 AM, jay vyas <ja...@gmail.com>
> >> wrote:
> >> > hold the phone on this...............
> >> >
> >> > I just spun up an rc08 instance, and it looks like some of the mahout
> >> > jobs
> >> > actually failed .... because the mahout version is for hadoop 1x is
> >> > being
> >> > built (rather than 2x) . JIRA created.
> >> >
> >> > https://issues.apache.org/jira/browse/BIGTOP-1453
> >> >
> >
> >
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Konstantin Boudnik <co...@apache.org>.
Thanks for testing the release candidate folks!
And sorry for the delays guys - this is officially the most crazy two weeks in
my life (and still not completely over). I have done some testing using 0.8.0
bits - HDFS ops, YARN w/ MR, some standalone Spark - all looks good.
I will try to update the Changes and cut the officially signed bits over the
weekend.
Cos
On Wed, Sep 24, 2014 at 07:59PM, Roman Shaposhnik wrote:
> Thanks for keeping us up-to-date on your progress. I plan to do some
> additional testing over the weekend as well.
>
> Cos is also working on cutting formal release bits.
>
> Thanks,
> Roman.
>
> On Wed, Sep 24, 2014 at 5:50 PM, Jean-Marc Spaggiari
> <je...@spaggiari.org> wrote:
> > Build 0.8 again today on CentOs without any issue. Will run applications
> > testing tomorrow.
> >
> > 2014-09-20 20:38 GMT-04:00 Roman Shaposhnik <ro...@shaposhnik.org>:
> >>
> >> First of all -- thanks a million for all the testing so far!
> >>
> >> Also, I seem to be able to use Mahout from RC0.8 just
> >> fine. Lets get to the bottom of this.
> >>
> >> Thanks,
> >> Roman.
> >>
> >> On Sat, Sep 20, 2014 at 5:57 AM, jay vyas <ja...@gmail.com>
> >> wrote:
> >> > hold the phone on this...............
> >> >
> >> > I just spun up an rc08 instance, and it looks like some of the mahout
> >> > jobs
> >> > actually failed .... because the mahout version is for hadoop 1x is
> >> > being
> >> > built (rather than 2x) . JIRA created.
> >> >
> >> > https://issues.apache.org/jira/browse/BIGTOP-1453
> >> >
> >
> >
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Roman Shaposhnik <ro...@shaposhnik.org>.
Thanks for keeping us up-to-date on your progress. I plan to do some
additional testing over the weekend as well.
Cos is also working on cutting formal release bits.
Thanks,
Roman.
On Wed, Sep 24, 2014 at 5:50 PM, Jean-Marc Spaggiari
<je...@spaggiari.org> wrote:
> Build 0.8 again today on CentOs without any issue. Will run applications
> testing tomorrow.
>
> 2014-09-20 20:38 GMT-04:00 Roman Shaposhnik <ro...@shaposhnik.org>:
>>
>> First of all -- thanks a million for all the testing so far!
>>
>> Also, I seem to be able to use Mahout from RC0.8 just
>> fine. Lets get to the bottom of this.
>>
>> Thanks,
>> Roman.
>>
>> On Sat, Sep 20, 2014 at 5:57 AM, jay vyas <ja...@gmail.com>
>> wrote:
>> > hold the phone on this...............
>> >
>> > I just spun up an rc08 instance, and it looks like some of the mahout
>> > jobs
>> > actually failed .... because the mahout version is for hadoop 1x is
>> > being
>> > built (rather than 2x) . JIRA created.
>> >
>> > https://issues.apache.org/jira/browse/BIGTOP-1453
>> >
>
>
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Roman Shaposhnik <ro...@shaposhnik.org>.
Thanks for keeping us up-to-date on your progress. I plan to do some
additional testing over the weekend as well.
Cos is also working on cutting formal release bits.
Thanks,
Roman.
On Wed, Sep 24, 2014 at 5:50 PM, Jean-Marc Spaggiari
<je...@spaggiari.org> wrote:
> Build 0.8 again today on CentOs without any issue. Will run applications
> testing tomorrow.
>
> 2014-09-20 20:38 GMT-04:00 Roman Shaposhnik <ro...@shaposhnik.org>:
>>
>> First of all -- thanks a million for all the testing so far!
>>
>> Also, I seem to be able to use Mahout from RC0.8 just
>> fine. Lets get to the bottom of this.
>>
>> Thanks,
>> Roman.
>>
>> On Sat, Sep 20, 2014 at 5:57 AM, jay vyas <ja...@gmail.com>
>> wrote:
>> > hold the phone on this...............
>> >
>> > I just spun up an rc08 instance, and it looks like some of the mahout
>> > jobs
>> > actually failed .... because the mahout version is for hadoop 1x is
>> > being
>> > built (rather than 2x) . JIRA created.
>> >
>> > https://issues.apache.org/jira/browse/BIGTOP-1453
>> >
>
>
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Jean-Marc Spaggiari <je...@spaggiari.org>.
Build 0.8 again today on CentOs without any issue. Will run applications
testing tomorrow.
2014-09-20 20:38 GMT-04:00 Roman Shaposhnik <ro...@shaposhnik.org>:
> First of all -- thanks a million for all the testing so far!
>
> Also, I seem to be able to use Mahout from RC0.8 just
> fine. Lets get to the bottom of this.
>
> Thanks,
> Roman.
>
> On Sat, Sep 20, 2014 at 5:57 AM, jay vyas <ja...@gmail.com>
> wrote:
> > hold the phone on this...............
> >
> > I just spun up an rc08 instance, and it looks like some of the mahout
> jobs
> > actually failed .... because the mahout version is for hadoop 1x is being
> > built (rather than 2x) . JIRA created.
> >
> > https://issues.apache.org/jira/browse/BIGTOP-1453
> >
>
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Jean-Marc Spaggiari <je...@spaggiari.org>.
Build 0.8 again today on CentOs without any issue. Will run applications
testing tomorrow.
2014-09-20 20:38 GMT-04:00 Roman Shaposhnik <ro...@shaposhnik.org>:
> First of all -- thanks a million for all the testing so far!
>
> Also, I seem to be able to use Mahout from RC0.8 just
> fine. Lets get to the bottom of this.
>
> Thanks,
> Roman.
>
> On Sat, Sep 20, 2014 at 5:57 AM, jay vyas <ja...@gmail.com>
> wrote:
> > hold the phone on this...............
> >
> > I just spun up an rc08 instance, and it looks like some of the mahout
> jobs
> > actually failed .... because the mahout version is for hadoop 1x is being
> > built (rather than 2x) . JIRA created.
> >
> > https://issues.apache.org/jira/browse/BIGTOP-1453
> >
>
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Roman Shaposhnik <ro...@shaposhnik.org>.
First of all -- thanks a million for all the testing so far!
Also, I seem to be able to use Mahout from RC0.8 just
fine. Lets get to the bottom of this.
Thanks,
Roman.
On Sat, Sep 20, 2014 at 5:57 AM, jay vyas <ja...@gmail.com> wrote:
> hold the phone on this...............
>
> I just spun up an rc08 instance, and it looks like some of the mahout jobs
> actually failed .... because the mahout version is for hadoop 1x is being
> built (rather than 2x) . JIRA created.
>
> https://issues.apache.org/jira/browse/BIGTOP-1453
>
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Roman Shaposhnik <ro...@shaposhnik.org>.
First of all -- thanks a million for all the testing so far!
Also, I seem to be able to use Mahout from RC0.8 just
fine. Lets get to the bottom of this.
Thanks,
Roman.
On Sat, Sep 20, 2014 at 5:57 AM, jay vyas <ja...@gmail.com> wrote:
> hold the phone on this...............
>
> I just spun up an rc08 instance, and it looks like some of the mahout jobs
> actually failed .... because the mahout version is for hadoop 1x is being
> built (rather than 2x) . JIRA created.
>
> https://issues.apache.org/jira/browse/BIGTOP-1453
>
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by jay vyas <ja...@gmail.com>.
hold the phone on this...............
I just spun up an rc08 instance, and it looks like some of the mahout jobs
actually failed .... because the mahout version is for hadoop 1x is being
built (rather than 2x) . JIRA created.
https://issues.apache.org/jira/browse/BIGTOP-1453
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by jay vyas <ja...@gmail.com>.
hold the phone on this...............
I just spun up an rc08 instance, and it looks like some of the mahout jobs
actually failed .... because the mahout version is for hadoop 1x is being
built (rather than 2x) . JIRA created.
https://issues.apache.org/jira/browse/BIGTOP-1453
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Jay Vyas <ja...@gmail.com>.
Test results: mapred pig hive hbase and mahout all checked out , and bigpetstore ran as well start to finish. !
--- boring details ---
- ran tests for mapreduce, pig, hive, mahout, hbase: all worked for me. I tested both 4 and 3 node clusters w 2 g ram per node my testing was done manually for hive and mahout and using the protocol on bigtop-1449 for pig, mapreduce.
- I put some of my manual steps into automation via updating our vagrant recipe to actually run bigtop tests. see bigtop-1449 - that should make next rc testing completely automated on n nodes.
- I think for 0.9 We will need to update the hive tests to be simpler to run and interpret.. see bigtop-1450
Look forward to hearing about other tests you folks might have done.
...from my angle: rc0.8 works great!!!
> On Sep 14, 2014, at 10:00 PM, Roman Shaposhnik <rv...@apache.org> wrote:
>
> Hi!
>
> I just wanted to remind everybody that the bits
> currently published over here:
> http://bigtop01.cloudera.org:8080/view/Releases/job/Bigtop-0.8.0/
> are as close as it gets to Bigtop 0.8.0.
>
> We can't make a good release, without all
> of the help from everybody who can invest
> some amount of time in actually testing the
> thing.
>
> This coming week, Cos is likely to produce an
> official RC0 set of artifacts, but packages are
> already there -- lets test 'em!
>
> Thanks,
> Roman.
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by Jay Vyas <ja...@gmail.com>.
Test results: mapred pig hive hbase and mahout all checked out , and bigpetstore ran as well start to finish. !
--- boring details ---
- ran tests for mapreduce, pig, hive, mahout, hbase: all worked for me. I tested both 4 and 3 node clusters w 2 g ram per node my testing was done manually for hive and mahout and using the protocol on bigtop-1449 for pig, mapreduce.
- I put some of my manual steps into automation via updating our vagrant recipe to actually run bigtop tests. see bigtop-1449 - that should make next rc testing completely automated on n nodes.
- I think for 0.9 We will need to update the hive tests to be simpler to run and interpret.. see bigtop-1450
Look forward to hearing about other tests you folks might have done.
...from my angle: rc0.8 works great!!!
> On Sep 14, 2014, at 10:00 PM, Roman Shaposhnik <rv...@apache.org> wrote:
>
> Hi!
>
> I just wanted to remind everybody that the bits
> currently published over here:
> http://bigtop01.cloudera.org:8080/view/Releases/job/Bigtop-0.8.0/
> are as close as it gets to Bigtop 0.8.0.
>
> We can't make a good release, without all
> of the help from everybody who can invest
> some amount of time in actually testing the
> thing.
>
> This coming week, Cos is likely to produce an
> official RC0 set of artifacts, but packages are
> already there -- lets test 'em!
>
> Thanks,
> Roman.
Re: Civic duty reminder: please take Bigtop 0.8.0 RC0 for a spin
Posted by jay vyas <ja...@gmail.com>.
First tests look good ! Our vagrant recipes I was able to spin up a single
node cluster.
Now smoke testing and testing multinode !
*[root@bigtop1 vagrant]# jps25256 JobHistoryServer24596 NodeManager23792
ResourceManager27048 ThriftServer16903 WebAppProxyServer17464 NameNode25790
HRegionServer26370 HMaster28797 Jps15697 QuorumPeerMain18077 DataNode16247
SecondaryNameNode[root@bigtop1 vagrant]# hadoop versionHadoop
2.4.1Subversion https://git-wip-us.apache.org/repos/asf/bigtop.git
<https://git-wip-us.apache.org/repos/asf/bigtop.git> -r
8928bf946848ecf035d5da190c2fcbadcef7b994Compiled by jenkins on
2014-09-09T16:22ZCompiled with protoc 2.5.0From source with checksum
bb7ac0a3c73dc131f4844b873c74b630This command was run using
/usr/lib/hadoop/hadoop-common-2.4.1.jar*
On Sun, Sep 14, 2014 at 10:00 PM, Roman Shaposhnik <rv...@apache.org> wrote:
> Hi!
>
> I just wanted to remind everybody that the bits
> currently published over here:
> http://bigtop01.cloudera.org:8080/view/Releases/job/Bigtop-0.8.0/
> are as close as it gets to Bigtop 0.8.0.
>
> We can't make a good release, without all
> of the help from everybody who can invest
> some amount of time in actually testing the
> thing.
>
> This coming week, Cos is likely to produce an
> official RC0 set of artifacts, but packages are
> already there -- lets test 'em!
>
> Thanks,
> Roman.
>
--
jay vyas