You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@whirr.apache.org by Paul Baclace <pa...@gmail.com> on 2011/10/29 04:10:48 UTC

building trunk whirr, and deployment

What is the right way to build whirr using mvn on my laptop and then 
how/what to deploy on my ec2 launcher machine (a tiny instance)?

What I have tried so far:

On laptop::
   cd trunk
   #  is this correct goal?
   mvn install
   #  copy the whole thing?  really?  or some clever subset of files?
   scp -r . my_ec2_machine:/tmp/whirr7

On my_ec2_machine::
   export WHIRR_HOME=/tmp/whirr7
   WHIRR_BIN=$WHIRR_HOME/bin/whirr
   ... run my script that runs $WHIRR_BIN

if $WHIRR_HOME/services/*/src/main/resources/functions/*.sh files are 
changed in place where deployed, I think that works for quick testing of 
experimental changes.

One more vital question:  how do I build trunk jclouds and correctly 
update the multitude of cli/target/lib/jclouds-*.jar files in whirr so 
that whirr actually uses them?  If the jar filename changes what xml 
files need to be changed in jclouds and whirr?

Thanks!

Paul


Re: building trunk whirr, and deployment

Posted by Adrian Cole <ad...@jclouds.org>.
P.S.  I have opened and attached patches for a couple things that can also help:

  * WHIRR-417 -> allows you to pass a property to change jclouds
modules.  Ex. if for some reason you have doubts about sshj, you can
switch to jsch.
  * WHIRR-418 -> logs ssh connection details to jclouds-ssh.log

I hope these help!
-Adrian

On Sat, Oct 29, 2011 at 9:10 AM, Adrian Cole <ad...@jclouds.org> wrote:
> Hi, Paul.
>
> There's also this post, which automates this process:
>
> http://www.google.com/url?sa=t&rct=j&q=whirr%20%20jclouds%20compute-basics&source=web&cd=1&ved=0CBkQFjAA&url=http%3A%2F%2Fmail-archives.apache.org%2Fmod_mbox%2Fwhirr-dev%2F201108.mbox%2F%253CCAARMrHXi1ad5fkGvSYJRYJB7OsDmsOtzLgbR%3D0G9KNP-FdjBeA%40mail.gmail.com%253E&ei=laarTrrECebi4QTUxOiJDw&usg=AFQjCNF4Ix6I6m0o0pOws1v1Oi_sb05awg&sig2=caDzIifCj-RxEQ_bSqDI7Q
>
> This could be brushed up and/or made as formalized tool.
>
> -A
>
> On Sat, Oct 29, 2011 at 4:10 AM, Paul Baclace <pa...@gmail.com> wrote:
>> What is the right way to build whirr using mvn on my laptop and then
>> how/what to deploy on my ec2 launcher machine (a tiny instance)?
>>
>> What I have tried so far:
>>
>> On laptop::
>>  cd trunk
>>  #  is this correct goal?
>>  mvn install
>>  #  copy the whole thing?  really?  or some clever subset of files?
>>  scp -r . my_ec2_machine:/tmp/whirr7
>>
>> On my_ec2_machine::
>>  export WHIRR_HOME=/tmp/whirr7
>>  WHIRR_BIN=$WHIRR_HOME/bin/whirr
>>  ... run my script that runs $WHIRR_BIN
>>
>> if $WHIRR_HOME/services/*/src/main/resources/functions/*.sh files are
>> changed in place where deployed, I think that works for quick testing of
>> experimental changes.
>>
>> One more vital question:  how do I build trunk jclouds and correctly update
>> the multitude of cli/target/lib/jclouds-*.jar files in whirr so that whirr
>> actually uses them?  If the jar filename changes what xml files need to be
>> changed in jclouds and whirr?
>>
>> Thanks!
>>
>> Paul
>>
>>
>

Re: building trunk whirr, and deployment

Posted by Adrian Cole <ad...@jclouds.org>.
Hi, Paul.

There's also this post, which automates this process:

http://www.google.com/url?sa=t&rct=j&q=whirr%20%20jclouds%20compute-basics&source=web&cd=1&ved=0CBkQFjAA&url=http%3A%2F%2Fmail-archives.apache.org%2Fmod_mbox%2Fwhirr-dev%2F201108.mbox%2F%253CCAARMrHXi1ad5fkGvSYJRYJB7OsDmsOtzLgbR%3D0G9KNP-FdjBeA%40mail.gmail.com%253E&ei=laarTrrECebi4QTUxOiJDw&usg=AFQjCNF4Ix6I6m0o0pOws1v1Oi_sb05awg&sig2=caDzIifCj-RxEQ_bSqDI7Q

This could be brushed up and/or made as formalized tool.

-A

On Sat, Oct 29, 2011 at 4:10 AM, Paul Baclace <pa...@gmail.com> wrote:
> What is the right way to build whirr using mvn on my laptop and then
> how/what to deploy on my ec2 launcher machine (a tiny instance)?
>
> What I have tried so far:
>
> On laptop::
>  cd trunk
>  #  is this correct goal?
>  mvn install
>  #  copy the whole thing?  really?  or some clever subset of files?
>  scp -r . my_ec2_machine:/tmp/whirr7
>
> On my_ec2_machine::
>  export WHIRR_HOME=/tmp/whirr7
>  WHIRR_BIN=$WHIRR_HOME/bin/whirr
>  ... run my script that runs $WHIRR_BIN
>
> if $WHIRR_HOME/services/*/src/main/resources/functions/*.sh files are
> changed in place where deployed, I think that works for quick testing of
> experimental changes.
>
> One more vital question:  how do I build trunk jclouds and correctly update
> the multitude of cli/target/lib/jclouds-*.jar files in whirr so that whirr
> actually uses them?  If the jar filename changes what xml files need to be
> changed in jclouds and whirr?
>
> Thanks!
>
> Paul
>
>

Re: building trunk whirr, and deployment

Posted by Andrei Savu <sa...@gmail.com>.
Some basic instructions to build a SNAPSHOT release extracted from:

https://cwiki.apache.org/confluence/display/WHIRR/How+To+Release

1. Checkout the trunk from SVN or GIT:


svn checkout http://svn.apache.org/repos/asf/whirr/trunk/ whirr-trunk

OR


git clone git://git.apache.org/whirr.git


2. Updated files as needed. If you want to use a SNAPSHOP version of jclouds
create a local checkout, update and build Whirr using those artefacts with
the following commands:

git clone git://github.com/jclouds/jclouds.git
cd jclouds
mvn -DskipTest clean install  (this will install 1.3.0-SNAPSHOT to the local
maven repository)

* update whirr-trunk/pom.xml to use the snapshot release 1.2.1 ->
1.3.0-SNAPSHOT

3. Build the Whirr binary release:

mvn clean package assembly:assembly

4. Done! You can find the SNAPSHOT release tarball in
 target/whirr-0.7.0-SNAPSHOT.tar.gz

5. If you later on decide that you need to update some of the bash scripts
you can create a functions folder in the Whirr root and override the
existing code (e.g. functions/install_hadoop.sh will override
services/hadoop/../functions/install_hadoop.sh).

See http://whirr.apache.org/faq.html  - "How can I modify the instance
installation and configuration scripts?"

I hope you will find this to be clear enough.

Cheers,

-- Andrei Savu / andreisavu.ro

On Sat, Oct 29, 2011 at 5:10 AM, Paul Baclace <pa...@gmail.com>wrote:

> What is the right way to build whirr using mvn on my laptop and then
> how/what to deploy on my ec2 launcher machine (a tiny instance)?
>
> What I have tried so far:
>
> On laptop::
>  cd trunk
>  #  is this correct goal?
>  mvn install
>  #  copy the whole thing?  really?  or some clever subset of files?
>  scp -r . my_ec2_machine:/tmp/whirr7
>
> On my_ec2_machine::
>  export WHIRR_HOME=/tmp/whirr7
>  WHIRR_BIN=$WHIRR_HOME/bin/**whirr
>  ... run my script that runs $WHIRR_BIN
>
> if $WHIRR_HOME/services/*/src/**main/resources/functions/*.sh files are
> changed in place where deployed, I think that works for quick testing of
> experimental changes.
>
> One more vital question:  how do I build trunk jclouds and correctly update
> the multitude of cli/target/lib/jclouds-*.jar files in whirr so that whirr
> actually uses them?  If the jar filename changes what xml files need to be
> changed in jclouds and whirr?
>
> Thanks!
>
> Paul
>
>