cloudcredo / cloudrocker Goto Github PK

View Code? Open in Web Editor NEW

34.0 34.0 6.0 1.73 MB

License: Apache License 2.0

Go 92.24% Shell 4.07% Java 0.39% HTML 0.76% JavaScript 0.39% PHP 0.09% Ruby 1.63% Makefile 0.18% Python 0.24%

cloudrocker's People

Contributors

Stargazers

Watchers

Forkers

jbayer ronakbanka svrc dambor raphaelaudet morika-t

cloudrocker's Issues

had to create /app/tmp directory in docker image to get php sample to work

here are some command outputs from the fock build image. note the error:
mkfifo: cannot create fifo `/app/tmp/heroku.waitpipe-8080.fycbjO': No such file or directory"
and the mkdir command and then see it working.

 vcap@b3033f667ef4:~$ source ./.profile.d/php.sh
 vcap@b3033f667ef4:~$ /bin/bash cloudfocker-start-1c4352a23e52040ddb1857d7675fe3cc.sh /app vendor/bin/heroku-php-apache2
 Booting on port 8080...
 Using PHP-FPM configuration file 'vendor/heroku/heroku-buildpack-php/conf/php/php-fpm.conf'
 Using PHP configuration (php.ini) file 'vendor/heroku/heroku-buildpack-php/conf/php/php.ini'
 Using Apache2 VirtualHost-level configuration include 'vendor/heroku/heroku-buildpack-php/conf/apache2/default_include.conf'
 Using Apache2 configuration file 'vendor/heroku/heroku-buildpack-php/conf/apache2/heroku.conf'
 mkfifo: cannot create fifo `/app/tmp/heroku.waitpipe-8080.fycbjO': No such file or directory
 vcap@b3033f667ef4:~$ mkdir /app/tmp
 vcap@b3033f667ef4:~$ /bin/bash cloudfocker-start-1c4352a23e52040ddb1857d7675fe3cc.sh /app vendor/bin/heroku-php-apache2
 Booting on port 8080...
 Using PHP-FPM configuration file 'vendor/heroku/heroku-buildpack-php/conf/php/php-fpm.conf'
 Using PHP configuration (php.ini) file 'vendor/heroku/heroku-buildpack-php/conf/php/php.ini'
 Using Apache2 VirtualHost-level configuration include 'vendor/heroku/heroku-buildpack-php/conf/apache2/default_include.conf'
 Using Apache2 configuration file 'vendor/heroku/heroku-buildpack-php/conf/apache2/heroku.conf'
 Starting log redirection...
 Starting php-fpm...
 Starting httpd...
 [14-Dec-2014 08:16:47] NOTICE: [pool www] 'user' directive is ignored when FPM is not running as root
 [14-Dec-2014 08:16:47] NOTICE: fpm is running, pid 131
 [14-Dec-2014 08:16:47] NOTICE: ready to handle connections
 127.0.0.1 - - [14/Dec/2014:08:17:49 +0000] "GET / HTTP/1.1" 200 65 "-" "curl/7.19.7 (x86_64-pc-linux-gnu) libcurl/7.19.7 OpenSSL/0.9.8k zlib/1.2.3.3 libidn/1.15"

Can't build spring-music application

I want run spring-music application on lattice, so I try to pack it into docker container with cloud rocket. But when I fock up in spring-music directory I got an error:

/vagrant/spring-music$ fock up
Starting the CloudFocker container...
Running Buildpacks...
Started the CloudFocker container.
Deleting the CloudFocker container...
cloudfocker-staging
Deleted container.
2015/02/16 15:53:32  Staging failed - have you added a buildpack for this type of application?

I have already installed java buildpack.

$ fock buildpacks
java-buildpack

Application was built with ./gradlew assemble

forgetting to 'rock this' results in unintuitive error message

i forgot to rock this and it would be nice to defensively inform the user that they should do it.

$ rock build
Starting the CloudRocker container...
Unable to find image 'cloudrocker-base:latest' locally
Pulling repository cloudrocker-base
2015/05/09 21:10:59 Error: Error: image library/cloudrocker-base:latest not found

Files beginning with a . in the application root are ignored

While I was testing the multi-buildpack I run into an issue where it would not be detected.
After further investigation I found the problem was related to the .buildpacks file which was not copied into the staging environment.
I suspect it being related to these lines.

go sample does not fock build

$ fock build
Starting the CloudFocker container...
Running Buildpacks...
-------> Buildpack version 1.1.0
! A .godir is required. For instructions:
! http://mmcgrana.github.io/2012/09/getting-started-with-go-on-heroku
Started the CloudFocker container.
Deleting the CloudFocker container...
cloudfocker-staging
Deleted container.
2014/12/13 19:38:34 Staging failed - have you added a buildpack for this type of application?
vagrant@ubuntu-14:/vagrant/sample-apps/go$ fock buildpacks
go-buildpack
java-buildpack

rock cannot adapt to docker 1.8.1.

bash-4.3$ docker -v
Docker version 1.8.1, build d12ea79
bash-4.3$ rock docker
Checking Docker version
Client API version: 1.15
Go version (client): go1.5
OS/Arch (client): darwin/amd64
2015/09/02 09:41:44 Error: Get http://%2Fvar%2Frun%2Fdocker.sock/v1.15/version: dial unix /var/run/docker.sock: connect: no such file or directory

`rock this` fails when run as root

UID 0 already exists in the raw image.

mamacdon@ubuntud:~$ sudo ./rocker/bin/rock this
Bootstrapping Docker setup - this will take a few minutes...
Downloading from https://s3.amazonaws.com/blob.cfblob.com/1030f8d6-15c8-48d6-a99f-db8742b136f4
acb06367c15f26cae90ef22f62f0e2a20bcf8e828174f730fe6b8e1fd119095b
Creating image configuration...
Creating image...
Sending build context to Docker daemon  2.56 kB
Sending build context to Docker daemon 
Step 0 : FROM cloudrocker-raw:latest
 ---> acb06367c15f
Step 1 : RUN /usr/sbin/useradd -mU -u 0 -d /app -s /bin/bash vcap
 ---> Running in 6782b2a5741b
useradd: UID 0 is not unique
2015/05/15 11:45:41 Error: Status: The command [/bin/sh -c /usr/sbin/useradd -mU -u 0 -d /app -s /bin/bash vcap] returned a non-zero code: 4, Code: 1

Unable to use fuse bases filesystems inside container

While trying to use a fuse based filesystem I found a difference between the permissions inside the docker container spawned by cloudrocker vs the permissions in the warden based container.

# cloudrocker
> ls -la /dev/fuse
c--------- 1 root root 10, 229 Aug  6 21:57 /dev/fuse

# cloudfoundry
> ls -la /dev/fuse
crw-rw-rw- 1 root fuse 10, 229 2015-07-05 15:22 /dev/fuse

Which results in fuse: failed to open /dev/fuse: Permission denied in the cloudrocker environment.

ruby sample does not work on diego without setting HOME=/app

diego value for HOME is /home/vcap
docker image HOME is /app

when running the docker image /app does not appear to be symlinked to /home/vcap, so when running the ruby sample on lattice with diego i had to do a command like this to get the /app/.profile.d/ruby.sh script to have the correct values:

 lattice-cli start ruby-example -i "docker:///jbayer/ruby-example" -- /bin/bash /app/cloudfocker-start-1c4352a23e52040ddb1857d7675fe3cc.sh /app export HOME=/app \&\& source /app/.profile.d/ruby.sh \&\& env \&\& /app/vendor/bundle/ruby/1.9.1/bin/bundle version \&\& /app/vendor/bundle/ruby/1.9.1/bin/bundle exec rackup config.ru -p 8080

404 for https://s3.amazonaws.com/cloudrocker/vagrantboxes/cloudrocker-0.0.2-vbox.box

Hi,

I'm trying to do the vagrant provision and getting a 404 when trying to download the vbox: https://s3.amazonaws.com/cloudrocker/vagrantboxes/cloudrocker-0.0.2-vbox.box

Trying it in a browser also yields a 404.

invalid command in docker image after fock build

running fock build on a simple spring boot app.groovy results in an invalid command entry in the docker image.

"Cmd": [
            "/bin/sh",
            "-c",
            "#(nop) CMD [/bin/sh -c [\"/bin/bash\", \"/app/cloudfocker-start-1c4352a23e52040ddb1857d7675fe3cc.sh\", \"/app\", \"JAVA_HOME=$PWD/.java-buildpack/open_jdk_jre\", \"JAVA_OPTS=\"-Djava.io.tmpdir=/app/tmp\", \"-XX:OnOutOfMemoryError=$PWD/.java-buildpack/open_jdk_jre/bin/killjava.sh\", \"-XX:MaxMetaspaceSize=64M\", \"-XX:MetaspaceSize=64M\"\", \"SERVER_PORT=8080\", \"$PWD/.java-buildpack/spring_boot_cli/bin/spring\", \"run\", \"app.groovy\"]]"
        ]

the full detail is here: https://gist.github.com/jbayer/f834e5a08e32bf68519f#file-gistfile1-txt-L50

simple app.groovy is here:

@Controller
class ThisWillActuallyRun {

    @RequestMapping("/")
    @ResponseBody
    String home() {
        return "BLUE"
    }

    @RequestMapping("/services")
    @ResponseBody
    String services() {
        return System.getenv("VCAP_SERVICES")
    }

}

python sample app errors with uid not found: 1000

the sample flask app works fine on run.pivotal.io, so i assume it's some difference in the cloud rocker env from the normal cloud foundry environment.

$ fock add-buildpack https://github.com/cloudfoundry/python-buildpack.git
Downloading buildpack...
Cloning into 'python-buildpack'...
remote: Counting objects: 152, done.
remote: Compressing objects: 100% (90/90), done.
remote: Total 152 (delta 59), reused 123 (delta 58)
Receiving objects: 100% (152/152), 8.19 MiB | 3.03 MiB/s, done.
Resolving deltas: 100% (59/59), done.
Checking connectivity... done.
Submodule 'compile-extensions' (https://github.com/cf-buildpacks/compile-extensions.git) registered for path 'compile-extensions'
Cloning into 'compile-extensions'...
remote: Counting objects: 203, done.
remote: Compressing objects: 100% (102/102), done.
remote: Total 203 (delta 66), reused 199 (delta 64)
Receiving objects: 100% (203/203), 35.48 KiB | 0 bytes/s, done.
Resolving deltas: 100% (66/66), done.
Checking connectivity... done.
Submodule path 'compile-extensions': checked out '1b84b6955b115c84a552eaed2a28b1ae5dfb0456'
Downloaded buildpack.
vagrant@ubuntu-14:/vagrant/sample-apps/python$ fock up
Starting the CloudFocker container...
Running Buildpacks...
-------> Buildpack version 1.1.1
-----> Installing runtime (python-2.7.9)
-----> Installing dependencies with pip
Traceback (most recent call last):
  File "/app/.heroku/python/bin/pip", line 9, in <module>
    load_entry_point('pip==1.5.6', 'console_scripts', 'pip')()
  File "build/bdist.linux-x86_64/egg/pkg_resources.py", line 339, in load_entry_point
  File "build/bdist.linux-x86_64/egg/pkg_resources.py", line 2470, in load_entry_point
  File "build/bdist.linux-x86_64/egg/pkg_resources.py", line 2184, in load
  File "/app/.heroku/python/lib/python2.7/site-packages/pip-1.5.6-py2.7.egg/pip/__init__.py", line 10, in <module>
    from pip.util import get_installed_distributions, get_prog
  File "/app/.heroku/python/lib/python2.7/site-packages/pip-1.5.6-py2.7.egg/pip/util.py", line 15, in <module>
    from pip.locations import site_packages, running_under_virtualenv, virtualenv_no_global
  File "/app/.heroku/python/lib/python2.7/site-packages/pip-1.5.6-py2.7.egg/pip/locations.py", line 96, in <module>
    build_prefix = _get_build_prefix()
  File "/app/.heroku/python/lib/python2.7/site-packages/pip-1.5.6-py2.7.egg/pip/locations.py", line 65, in _get_build_prefix
    __get_username())
  File "/app/.heroku/python/lib/python2.7/site-packages/pip-1.5.6-py2.7.egg/pip/locations.py", line 60, in __get_username
    return pwd.getpwuid(os.geteuid()).pw_name
KeyError: 'getpwuid(): uid not found: 1000'
Traceback (most recent call last):
  File "/app/.heroku/python/bin/pip", line 9, in <module>
    load_entry_point('pip==1.5.6', 'console_scripts', 'pip')()
  File "build/bdist.linux-x86_64/egg/pkg_resources.py", line 339, in load_entry_point
  File "build/bdist.linux-x86_64/egg/pkg_resources.py", line 2470, in load_entry_point
  File "build/bdist.linux-x86_64/egg/pkg_resources.py", line 2184, in load
  File "/app/.heroku/python/lib/python2.7/site-packages/pip-1.5.6-py2.7.egg/pip/__init__.py", line 10, in <module>
    from pip.util import get_installed_distributions, get_prog
  File "/app/.heroku/python/lib/python2.7/site-packages/pip-1.5.6-py2.7.egg/pip/util.py", line 15, in <module>
    from pip.locations import site_packages, running_under_virtualenv, virtualenv_no_global
  File "/app/.heroku/python/lib/python2.7/site-packages/pip-1.5.6-py2.7.egg/pip/locations.py", line 96, in <module>
    build_prefix = _get_build_prefix()
  File "/app/.heroku/python/lib/python2.7/site-packages/pip-1.5.6-py2.7.egg/pip/locations.py", line 65, in _get_build_prefix
    __get_username())
  File "/app/.heroku/python/lib/python2.7/site-packages/pip-1.5.6-py2.7.egg/pip/locations.py", line 60, in __get_username
    return pwd.getpwuid(os.geteuid()).pw_name
KeyError: 'getpwuid(): uid not found: 1000'

cp: `/app/.heroku' and `/app/.heroku' are the same file
cp: `/app/.profile.d' and `/app/.profile.d' are the same file
cp: `/app/requirements.txt' and `/app/requirements.txt' are the same file
cp: `/app/runtime.txt' and `/app/runtime.txt' are the same file
Started the CloudFocker container.
Deleting the CloudFocker container...
cloudfocker-staging
Deleted container.
2015/01/03 00:55:45  Staging failed - have you added a buildpack for this type of application?

staging fails for python app

I am trying to build a python buildpack based application. rock up fails with below message.

It could well be that this is related to my special setup of package repo and vendor folder. But the error message is very unspecific. How to increase verbosity or debug otherwise?

Running Buildpacks...
-------> Buildpack version 1.5.23
-----> Installing python-3.5.3
Downloaded [https://buildpacks.cloudfoundry.org/dependencies/python/python-3.5.3-linux-x64-105f02b2.tgz]
-----> Installing Setuptools (32.1.0)
Downloaded [https://buildpacks.cloudfoundry.org/dependencies/manual-binaries/setuptools/setuptools-32.1.0-44054eaa.tar.gz]
-----> Installing Pip (9.0.1)
Downloaded [https://buildpacks.cloudfoundry.org/dependencies/manual-binaries/pip/pip-9.0.1-35f01da3.tar.gz]
-----> Installing pip-pop (0.1.1)
Downloaded [https://buildpacks.cloudfoundry.org/dependencies/manual-binaries/pip-pop/pip-pop-0.1.1-d410583a.tar.gz]
-----> Installing pipenv (4.0.1)
Downloaded [https://buildpacks.cloudfoundry.org/dependencies/manual-binaries/pipenv/pipenv-4.0.1-148f753f.tar.gz]
     $ pip install -r requirements.txt
 Failed to compile droplet: exit status 1Deleting the CloudRocker container...
Deleted container.
2017/08/10 17:02:59  Staging failed - have you added a buildpack for this type of application?

Make sure to set CF_STACK during staging on cflinuxfs2

When using the cflinuxfs2 stack I'm running into issues with the ruby buildpack (v1.6.5).
The error below is raised because it expects the CF_STACK env far to be set to cflinuxfs2.
I was able to verify the this by ENV['CF_STACK'] = 'cflinuxfs2' to the top of the compile script, which fixed the issue.

2015/08/22 13:10:23  Staging failed - have you added a buildpack for this type of application?
Starting the CloudRocker container...
Running Buildpacks...
It looks like you're deploying on a stack (currently set to *lucid64*) that's not supported by this buildpack.
That could be because you're using a recent buildpack release on a deprecated stack.
If you're using the buildpack installed by your CF admin, please let your admin know you saw this error message.
If you at one point specified a buildpack that's at git URL, please make sure you're pointed at a version that supports this stack.

 Failed to compile droplet: exit status 44

Clarify problems with boot2docker?

Can cloudrocker be made to work under boot2docker? The README says:

Feel free to try Cloud Rocker with boot2docker, and good luck with the volume mounts.

I took this to mean that it was at least possible (maybe I misread some sarcasm there). Anyway, when I rock up or rock build, I get an error:

docker@boot2docker:/c/Users/mamacdon/code/myapp$ rock build
Starting the CloudRocker container...
no such file or directory
2015/04/27 18:42:32 Error: Error response from daemon: Cannot start container 4f1d198adcc6b7c96a5c87666271d61b2c5cb151d66e9f91f92c9573df34d5bc: [8] System error: no such file or directory
docker@boot2docker:/c/Users/mamacdon/code/myapp$

It would be nice to have more detail in the docs about why this will or won't work.