running openstack-neat on centos 6.5

[fa...@staff.upnjatim.ac.id]

hi Anton, 

i try to install and running Openstack-neat on centos 6.5 got this error :

[root@controller openstack-neat]# ./all-start.sh

Traceback (most recent call last):

  File "./compute-data-collector-start.py", line 17, in <module>

    from neat.config import *

  File "/root/openstack-neat/neat/config.py", line 19, in <module>

    from contracts import contract

  File "/usr/lib/python2.6/site-packages/contracts/__init__.py", line 7, in <module>

    from . import syntax

  File "/usr/lib/python2.6/site-packages/contracts/syntax.py", line 79, in <module>

    from .library import (EqualTo, Unary, Binary, composite_contract,

  File "/usr/lib/python2.6/site-packages/contracts/library/__init__.py", line 28, in <module>

    from .array import (ShapeContract, Shape, Array, ArrayConstraint, DType,

  File "/usr/lib/python2.6/site-packages/contracts/library/array.py", line 6, in <module>

    from .array_ops import (ArrayOR, ArrayAnd, DType, ArrayConstraint,

  File "/usr/lib/python2.6/site-packages/contracts/library/array_ops.py", line 229, in <module>

    'np_float16': np.float16,  #  Half precision float: sign bit, 5 bits exponent, 10 bits mantissa

AttributeError: 'module' object has no attribute 'float16'

Starting openstack-neat-global-manager:                    [  OK  ]

Starting openstack-neat-db-cleaner:                        [  OK  ]

Traceback (most recent call last):

  File "./compute-local-manager-start.py", line 17, in <module>

    from neat.config import *

  File "/root/openstack-neat/neat/config.py", line 19, in <module>

    from contracts import contract

  File "/usr/lib/python2.6/site-packages/contracts/__init__.py", line 7, in <module>

    from . import syntax

  File "/usr/lib/python2.6/site-packages/contracts/syntax.py", line 79, in <module>

    from .library import (EqualTo, Unary, Binary, composite_contract,

  File "/usr/lib/python2.6/site-packages/contracts/library/__init__.py", line 28, in <module>

    from .array import (ShapeContract, Shape, Array, ArrayConstraint, DType,

  File "/usr/lib/python2.6/site-packages/contracts/library/array.py", line 6, in <module>

    from .array_ops import (ArrayOR, ArrayAnd, DType, ArrayConstraint,

  File "/usr/lib/python2.6/site-packages/contracts/library/array_ops.py", line 229, in <module>

    'np_float16': np.float16,  #  Half precision float: sign bit, 5 bits exponent, 10 bits mantissa

AttributeError: 'module' object has no attribute 'float16'

[root@controller openstack-neat]#

========================================================================

[root@controller setup]# ./deps-centos.sh

Loaded plugins: fastestmirror

Loading mirror speeds from cached hostfile

 * base: mirror.dionipe.net

 * epel: mirror.nus.edu.sg

 * extras: mirror.dionipe.net

 * updates: mirror.dionipe.net

Setting up Install Process

Package python-pip-1.3.1-4.el6.noarch already installed and latest version

Package numpy-1.4.1-9.el6.x86_64 already installed and latest version

Package scipy-0.7.2-8.el6.x86_64 already installed and latest version

Nothing to do

Requirement already up-to-date: pyqcy in /usr/lib/python2.6/site-packages

Requirement already up-to-date: PyContracts in /usr/lib/python2.6/site-packages

Requirement already up-to-date: SQLAlchemy in /usr/lib/python2.6/site-packages

Requirement already up-to-date: bottle in /usr/lib/python2.6/site-packages

Requirement already up-to-date: requests in /usr/lib/python2.6/site-packages

Requirement already up-to-date: Sphinx in /usr/lib/python2.6/site-packages

Requirement already up-to-date: python-novaclient in /usr/lib/python2.6/site-packages

Requirement already up-to-date: pyparsing<2.0.0 in /usr/lib/python2.6/site-packages (from PyContracts)

Requirement already up-to-date: decorator in /usr/lib/python2.6/site-packages (from PyContracts)

Requirement already up-to-date: Pygments>=1.2 in /usr/lib/python2.6/site-packages (from Sphinx)

Requirement already up-to-date: docutils>=0.7 in /usr/lib/python2.6/site-packages (from Sphinx)

Requirement already up-to-date: Jinja2>=2.3 in /usr/lib/python2.6/site-packages (from Sphinx)

Requirement already up-to-date: pbr>=0.6,!=0.7,<1.0 in /usr/lib/python2.6/site-packages (from python-novaclient)

Requirement already up-to-date: argparse in /usr/lib/python2.6/site-packages (from python-novaclient)

Requirement already up-to-date: iso8601>=0.1.9 in /usr/lib/python2.6/site-packages (from python-novaclient)

Requirement already up-to-date: PrettyTable>=0.7,<0.8 in /usr/lib/python2.6/site-packages (from python-novaclient)

Requirement already up-to-date: simplejson>=2.0.9 in /usr/lib/python2.6/site-packages (from python-novaclient)

Requirement already up-to-date: six>=1.7.0 in /usr/lib/python2.6/site-packages (from python-novaclient)

Requirement already up-to-date: Babel>=1.3 in /usr/lib/python2.6/site-packages (from python-novaclient)

Requirement already up-to-date: markupsafe in /usr/lib/python2.6/site-packages (from Jinja2>=2.3->Sphinx)

Requirement already up-to-date: pip in /usr/lib/python2.6/site-packages (from pbr>=0.6,!=0.7,<1.0->python-novaclient)

Requirement already up-to-date: pytz>=0a in /usr/lib/python2.6/site-packages (from Babel>=1.3->python-novaclient)

Cleaning up...

Requirement already satisfied (use --upgrade to upgrade): mocktest in /usr/lib/python2.6/site-packages

Requirement already satisfied (use --upgrade to upgrade): setuptools in /usr/lib/python2.6/site-packages (from mocktest)

Cleaning up...

[root@controller setup]#

======================================================================

[root@controller setup]# python

Python 2.6.6 (r266:84292, Jan 22 2014, 09:42:36)

[GCC 4.4.7 20120313 (Red Hat 4.4.7-4)] on linux2

Type "help", "copyright", "credits" or "license" for more information.

>>> import contracts

Traceback (most recent call last):

  File "<stdin>", line 1, in <module>

  File "/usr/lib/python2.6/site-packages/contracts/__init__.py", line 7, in <module>

    from . import syntax

  File "/usr/lib/python2.6/site-packages/contracts/syntax.py", line 79, in <module>

    from .library import (EqualTo, Unary, Binary, composite_contract,

  File "/usr/lib/python2.6/site-packages/contracts/library/__init__.py", line 28, in <module>

    from .array import (ShapeContract, Shape, Array, ArrayConstraint, DType,

  File "/usr/lib/python2.6/site-packages/contracts/library/array.py", line 6, in <module>

    from .array_ops import (ArrayOR, ArrayAnd, DType, ArrayConstraint,

  File "/usr/lib/python2.6/site-packages/contracts/library/array_ops.py", line 229, in <module>

    'np_float16': np.float16,  #  Half precision float: sign bit, 5 bits exponent, 10 bits mantissa

AttributeError: 'module' object has no attribute 'float16'

>>>

=============

how to solve this error ?

[fa...@staff.upnjatim.ac.id]

upgrading numpy using :

[root@compute2 openstack-neat]# pip install --upgrade numpy

solve that problem, but in the compute1 I got error like this:

Command /usr/bin/python -c "import setuptools, tokenize;__file__='/tmp/pip_build_root/numpy/setup.py';exec(compile(getattr(tokenize, 'open', open)(__file__).read().replace('\r\n', '\n'), __file__, 'exec'))" install --record /tmp/pip-8iN8nv-record/install-record.txt --single-version-externally-managed --compile failed with error code 1 in /tmp/pip_build_root/numpy

Traceback (most recent call last):

  File "/usr/bin/pip", line 9, in <module>

    load_entry_point('pip==1.5.6', 'console_scripts', 'pip')()

  File "/usr/lib/python2.6/site-packages/pip/__init__.py", line 185, in main

    return command.main(cmd_args)

  File "/usr/lib/python2.6/site-packages/pip/basecommand.py", line 161, in main

    text = '\n'.join(complete_log)

UnicodeDecodeError: 'ascii' codec can't decode byte 0xe2 in position 70: ordinal not in range(128)

so, import contracts :

[root@compute1 ~]# python

Python 2.6.6 (r266:84292, Jan 22 2014, 09:42:36)

[GCC 4.4.7 20120313 (Red Hat 4.4.7-4)] on linux2

Type "help", "copyright", "credits" or "license" for more information.

>>> import contracts

Traceback (most recent call last):

  File "<stdin>", line 1, in <module>

  File "/usr/lib/python2.6/site-packages/contracts/__init__.py", line 7, in <module>

    from . import syntax

  File "/usr/lib/python2.6/site-packages/contracts/syntax.py", line 79, in <module>

    from .library import (EqualTo, Unary, Binary, composite_contract,

  File "/usr/lib/python2.6/site-packages/contracts/library/__init__.py", line 28, in <module>

    from .array import (ShapeContract, Shape, Array, ArrayConstraint, DType,

  File "/usr/lib/python2.6/site-packages/contracts/library/array.py", line 6, in <module>

    from .array_ops import (ArrayOR, ArrayAnd, DType, ArrayConstraint,

  File "/usr/lib/python2.6/site-packages/contracts/library/array_ops.py", line 229, in <module>

    'np_float16': np.float16,  #  Half precision float: sign bit, 5 bits exponent, 10 bits mantissa

AttributeError: 'module' object has no attribute 'float16'

any idea ??

[Anton Beloglazov]

Hi, it seems like it's again a numpy problem. Have you tried to reinstall it on compute1 using pip?

Cheers,

Anton

--
You received this message because you are subscribed to the Google Groups "OpenStack Neat" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openstack-nea...@googlegroups.com.
Visit this group at http://groups.google.com/group/openstack-neat.
For more options, visit https://groups.google.com/d/optout.

[fa...@staff.upnjatim.ac.id]

yeah ... after rebooting all host and reinstalling numpy, finally all openstack-neat service running, but ...

in the controller i can't start openstack-keystone with this error :

[root@controller config]# keystone user-list

Authorization Failed: Unable to communicate with identity service: [Errno 111] Connection refused. (HTTP 400)

[root@controller config]# /etc/init.d/openstack-keystone restart

Stopping keystone:                                         [FAILED]

Starting keystone:                                         [  OK  ]

[root@controller config]# /etc/init.d/openstack-keystone start

Starting keystone:                                         [  OK  ]

[root@controller config]# /etc/init.d/openstack-keystone status

keystone dead but pid file exists

[root@controller config]# tail -f /var/log/keystone/keystone.log 

2014-07-15 18:00:00 CRITICAL [root] cannot import name exceptions

2014-07-15 18:00:02 CRITICAL [root] cannot import name exceptions

2014-07-15 18:00:05 CRITICAL [root] cannot import name exceptions

2014-07-15 18:02:49 CRITICAL [root] cannot import name exceptions

2014-07-15 18:02:52 CRITICAL [root] cannot import name exceptions

2014-07-15 18:02:54 CRITICAL [root] cannot import name exceptions

2014-07-15 18:07:56 CRITICAL [root] cannot import name exceptions

2014-07-15 18:47:38 CRITICAL [root] cannot import name exceptions

2014-07-15 18:51:01 CRITICAL [root] cannot import name exceptions

2014-07-15 18:51:07 CRITICAL [root] cannot import name exceptions

any idea?

On Tuesday, July 15, 2014 9:32:10 AM UTC+7, fa...@staff.upnjatim.ac.id wrote:

[Anton Beloglazov]

I haven't seen this error before, but this seems relevant: https://answers.launchpad.net/nova/+question/145697

Cheers,

Anton

--

[fa...@staff.upnjatim.ac.id]

in the first time installing openstack-neat, i also got that error especially after todo this step :

sudo pip-python install --upgrade pyqcy PyContracts SQLAlchemy bottle requests Sphinx python-novaclient

and i rollback using uninstalling pip command :

sudo pip-python uninstall pyqcy PyContracts SQLAlchemy bottle requests Sphinx python-novaclient

my openstack-keystone can start again.

which package do you think conflict with openstack-keystone service?

[Anton Beloglazov]

From the link I sent you, seems like it could be SQLAlchemy.

[fa...@staff.upnjatim.ac.id]

yeah ... based on your guide (https://answers.launchpad.net/nova/+question/145697) and upgrading sqlalchemy-migrate, my openstack-keystone can start i gain

thank you Anton :)

but i still confuse with openstack-nova-api and openstack-nova-metadata-api, that service cannot running together, so i stop openstack-nova-metada-api and restart openstack-nova-api ... openstack cloud can run especially when accessing dashboard.

any idea?

[Anton Beloglazov]

Great! openstack-nova-metadata-api is queried by VMs to obtain SSH keys and other user data, while openstack-nova-api is used to perform operations on VMs themselves. I'm not sure why you have a conflict between them. Maybe something's changed in recent versions of OpenStack.

Cheers,

Anton

[fa...@staff.upnjatim.ac.id]

Hi Anton, long time no see

after openstack-neat running, 

- what first to start for the next action?

- how to test the system that running well for vm migration?

i am really newbie

thanks again for your help

[Anton Beloglazov]

Hi Fadil,

Please have a look at Chapter 6 of my thesis (http://beloglazov.info/thesis.pdf) where I described my experiments with OpenStack Neat. There is also a more recent paper on the same topic (http://beloglazov.info/papers/2014-ccpe-openstack-neat.pdf), but it contains fewer technical details than the thesis. Scripts that I used for running experiments are available in the following repository: https://github.com/beloglazov/ccpe-2014-experiments

Best regards,

Anton

[fa...@staff.upnjatim.ac.id]

Thanks before,

is there a tutorial to running https://github.com/beloglazov/ccpe-2014-experiments ?

[Anton Beloglazov]

The following post has some information on running experiments: https://groups.google.com/d/msg/openstack-neat/PKz2vpKPMcA/-eImY47z1rMJ

Cheers,

Anton

[fa...@staff.upnjatim.ac.id]

yes, i follow that link with step :

1. ./vms-boot-28-slow.sh # to boot VMs >>> done

2. ./disable-distributor.sh # temporarily disable workload distribution >>> done

2. ./workload-distributor.py # then check to log file (workload-distributor.log) to see whether all VMs are ready and send requests for the workload >>> i do with this command on controller #python2 workload-distributor.py planetlab-filtered/ , is it right?

3. ./enable-distributor.sh # enable the workload distributor >>> done

4. ./workload-distributor.py # distribute the workload to VMs (send a unique workload trace file to each VM) >>> what is difference with step 2 ?

5. Once the experiment is completed (24 hours), scp the log files from the compute and controller nodes and use scripts from the results directory to compute the result statistics

and the next question is same with Albert Vonpupp on that threads. as you said on next threads that workload traces and cpu load generator should be uploaded or clone to the VM image, is it done by manually or there are scripts to do that?

thanks before 

[Anton Beloglazov]

Hi Fadil,

These steps are correct. There is no script to create a VM image, so you would have to do it yourself. Sorry, it's not really documented, but on line 40-41 in https://github.com/beloglazov/ccpe-2014-experiments/blob/master/workload-distributor.py#L40 you can see the command that will be executed by the workload distributor on the VMs. Basically, it starts the CPU load generator and passes in a workload trace. Both the cpu-load-generator.py script and the traces should already be on the VM. Please modify those lines to match your paths to the CPU load generator and the traces.

Cheers,

Anton