OpenStack Messaging (Zaqar)
Go to file
Flavio Percoco 35f2764204 Remove sqlalchemy from our data plane
This patch removes the sqlalchemy as a valid data driver. In order to
make this work properly, we also need the data plane to be properly
split into 2 completely independent pipelines.

This patch, unfortunately, disables some of the sqlalchemy's tests for
queues and pools. The reason is that, until these planes are correctly
separated and pooling is enabled by default, we won't be able to
properly run those tests. A follow-up patch, expected to land in kilo,
will do this.

DocImpact

Implements blueprint: disable-sqlalchemy-as-pool

Change-Id: I5ca73102241692b0e1b23b3eaaaf1938f0106dfa
2015-03-29 18:26:36 -03:00
doc Fix common misspellings 2015-03-19 17:25:57 +01:00
etc Refactor code structure for notification 2014-12-09 16:19:42 +13:00
rally-jobs Rename rally-scenarios/ to rally-jobs/ 2014-11-24 17:47:05 +04:00
tests Remove sqlalchemy from our data plane 2015-03-29 18:26:36 -03:00
tools/doc Update oslo-config-generation code 2014-11-17 21:49:53 +01:00
zaqar Remove sqlalchemy from our data plane 2015-03-29 18:26:36 -03:00
.coveragerc Rename Marconi to Zaqar 2014-08-04 10:36:50 +02:00
.gitignore Remove config sample and generate it using code 2014-11-14 17:43:18 +01:00
.gitreview Fix .gitreview due to the repo rename/move 2014-08-16 21:30:22 +04:00
.testr.conf Workaround pypy bug on trusty when running tests 2014-09-11 13:40:30 -07:00
AUTHORS.rst refactor: Rename AUTHORS so that it doesn't keep getting overwritten 2013-03-19 16:33:43 -04:00
HACKING.rst Refactor code structure for notification 2014-12-09 16:19:42 +13:00
LICENSE Include full license text 2014-03-21 10:16:28 +01:00
MANIFEST.in Move Redis driver's claim transaction to Lua 2014-09-29 11:49:40 -05:00
README.rst Replace API v1 with v1.1 in README 2015-02-13 13:36:38 -05:00
babel.cfg Prepare marconi for localization 2014-06-04 22:31:55 +02:00
bench-requirements.txt Add observer role to benchmark tool 2014-08-29 10:30:51 -05:00
doc-test.conf Rename Marconi to Zaqar 2014-08-04 10:36:50 +02:00
dox.yml Add a dox.yml config file 2014-09-08 13:55:13 +02:00
openstack-common.conf Use the oslo.serialization instead of openstack/common/jsonutils 2014-12-24 09:24:47 +08:00
requirements-py3.txt Updated from global requirements 2015-03-21 00:19:13 +00:00
requirements.txt Updated from global requirements 2015-03-21 00:19:13 +00:00
setup.cfg Remove sqlalchemy from our data plane 2015-03-29 18:26:36 -03:00
setup.py Updated from global requirements 2014-04-30 02:40:38 +00:00
test-requirements-py3.txt Updated from global requirements 2015-03-06 16:54:45 +00:00
test-requirements.txt Updated from global requirements 2015-03-21 00:19:13 +00:00
tox.ini Update oslo-config-generation code 2014-11-17 21:49:53 +01:00

README.rst

Zaqar

Message queuing service for OpenStack. To find more information read our wiki.

Running a local Zaqar server with MongoDB

Note: These instructions are for running a local instance of Zaqar and not all of these steps are required. It is assumed you have MongoDB and tox (see "Running tests" section below) installed and running.

  1. From your home folder create the ~/.zaqar folder and clone the repo:

    $ cd
    $ mkdir .zaqar
    $ git clone https://github.com/openstack/zaqar.git
  2. Generate and copy the Zaqar config files to the directory ~/.zaqar:

    $ cd zaqar
    $ tox -e genconfig
    $ cp etc/zaqar.conf.sample ~/.zaqar/zaqar.conf
    $ cp etc/logging.conf.sample ~/.zaqar/logging.conf
  3. Find [drivers] section in ~/.zaqar/zaqar.conf and specify to use mongodb storage:

    storage = mongodb

    Then find the [drivers:storage:mongodb] section and modify the URI to point to your local mongod instance:

    uri = mongodb://$MONGODB_HOST:$MONGODB_PORT

    By default, you will have:

    uri = mongodb://127.0.0.1:27017
  4. For logging, find the [DEFAULT] section in ~/.zaqar/zaqar.conf and modify as desired:

    log_file = server.log
  5. Change directories back to your local copy of the repo:

    $ cd ~/zaqar
  6. Run the following so you can see the results of any changes you make to the code without having to reinstall the package each time:

    $ pip install -e .
  7. Start the Zaqar server with logging level set to INFO so you can see the port on which the server is listening:

    $ zaqar-server -v
  8. Test out that Zaqar is working by creating a queue:

    $ curl -i -X PUT http://127.0.0.1:8888/v1.1/queues/samplequeue -H
    "Content-type: application/json"

You should get an HTTP 201 along with some headers that will look similar to this:

HTTP/1.0 201 Created
Date: Fri, 25 Oct 2013 15:34:37 GMT
Server: WSGIServer/0.1 Python/2.7.3
Content-Length: 0
Location: /v1.1/queues/samplequeue

Running tests

First install additional requirements:

$ pip install tox

And then run tests:

$ tox -e py27

You can read more about running functional tests in separate TESTS_README.

Running the benchmarking tool

First install and run zaqar-server (see above).

Then install additional requirements:

$ pip install -r bench-requirements.txt

Copy the configuration file to ~/.zaqar:

$ cp etc/zaqar-benchmark.conf.sample ~/.zaqar/zaqar-benchmark.conf

In the configuration file specify where zaqar-server can be found:

server_url = http://localhost:8888

The benchmarking tool needs a set of messages to work with. Specify the path to the file with messages in the configuration file. Alternatively, put it in the directory with the configuration file and name it zaqar-benchmark- messages.json. As a starting point, you can use the sample file from the etc directory:

$ cp etc/zaqar-benchmark-messages.json ~/.zaqar/

If the file is not found or no file is specified, a single hard-coded message is used for all requests.

Run the benchmarking tool using the following command:

$ zaqar-bench

By default, the command will run a performance test for 5 seconds, using one producer process with 10 greenlet workers, and one observer process with 5 workers. The consumer role is disabled by default.

You can override these defaults in the config file or on the command line using a variety of options. For example, the following command runs a performance test for 30 seconds using 4 producer processes with 20 workers each, plus 4 consumer processes with 20 workers each. Note that the observer role is also disabled in this example by setting its number of workers to zero:

$ zaqar-bench -pp 4 -pw 10 -cp 4 -cw 20 -ow 0 -t 30

By default, the results are in JSON. For more human-readable output add the --verbose flag. Verbose output looks similar to the following:

$ zaqar-bench --verbose

Starting producer (pp=1 , pw=10)...

Starting observer (op=1 , ow=5)...

Producer
========
duration_sec: 5.1
ms_per_req: 2.9
reqs_per_sec: 344.5
successful_reqs: 1742.0
total_reqs: 1742.0

Observer
========
duration_sec: 5.0
ms_per_req: 2.9
reqs_per_sec: 339.3
successful_reqs: 1706.0
total_reqs: 1706.0