OAR Command line interface.
OAR Command line interface
- python >= 2.7
- python-mysqldb/python-psycopg2 (for oar database archive)
You can install, upgrade, uninstall oar-cli with these commands:
$ pip install [--user] oar-cli $ pip install [--user] --upgrade oar-cli $ pip uninstall oar-cli
Or from git (last development version):
$ pip install git+https://github.com/oar-team/python-oar-cli.git
Or if you already pulled the sources:
$ pip install path/to/sources
Or if you don’t have pip:
$ easy_install oar-cli
OAR Database archive
This script allow you to copy your OAR database to another database for archiving purposes.
In order to do this, the tool offers two commands that can be launched regularly to simplify both databases maintenance.
This command is used to copy data to archive on the archive database. It is only useful when handling a small amount of data.
If your database is too big, we advise you to first create manually your archive database with a dump. If your archive database does not exist yet, and if it will be located on the same machine than your current OAR database, the ‘sync’ command will clone the OAR database to the archive (extremely efficient for postgresql).
Usage: oar-database-archive sync [OPTIONS] Send all resources and finished jobs to archive database. Options: --chunk INTEGER Chunk size [default: 10000] --ignore-jobs TEXT Ignore job state [default: ^Terminated, ^Error] --current-db-url TEXT The url for your current OAR database. --archive-db-url TEXT The url for your archive OAR database. -h, --help Show this message and exit.
Jobs unfinished (i.e. current jobs) are not copied by default. The symbol ‘^’ means negation.
For more information about the URL format, please refer to the SQLAlchemy documentation: http://docs.sqlalchemy.org/en/latest/core/engines.html#database-urls
$ oar-database-archive sync --archive-db-url postgresql://oar:oar@server:5432/oar_archive
This command will dump your current OAR database (configured in oar.conf) in the given database. The second run will only copy the missing rows.
From postgresql to mysql:
$ oar-database-archive sync --current-db-url postgresql://user:password@server1:5432/oar \ --archive-db-url mysql://user:password@server2:3306/oar_archive
From mysql to sqlite:
$ oar-database-archive sync --current-db-url mysql://user:password@server1:3306/oar \ --archive-db-url sqlite:////tmp/archive.db
The purge command is used to delete jobs and resources (previously archived) from the current OAR database.
Usage: oar-database-archive purge [OPTIONS] Purge old resources and old jobs from your current database. Options: --ignore-jobs TEXT Ignore job state [default: ^Terminated, ^Error] --max-job-id INTEGER Purge only jobs lower than this id --ignore-resources TEXT Ignore resource state [default: ^Dead] --current-db-url TEXT The url for your current OAR database. -h, --help Show this message and exit.
These options are almost similar. By default, all finished jobs are purged. You can use --max-job-id to ignore all jobs with an ID greater than the ID used as parameter..
All resources in ‘Dead’ state are also purged. If we do not want to purge resources, the filter --ignore-resources all needs to be used.
$ oar-database-archive purge \ --ignore-resources ^Dead \ --ignore-jobs ^Terminated --ignore-jobs ^Error \ --max-job-id 100000 \ --current-db-url postgresql://scott:tiger@localhost/oar_prod_database
This commande is used to delete all resourced marked as ‘Dead’ and all finished jobs (and associated events) with a ID smaller than 100000.
OAR CLI changelog
Released on June 24th 2015
- Added --conf option to used custom oar configuration file
- Used postgresql COPY with csv format by default
- Fixed modify nullable operation during schema upgrade
Released on June 23rd 2015
- Minor bug fixes
Released on June 23rd 2015
- Fixed project description in Pypi
Released on June 23rd 2015
- Renamed the project to OAR CLI !
- Added oar-database-migrate script.
- Added --schema-only and --data-only features to oar-database-migrate script
- Supported postgresql bulk insert using COPY clause to improved performance.
- Handled database connection errors.
- Managed the schema upgrade with alembic
- Fixed max_job_to_sync query if we want to copy all jobs
Released on June 05th 2015
- Made deterministic order_by to sync queries in order to avoid IntegrityError during copy (Fixed #1)
- Handled IntegrityError during bulk INSERT (Fixed #1)
- Used Postgresql DELETE with the USING clause to improve performance (Fixed #2)
- Made the delete orphan queries faster with LEFT JOIN in Mysql (Fixed #2)
- Removed count query when performing a bulk delete query (Fixed #2)
- Configured debug logging to displayed SQL queries
- Added option to disable pagination during sync operation
- Displayed default values in the CLI
Released on May 05th 2015
- [database-archive] : Added ability to ignore all resources during purge
Released on May 04th 2015
- Fixed pypi package
First public preview release.
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
|Filename, size & hash SHA256 hash help||File type||Python version||Upload date|
|oar_cli-0.3.3-py2.py3-none-any.whl (20.0 kB) Copy SHA256 hash SHA256||Wheel||2.7||Jun 24, 2015|
|oar-cli-0.3.3.tar.gz (16.6 kB) Copy SHA256 hash SHA256||Source||None||Jun 24, 2015|