Custom version of plone.recipe.zeoserver
Project description
Overview
This recipe creates and configures a ZEO server in parts. It also installs a control script in the bin/ directory. The name of the control script is the name of the part in buildout.
You can use it with a part like this:
[zeo] recipe = plone.recipe.zeoserver zeo-address = 8100
This will create a control script bin/zeo.
You can either start the database in foreground mode via bin/zeo fg or use the built-in zdaemon process control and use the start/stop/restart/status commands. The foreground mode is suitable for running the process under general process control software like supervisord.
Note: Windows support for this recipe is currently limited.
Options
The following options all affect the generated zeo.conf. If you want to have full control over the configuration file, see the zeo-conf option in the advanced options.
Process
- zeo-address
Give a port for the ZEO server (either specify the port number only (with ‘127.0.0.1’ as default) or you use the format host:port). Defaults to 8100.
- effective-user
The name of the effective user for the ZEO process. Defaults to not setting an effective user. This causes the process to run under the user account the process has been started with.
- socket-name
The filename where ZEO will write its socket file. Defaults to var/zeo.zdsock.
Storage
- storage-number
The number used to identify a storage. Defaults to 1.
- file-storage
The filename where the ZODB data file will be stored. Defaults to var/filestorage/Data.fs.
- blob-storage
The folder where the ZODB blob data files will be stored. Defaults to var/blobstorage.
Logging
- zeo-log
The filename of the ZEO log file. Defaults to var/log/${partname}.log.
- zeo-log-format
Format of logfile entries. Defaults to %(asctime)s %(message)s.
- zeo-log-custom
A custom section for the eventlog, to be able to use another event logger than logfile. zeo-log is still used to set the logfile value in the runner section.
Authentication
- authentication-database
The filename for a authentication database. Only accounts listed in this database will be allowed to access the ZEO server.
The format of the database file is:
realm <realm> <username>:<hash>
Where the hash is generated via:
import sha string = "%s:%s:%s" % (username, realm, password) sha.new(string).hexdigest()
- authentication-realm
The authentication realm. Defaults to ZEO.
Packing
- pack-days
How many days of history should the zeopack script retain. Defaults to one day.
- pack-gc
Can be set to false to disable garbage collection as part of the pack. Defaults to true.
- pack-keep-old
Can be set to false to disable the creation of *.fs.old files before the pack is run. Defaults to true.
- pack-user
If the ZEO server uses authentication, this is the username used by the zeopack script to connect to the ZEO server.
- pack-password
If the ZEO server uses authentication, this is the password used by the zeopack script to connect to the ZEO server.
Monitoring
- monitor-address
The address at which the monitor server should listen. The monitor server provides server statistics in a simple text format.
Performance
- invalidation-queue-size
The invalidation-queue-size used for the ZEO server. Defaults to 100.
Customization
- zeo-conf-additional
Give additional lines to zeo.conf. Make sure you indent any lines after the one with the parameter. This allows you to use generated zeo.conf file but add some minor additional lines to it.
- eggs
Set if you need to include other packages as eggs e.g. for making application code available on the ZEO server side for performing conflict resolution (through the _p_resolveConflict() handler).
- extra-paths
Specify additional directories which should be available to the control scripts. Use this only for non-eggified Python packages.
- zeo-conf
A relative or absolute path to a zeo.conf file. This lets you provide a completely custom configuration file and ignore most of the options in this recipe.
- repozo
The path to the repozo.py backup script. A wrapper for this will be generated in bin/repozo, which sets up the appropriate environment for running this. Defaults to using the repozo script from the ZODB3 egg. Set this to an empty value if you do not want this script to be generated.
- zeopack
The path to the zeopack.py backup script. A wrapper for this will be generated in bin/zeopack, which sets up the appropriate environment to run this. Defaults to using the zeopack script from the ZODB3 egg. Set this option to an empty value if you do not want this script to be generated.
- relative-paths
Set this to true to make the generated scripts use relative paths. You can also enable this in the [buildout] section.
Usage
- zeopack
A zeopack script will be generated for you in the buildout bin directory. If you’d like to use this script to pack a different mount point, you’ll need to specify -S mount_name. You can also specify a -B option to not use the default blob directory.
Reporting bugs or asking questions
We have a shared bugtracker and help desk on Launchpad: https://bugs.launchpad.net/collective.buildout/
Changelog
2.0.4 (2011-11-15)
Add files missing from sdist
2.0.3 (2011-11-14)
Fix a versioning mishap
2.0.2 (2011-11-14)
Add a list of imports for monkey patch injection:
[zeoserver] eggs = isotoma.zope.groupreadableblobstorage import = isotoma.zope.groupreadableblobstorage
2.0.0 - 2011-03-08
Friendly fork to patch sys.path behaviour in our deployments.
1.2.0 - 2010-10-18
Only require a nt_svcutils distribution on Windows. [hannosch]
1.1.1 - 2010-07-20
Fixed -B option being required for along with the -S option. [vangheem]
Added documentation for using the zeopack script with mount points. [vangheem]
1.1 - 2010-07-18
No changes.
1.1b1 - 2010-07-02
Implemented Windows support and support for running ZEO as a Windows service. We depend on the new nt_svcutils distribution to provide this support. [baijum, hannosch]
The FileStorage component of ZODB 3.9 now supports blobs natively, so no need to use BlobStorage proxy for it anymore. [baijum, hannosch]
Added extra-paths option to add additional modules paths. [baijum]
Fixed ZEO packing of mounted storage. [vangheem]
Added -B option to the zeopack script to specify the location of the blob storage. [vangheem]
1.1a2 - 2010-05-10
Added support for the pack-keep-old option introduced in ZODB 3.9. [hannosch]
1.1a1 - 2010-04-27
Added support for the pack-gc option introduced in ZODB 3.9. [hannosch]
Always create a blob-storage by default. [hannosch]
Require at least ZODB 3.8 and simplify the zeopack script. [hannosch]
Various documentation updates. [hannosch]
Use the new zope.mkzeoinstance package, which makes the recipe compatible with ZODB 3.9.5+. [hannosch]
Removed unmaintained win32 specific tests and old zope2 test mockups. [hannosch]
Removed testing dependency on zope.testing and refactored testing setup. [hannosch]
1.0 - 2010-04-05
Depend on and always include ZopeUndo. While it’s only needed for Zope 2, the distribution is so tiny, it doesn’t hurt for non-Zope 2 ZEO servers. [hannosch]
1.0b1 - 2010-03-19
Fixed issue with egg paths for the zeopack script. [davisagli]
Added support for setting ZEO log level. [baijum]
1.0a2 - 2009-12-03
Set up logging configuration that is needed by ZODB.blob. [davisagli]
Set shared_blob_dir to True when initializing the ClientStorage used by the pack script, since it will be using the same blob directory as the ZEO server. [davisagli]
1.0a1 - 2009-12-03
Updated and cleaned up after renaming. [hannosch]
Added compatibility with eggified Zopes (Zope >= 2.12). [davisagli]
Initial implementation based on plone.recipe.zope2zeoserver. [plone]
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.