bcfg2.conf includes configuration parameters for the Bcfg2 server and client.
The file is INI-style and consists of sections and options. A section begins with the name of the sections in square brackets and continues until the next section begins.
Options are specified in the form “name=value”.
The file is line-based each newline-terminated line represents either a comment, a section name or an option.
Any line beginning with a hash (#) is ignored, as are lines containing only whitespace.
These options are only necessary on the Bcfg2 server. They are specified in the [server] section of the configuration file.
The file monitor used to watch for changes in the repository. The default is the best available monitor. The following values are valid:
inotify gamin pseudo
A comma-separated list of globs that should be ignored by the file monitor. Default values are:
*~ *# #* *.swp *.swpx *.swx SCCS .svn 4913 .gitignore
A comma-delimited list of enabled server plugins. Currently available plugins are:
ACL Bundler Bzr Cfg Cvs Darcs Decisions Defaults Deps FileProbes Fossil Git GroupLogic GroupPatterns Guppy Hg Ldap Metadata NagiosGen Ohai Packages Pkgmgr POSIXCompat Probes Properties PuppetENC Reporting Rules SEModules ServiceCompat SSHbase Svn TemplateHelper Trigger
Descriptions of each plugin can be found in their respective sections below.
Specifies which server core backend to use. Current available options are:
cherrypy builtin best
The default is best, which is currently an alias for builtin. More details on the backends can be found in the official documentation.
This section has a listing of all the plugins currently provided with Bcfg2.
The ACL plugin controls which hosts can make which XML-RPC calls.
The Bundler plugin is used to describe groups of inter-dependent configuration entries, such as the combination of packages, configuration files, and service activations that comprise typical Unix daemons. Bundles are used to add groups of configuration entries to the inventory of client configurations, as opposed to describing particular versions of those entries.
The Bzr plugin allows you to track changes to your Bcfg2 repository using a GNU Bazaar version control backend. Currently, it enables you to get revision information out of your repository for reporting purposes.
The Cfg plugin provides a repository to describe configuration file contents for clients. In its simplest form, the Cfg repository is just a directory tree modeled off of the directory tree on your client machines.
The Cvs plugin allows you to track changes to your Bcfg2 repository using a Concurrent version control backend. Currently, it enables you to get revision information out of your repository for reporting purposes.
The Darcs plugin allows you to track changes to your Bcfg2 repository using a Darcs version control backend. Currently, it enables you to get revision information out of your repository for reporting purposes.
The Decisions plugin has support for a centralized set of per-entry installation decisions. This approach is needed when particular changes are deemed “high risk”; this gives the ability to centrally specify these changes, but only install them on clients when administrator supervision is available.
The Defaults plugin can be used to populate default attributes for entries. Defaults is not a Generator plugin, so it does not actually bind an entry; Defaults are applied after an entry has been bound, and only populate attributes that are not yet set.
The Deps plugin allows you to make a series of assertions like “Package X requires Package Y (and optionally also Package Z etc.)”
The FileProbes plugin allows you to probe a client for a file, which is then added to the Cfg specification. If the file changes on the client, FileProbes can either update it in the specification or allow Cfg to replace it.
The Fossil plugin allows you to track changes to your Bcfg2 repository using a Fossil SCM version control backend. Currently, it enables you to get revision information out of your repository for reporting purposes.
The Git plugin allows you to track changes to your Bcfg2 repository using a Git version control backend. Currently, it enables you to get revision information out of your repository for reporting purposes.
The GroupLogic plugin lets you flexibly assign group membership with a Genshi template.
The GroupPatterns plugin is a connector that can assign clients group membership based on patterns in client hostnames.
The Guppy plugin is used to trace memory leaks within the bcfg2-server process using Guppy.
The Hg plugin allows you to track changes to your Bcfg2 repository using a Mercurial version control backend. Currently, it enables you to get revision information out of your repository for reporting purposes.
The Ldap plugin makes it possible to fetch data from a LDAP directory, process it and attach it to your metadata.
The Metadata plugin is the primary method of specifying Bcfg2 server metadata.
The NagiosGen plugin dynamically generates Nagios configuration files based on Bcfg2 data.
The Ohai plugin is used to detect information about the client operating system. The data is reported back to the server using JSON.
The Packages plugin is an alternative to Pkgmgr for specifying package entries for clients. Where Pkgmgr explicitly specifies package entry information, Packages delegates control of package version information to the underlying package manager, installing the latest version available from through those channels.
The Pkgmgr plugin resolves the Abstract Configuration Entity “Package” to a package specification that the client can use to detect, verify and install the specified package.
The POSIXCompat plugin provides a compatibility layer for 1.3 POSIX Entries so that they are compatible with older clients.
The Probes plugin gives you the ability to gather information from a client machine before you generate its configuration. This information can be used with the various templating systems to generate configuration based on the results.
The Properties plugin is a connector plugin that adds information from properties files into client metadata instances.
The PuppetENC plugin is a connector plugin that adds support for Puppet External Node Classifiers.
The Reporting plugin enables the collection of data for use with Bcfg2’s dynamic reporting system.
The Rules plugin provides literal configuration entries that resolve the abstract configuration entries normally found in Bundler. The literal entries in Rules are suitable for consumption by the appropriate client drivers.
The SEModules plugin provides a way to distribute SELinux modules via Bcfg2.
The ServiceCompat plugin converts service entries for older clients.
The SSHbase generator plugin manages ssh host keys (both v1 and v2) for hosts. It also manages the ssh_known_hosts file. It can integrate host keys from other management domains and similarly export its keys.
The Svn plugin allows you to track changes to your Bcfg2 repository using a Subversion backend. Currently, it enables you to get revision information out of your repository for reporting purposes.
The Trigger plugin provides a method for calling external scripts when clients are configured.
These options are specified in the [caching] section.
The following four caching modes are available for client metadata:
- off: No caching of client metadata objects is performed. This is the default.
- initial: Only initial metadata objects are cached. Initial metadata objects are created only from the data in the Metadata plugin, before additional groups from other plugins are merged in.
- cautious: Final metadata objects are cached, but each client’s cache is cleared at the start of each client run, immediately after probe data is received. Cache is also cleared as in aggressive mode. on is a synonym for cautious.
- aggressive: Final metadata objects are cached. Each plugin is responsible for clearing cache when appropriate.
These options only affect client functionality. They can be specified in the [client] section.
- Specify the server decision list mode (whitelist or blacklist). (This settiing will be ignored if the client is called with the -f option).
- Specify tool driver set to use. This option can be used to explicitly specify the client tool drivers you want to use when the client is run.
- Run the client in paranoid mode.
- Assert the given profile for the host.
Specified in the [communication] section. These options define settings used for client-server communication.
- The path to a file containing the CA certificate. This file is required on the server, and optional on clients. However, if the cacert is not present on clients, the server cannot be verified.
- The path to a file containing a PEM formatted certificate which signs the key with the ca certificate. This setting is required on the server in all cases, and required on clients if using client certificates.
- Specifies the path to a file containing the SSL Key. This is required on the server in all cases, and required on clients if using client certificates.
- Required on both the server and clients. On the server, sets the password clients need to use to communicate. On a client, sets the password to use to connect to the server.
- Communication protocol to use. Defaults to xmlrpc/tlsv1.
- A client-only option. Number of times to retry network communication. Default is 3 retries.
- A client-only option. Number of seconds to wait in between retrying network communication. Default is 1 second.
- A client-only option. A colon-separated list of Common Names the client will accept in the SSL certificate presented by the server.
- A client-only option. The network communication timeout.
- A client-only option. The UUID of the client.
Specified in the [components] section.
URL of the server. On the server this specifies which interface and port the server listens on. On the client, this specifies where the client will attempt to contact the server.
e.g. bcfg2 = https://10.3.1.6:6789
- Text encoding of configuration files. Defaults to UTF-8.
- The path to the client lock file, which is used to ensure that only one Bcfg2 client runs at a time on a single client.
Specified in the [logging] section. These options control the server logging functionality.
- Whether or not to enable debug-level log output. Default is false.
- Server log file path.
- Whether or not to send logging data to syslog. Default is true.
- Whether or not to enable verbose log output. Default is false.
Specified in the [mdata] section. These options affect the default metadata settings for Paths with type=’file’.
- Global owner for Paths (defaults to root)
- Global group for Paths (defaults to root)
- Global permissions for Paths (defaults to 644)
- Global SELinux context for Path entries (defaults to __default__, which restores the expected context)
- Global paranoid settings for Paths (defaults to false)
- Global sensitive settings for Paths (defaults to false)
- Global important settings for Paths. Defaults to false.
The following options are specified in the [packages] section.
- Comma separated list of backends for the dependency resolution. Default is “Yum,Apt,Pac,Pkgng”.
- Enable dependency resolution. Default is 1 (true).
- Enable metadata processing. Default is 1 (true). If metadata is disabled, it’s implied that resolver is also disabled.
- The path at which to generate Yum configs. No default.
- The path at which to generate APT configs. No default.
- The path on the client where RPM GPG keys will be copied before they are imported on the client. Default is /etc/pki/rpm-gpg.
- Set the version attribute used when binding Packages. Default is auto.
The following options are specified in the [packages:yum] section.
- By default, Bcfg2 uses an internal implementation of Yum’s dependency resolution and other routines so that the Bcfg2 server can be run on a host that does not support Yum itself. If you run the Bcfg2 server on a machine that does have Yum libraries, however, you can enable use of those native libraries in Bcfg2 by setting this to 1.
- Path to bcfg2-yum-helper. By default, Bcfg2 looks first in $PATH and then in /usr/sbin/bcfg2-yum-helper for the helper.
The following options are specified in the [packages:pulp] section.
- The username of a Pulp user that will be used to register new clients and bind them to repositories.
- The password of a Pulp user that will be used to register new clients and bind them to repositories.
All other options in the [packages:yum] section will be passed along verbatim to the Yum configuration if you are using the native Yum library support.
These options allow for finer-grained control of the paranoid mode on the Bcfg2 client. They are specified in the [paranoid] section of the configuration file.
- Custom path for backups created in paranoid mode. The default is in /var/cache/bcfg2.
- Specify a maximum number of copies for the server to keep when running in paranoid mode. Only the most recent versions of these copies will be kept.
These options are necessary to configure the SSL CA feature of the Cfg plugin and can be found in the [sslca_default] section of the configuration file.
- Specifies the location of the openssl configuration file for your CA.
- Specifies the passphrase for the CA’s private key (if necessary). If no passphrase exists, it is assumed that the private key is stored unencrypted.
- Specifies the location of your ssl chaining certificate. This is used when pre-existing certifcate hostfiles are found, so that they can be validated and only regenerated if they no longer meet the specification. If you’re using a self signing CA this would be the CA cert that you generated.
Server-only, specified in the [database] section. These options control the database connection of the server.
The database engine used by server plugins. One of the following:postgresql mysql sqlite3 ado_mssql
- The name of the database to use for server data. If ‘database_engine’ is set to ‘sqlite3’ this is a file path to the sqlite file and defaults to $REPOSITORY_DIR/etc/bcfg2.sqlite.
- User for database connections. Not used for sqlite3.
- Password for database connections. Not used for sqlite3.
- Host for database connections. Not used for sqlite3.
- Port for database connections. Not used for sqlite3.
- Various options for the database connection. The value expected is the literal value of the django OPTIONS setting.
The database engine used by the Reporting plugin. One of the following:postgresql mysql sqlite3 ado_mssql If reporting_engine is not specified, the Reporting plugin uses the same database as the other server plugins.
- The name of the database to use for reporting data. If ‘database_engine’ is set to ‘sqlite3’ this is a file path to the sqlite file and defaults to $REPOSITORY_DIR/etc/reporting.sqlite.
- User for reporting database connections. Not used for sqlite3.
- Password for reporting database connections. Not used for sqlite3.
- Host for reporting database connections. Not used for sqlite3.
- Port for reporting database connections. Not used for sqlite3.
- Various options for the database connection. The value expected is the literal value of the django OPTIONS setting.
- Specifies the location of the reporting configuration (default is /etc/bcfg2-web.conf.
- Specifies a time zone other than that used on the system. (Note that this will cause the Bcfg2 server to log messages in this time zone as well).
- Turn on Django debugging.
- Maximum number of children for the reporting collector. Use 0 to disable the limit. (default is 0)
- Arbitrary options for the Django installation. The value expected is a literal python dictionary, that is merged with the already set django settings.