Akonadi（数据存储）
介绍
Akonadi 框架为应用程序提供中心数据库来统一保存、索引和获取用户的个人信息。这包括邮件、联系人、日历、事件、日志、闹钟和笔记等。在 SC 4.4 中， KAddressBook 成为首个使用 Akonadi 框架的程序。在 SC 4.7 中， KMail, KOrganizer, KJots等也开始更新使用 Akonadi 。此外，一些 等离子 部件也使用 Akonadi 保存和获取日历事件、笔记等。
Currently, all KDE PIM applications with the exception of Akregator are using Akonadi to access user's PIM data.
控制 Akonadi 服务器
Akonadi will start automatically in the background when any application using it is started.
To manually start, stop or restart Akonadi, you can use the akonadictl command from the commandline.
Using this method, you can get additional useful information on the console.
启动 Akonadi 服务器：
akonadictl start
停止Akonadi 服务器：
akonadictl stop
重新启动正在运行的 Akonadi 服务器，
akonadictl restart
查询 Akonadi 服务器的状态：
akonadictl status
Disabling the Akonadi subsystem
The Akonadi server is automatically started by any Akonadi-enabled application. If you don't want Akonadi to be started after login, you have to ensure that no Akonadi-enabled application is launched at login or thereafter. Remember to check Plasma applets as well — the Digital Clock widget in the default panel, for instance uses Akonadi to (optionally) display calendar events and this is enabled in its settings by default (see the "Display Events" option) . You must remove any widgets that may start it from your start-up, if you wish Akonadi to start only when you start KMail or other applications.
别忘了！
To ensure that Akonadi is not started, check that no applications require it at login. In particular, open the Plasma clock applet preferences, go to and uncheck to prevent Plasma from requesting information from Akonadi and thus allowing it to start.
Some Definitions
- Real data
- By real data we mean the data, like the contacts or events. These data are stored either on a groupware server or in local files. Where exactly depends on the resource you are using. E.g. the Personal Contacts resource stores its data under $XDG_DATA_HOME/contacts.
- Cached data
- The cached data are copies of the real data that are kept in the database for faster access and offline caching. The database also keeps the meta data which are management data needed by Akonadi to work correctly.
- Configuration data
- The configuration data are the data that configure the Akonadi server and the individual resources. The general configuration data for the server can be found under $XDG_CONFIG_HOME/akonadi. The configuration data for each indvidual resources are stored under $XDG_CONFIG_HOME/akonadi_xyz_resourcerc# (xyz is name of resource and # its instance number).
- The Akonadi server configuration is a couple of files in $XDG_CONFIG_HOME/akonadi. It contains which data sources and helper programs are active and will be started and watched (so they can be restarted on crashes) by one of Akonadi's server processes (akonadi_control).
- Each data source handler (called resources) or helper program (called agents) can have its own configuration although some agents or resources don't require configuration. The general rule is that for every entry in $XDG_CONFIG_HOME/akonadi/agentsrc there is a corresponding configuration file in $XDG_CONFIG_HOME. For example, if the [Instances] section in agentrc contains an entry for akonadi_ical_resource_2, there is also a config file called akonadi_ical_resource_2rc in the $XDG_CONFIG_HOME directory.
- Depending on the type of data, such config files for resources will have filenames or directory names of where the data is stored. Common locations are KDE's legacy default files, e.g. $HOME/.kde/share/apps/korganizer/std.ics. New default locations are files and directories in $XDG_DATA_HOME, e.g. $XDG_DATA_HOME/contacts.
Backup
So now we need to decide what to back up. If you want to backup the "real data", then it depends on the resources you have configured... if you use a groupware server, then the backup should be done there. For contacts, the files under $XDG_DATA_HOME/contacts will normally be what you need.
To back up the entire Akonadi configuration, including which resources are active and their configuration, you can use the pimdataexporter' tool. This, however doesn't back up the Akonadi database containing the cached data and, unfortunately, after restoring the configuration (using the pimdataexporter again), Akonadi will have to re-fetch all data again into its cache. This can cause configuration that points to actual mail folders or calendars to get broken and accidentally point to another folder.
After restoring configuration and syncing all data, it's vital to manually check all folder configuration, especially in KMail identities and make sure the folders are configured properly.
常见问题
我现在的数据在哪？
Your data are safely stored outside of Akonadi control on your disk (e.g. local maildir folder or iCal calendar), or on a remote server (in case of e.g. email over IMAP or events from a CalDAV calendar). Akonadi will optionally store a copy of this data in its database to allow applications to quickly retrieve and display them. Any modifications done to data in the Akonadi database will be synced to the actual storage. The main advantage of using the database as a cache is that remote PIM data are available even when you are offline, and you can still interact with them (e.g. mark emails as read or move them, create new events, reschedule existing meetings etc.) and all the changes will get synced automatically once you connect to the internet again.
Thus, deleting the Akonadi database will not cause any data to be lost (as long as all pending changes are synced).
How to upgrade my PostgreSQL database?
After updating your PostgreSQL server to a new major version, sometimes you will have to convert your Akonadi database for use with this new version. Instructions can be found on this page.
迁移问题
How do I switch from MySQL/PostgreSQL to SQLite?
Since Akonadi 6 akonadi-db-migrator is included in akonadi that allows you to switch the database backend. Where for previous versions this involved deleting and recreating the database a migration is now properly supported by akonadi.
The tool needs to be called with the parameter --newengine to chose to which backend it should to: {{{
--newengine <ENGINE> The new DB engine to use. Possible values are
"sqlite", "mysql" and "postgres"
}}}
For example: {{{ akonadi-db-migrator --newengine sqlite }}}