To perform an export, open Windows Service Manager and open the property sheet for Searchbox Engine service. Stop the service if it is running by pressing the Stop button. Enter /export dir in the Start parameters text field and press the Start button. The searchbox service will stop automatically after the export is finished.

dir is a directory that will contain one file for each source, archive, collection, watch, plugin etc. configured on the system. The directory will be created by searchboxd and must not exist before the export.

To perform an import, open Windows Service Manager and open the property sheet for Searchbox Engine service. Stop the service if it is running by pressing the Stop button. Enter /import dir in the Start parameters text field and press the Start button. The searchbox service will stop automatically after the import is finished.

dir is a directory created by a previous export between double quotes if the path contains spaces.

To perform an export first stop searchboxd, then run

# searchboxd --export=dir

Where dir is a directory that will contain one file for each source, archive, collection, watch, plugin etc. configured on the system. The directory will be created by searchboxd and must not exist before the export.

To perform an import first stop searchboxd, then run

# searchboxd --import=dir

Where dir is a directory created by a previous --export.

To perform a reindex, open Windows Service Manager and open the property sheet for Searchbox Server service. Stop the service if it is running by pressing the Stop button. Enter /reindex in the Start parameters text field and press the Start button. The searchbox service will stop automatically after the reindex is finished.

To rebuild them first stop searchboxd, then run:

# searchboxd --reindex

To perform a reset, open Windows Service Manager and open the property sheet for Searchbox Server service. Stop the service if it is running by pressing the Stop button. Enter /reset in the Start parameters text field and press the Start button. The searchbox service will stop automatically after the reset is finished.

First stop searchboxd, then run:

# searchboxd --reset

To perform an optimization, open Windows Service Manager and open the property sheet for Searchbox Server service. Stop the service if it is running by pressing the Stop button. Enter /optimize in the Start parameters text field and press the Start button. The searchbox service will stop automatically after the optimization is finished.

Run the following command:

# searchboxd --optimize

The User and Group configuration parameters determine the process identity, these are Unix-only options. searchbox needs to be run as root on startup, but for security purposes it will switch to the specified user and group after the initial startup phase.

On Windows you configure the process identity through the Service Control Manager.

DEPLOYDIR and DATADIR point to the program installation location and the data file location. You don't usually need to change these, as they are set by the installation application.

searchbox records that it's running in a file called the pidfile whose location is specified in the PIDFILE parameter. It is set to /var/run/searchbox.pid and you should not change it unless you are fiddling with system startup/shutdown.

WORKERS is number of processes that concurrently perform the gathering/parsing/rendering chain. The default value is 5 but can be modified according with the size of your pool of sources and the available amount of RAM and CPU.

These directives let you control how to access searchbox from external applications:

If you're running searchbox from behind a proxy, the following configuration directives let you set the necessary parameters:

searchbox can send notifications by email or instant messages using external facilities:

LOGCFG determines how much information will be logged to the files inside the log directory in the searchbox data directory. It is a space-separated list of facility:level pairs. Facility names are:

ossible level values, in decreasing order of verbosity, are BORING, DEBUG, INFO, WARNING, ERROR, CRITICAL. The default for all facilities is info, and it's generally not necessary to change it, except possibly for debugging purposes.

focuseek searchbox is a well-behaved web spider and implements the standard for robots exclusion (http://www.robotstxt.org/wc/norobots.html), usually referred to as the robots.txt standard. The robots.txt standard lets webmasters disable spidering of portions of their site by placing a suitably formatted robots.txt file in the root directory of their site. The robots.txt standard requires that each spider identifies itself using a user agent string. This string is used also to disable some robots spidering the site while still allowing others to access it. See the standard for the details.

You can specify the user agent used by searchbox with the UserAgent parameter in the configuration file. It defaults to focuseekbot. Such parameters can be overridden by the specific configuration for each Archive (see the Control Panel documentation)

The UserAgent parameter is also used to identify the spider in the HTTP User-Agent request header. focuseek searchbox uses the following HTTP user agent:

Mozilla/4.0 (compatible; agent)

Where agent is the value of the UserAgent parameter in the configuration fine. For further details on the HTTP user agent see for example http://www.mozilla.org/build/user-agent-strings.html

Index optimization will occur daily. Since it is a very disk anc cpu consuming operation, you can choose the exact hour at which it will occur. The OptimizeAt parameter holds the GMT hour at which index optimization will occur. You can disable index optimization by setting the OptimizeAt parameter to -1.

When the optimization is enabled some other parameters can be tuned:

For performace reasons, documents are not added to index one at once, but in batches. However, an index flush will be forced from time to time. The FlushLatency parameter holds the number of seconds between index flushes.

The SyncLevel parameter settles how often the internal database flushes its data on disk.

searchbox has its own internal http server used to implement the Web Service interface. The following parameters determines its behaviour:

Inthe case you want to disable access to searchbox through the internal Enterprise Search Portal set the PORTAL_ENABLED parameter value as NO.

Worker is the searchbox component responsible of all processing activities on every single document. Sometime it happens that a worker process hangs due to the impossibility to process a document. The WORKER_TIMEOUT parameter sets the number of seconds that searchbox waits before killing a hanged process.

Gathering a large document set can lead to disk space exhaustion. Searchbox will check for free disk space and halt the gathering process when disk free space falls below the value set in the MINADDFREEDISK parameter (in MB).

You can allow the gathering process to run only during a certain period of the day. For example, you could allow gathering only during the night. To do this, you can change the CRAWLSTART and CRAWLSTOP parameters, which hold the start and stop hours (GMT). You can completely stop the gathering by setting one of these parameters to -1.

Note that the crawls will be put in a sleep state outside of this interval of time, and they will resume from the point where they left. This is different from stopping and restarting a crawl: in this case gathering will start again from seeds.

searchbox can process suitably formattted metadata placed in HTML files. To enable this feature you must set the USE_SEARCHBOX_METAS parameter to 1. The default value is 0. See the section called “Collecting metadata embedded in HTML documents” for further explanations.