habconf

NAME

habconf - Habitat configuration  

DESCRIPTION

Every major habitat utility and program gathers its configuration from external sources in the same way. The formats are shown below, but this section details how the data is gathered.

Firstly, each program can have individual directives specified on the command line with multiple '-C' flags.

A whole file of replacement directives can be used by using the '-c' flag. As this can be a route address, the data can originate from a web server. This feature allows one to keep the application and user defaults for mainstream use, and also allows special instances to be used simultaneously.

The user file ~/.habrc is read next (although the location can be overridden), generally used for specific customisations. For example, the data files held in user interface history.

Then the application config is read from $HAB/etc/habitat.conf or /etc/habitat.conf (either of which again can be overridden). This contains the default supplied by the developers and may be ammended by the local administrators.

Further configuration sources are possible, to allow for administrators to configure on a regional or global basis. Thus, the application configuration file does not have to be altered or making regional varients built. These sources may be web servers or other directory locations.  

PARAMETERS

To customise at a user level, the file ~/.habrc should contain directives in the following format.

[-]directive [[=] value]

Directive may appear on its own, or have a value associated with it. Values may be arrays, by separating their elements with whitespace. Each directive must appear on a new line. Comments may appear on any line, including ones with directives. They start with the '#' character and end with the newline character. The file must start with the magic string 'habitat 1', which confirms the contents and intent of the file to the configuration system.

Common directives include:-

elog.above <sev> <route>
elog.below <sev> <route>
elog.set <sev> <route>
elog.all <route>
send errors and other messages above (or below) level <sev> to the route <route>. The .set form configures just that severity level and the .all form sends messages from all severities to <route>. <sev> can be 'fatal', 'error', 'warning', 'info', 'diag' and 'debug'.

elog.format <sev> <format>
elog.allformat <format>
The specified severity level should print using <format>. The .allformat form sets <format> to all message sevelities. See the full manuals for more details of the format.

hab.cfetc <route>
hab.cfuser <route>
Locations of the user configuration file (~/.habrc) in route format and the application wide file ($HAB/etc/habitat.conf). Because they are read early in the process of starting an application, these have to be specifed on the command line using -c or -C flags.

nmalloc
Turn on the memory checking, which will identify memory leaks. Normally this is off on stable releases, but developer releases may have it activated within the code.

replicate.out <links>
replicate.in <links>
Specifies a list of replication links, used by the replication job. The purpose is to associate a local storage ring with a remote one on a repository. See the replication method's manual for information on its format.  

FILES

~/.habrc
$HAB/etc/habitat.conf or /etc/habitat.conf  

SEE ALSO

clockwork(8), killclock(8), statclock(8), habedit(8), habrep(8), myhabitat(1), habget(1), habput(1), habrs(1), habprobe(1), habmeth(1)

---

 

Index

NAME

DESCRIPTION

PARAMETERS

FILES

SEE ALSO