Main index | Section 3 | Options |
#include <krb5.h>
The krb5_openlog() function creates a logging facility, that is used to log messages. A facility consists of one or more destinations (which can be files or syslog or some other device). The program parameter should be the generic name of the program that is doing the logging. This name is used to lookup which destinations to use. This information is contained in the logging section of the krb5.conf configuration file. If no entry is found for program, the entry for default is used, or if that is missing too, SYSLOG will be used as destination.
To close a logging facility, use the krb5_closelog() function.
To log a message to a facility use one of the functions krb5_log(), krb5_log_msg(), krb5_vlog(), or krb5_vlog_msg(). The functions ending in _msg return in reply a pointer to the message that just got logged. This string is allocated, and should be freed with free(). The format is a standard printf() style format string (but see the BUGS section).
If you want better control of where things gets logged, you can instead of using krb5_openlog() call krb5_initlog(), which just initializes a facility, but doesn't define any actual logging destinations. You can then add destinations with the krb5_addlog_dest() and krb5_addlog_func() functions. The first of these takes a string specifying a logging destination, and adds this to the facility. If you want to do some non-standard logging you can use the krb5_addlog_func() function, which takes a function to use when logging. The log function is called for each message with time being a string specifying the current time, and message the message to log. close is called when the facility is closed. You can pass application specific data in the data parameter. The min and max parameter are the same as in a destination (defined below). To specify a max of infinity, pass -1.
krb5_openlog() calls krb5_initlog() and then calls krb5_addlog_dest() for each destination found.
STDERR | This logs to the program's stderr. |
FILE: /file
FILE= /file | |
Log to the specified file. The form using a colon appends to the file, the form with an equal truncates the file. The truncating form keeps the file open, while the appending form closes it after each log message (which makes it possible to rotate logs). The truncating form is mainly for compatibility with the MIT libkrb5. | |
DEVICE= /device | |
This logs to the specified device, at present this is the same as FILE:/device. | |
CONSOLE | |
Log to the console, this is the same as DEVICE=/dev/console. | |
SYSLOG[:priority[:facility]] | |
Send messages to the syslog system, using priority, and facility. To get the name for one of these, you take the name of the macro passed to syslog(3), and remove the leading LOG_ (LOG_NOTICE becomes NOTICE). The default values (as well as the values used for unrecognised values), are ERR, and AUTH, respectively. See syslog(3) for a list of priorities and facilities. | |
Each destination may optionally be prepended with a range of logging levels, specified as min-max/. If the level parameter to krb5_log() is within this range (inclusive) the message gets logged to this destination, otherwise not. Either of the min and max valued may be omitted, in this case min is assumed to be zero, and max is assumed to be infinity. If you don't include a dash, both min and max gets set to the specified value. If no range is specified, all messages gets logged.
[logging] kdc = 0/FILE:/var/log/kdc.log kdc = 1-/SYSLOG:INFO:USER default = STDERR
This will log all messages from the kdc program with level 0 to /var/log/kdc.log, other messages will be logged to syslog with priority LOG_INFO, and facility LOG_USER. All other programs will log all messages to their stderr.
If logging is done to the syslog facility, these functions might not be thread-safe, depending on the implementation of openlog(), and syslog().
HEIMDAL | KRB5_OPENLOG (3) | August 6, 1997 |
Main index | Section 3 | Options |
Please direct any comments about this manual page service to Ben Bullock. Privacy policy.
“ | A typical Unix /bin or /usr/bin directory contains a hundred different kinds of programs, written by dozens of egotistical programmers, each with its own syntax, operating paradigm, rules of use ... strategies for specifying options, and different sets of constraints. | ” |
— The Unix Haters' handbook |