SIC Event Logging Overview (SRC CLI)
SIC log streams capture different groups of server-related events at various levels of granularity. You can configure the SIC to capture any number of log streams. If you configure multiple log streams, make sure you configure unique names for each log stream by using the shared sic group identifier server identifier logger identifier statement. Each log stream you create captures events in a separate log file, which is date stamped, and you can also assign a prefix to it for easy identification.
Log messages are divided into several log groups according to the subject of the log information. You can configure a log stream to display only log messages from particular log groups. Each log group captures different types of server-related events. You configure the level of granularity captured for the log group by setting the event level for the log group.
Log File Options
You use the configuration options described in Table 1 to define the properties of the log files.
Table 1: SIC Log File Options
Prefix added to the log file name. This string is prepended to each log file name.
Filter to define which event messages are logged or ignored. The filter specifies the logging level, such as debug.
If set to true, log messages are immediately written to the log file without buffering. Use this setting for real-time logging.
If set to false, SIC log messages are kept in the buffer until the buffer is full and then all messages in the buffer are written to the log file. Use this setting for performance optimization, when real-time logging is not needed.
Footer message added to the end of each log file.
Header message added to the beginning of each log file.
High-resolution time reporting system functions are used.
New log file created after these many bytes. When a log file reaches this size, logging begins in a new log file.
Prepend each log message with additional information. Add time, thread, and transaction information to each log message. You can achieve additional fine-tuning by using the work-id-label, work-id-padding, and utc options.
New log file is created after this amount of time elapses. Specified in seconds.
New log file is created every time the server starts.
Time and date values reflect Universal Time Coordinates (UTC), formerly known as Greenwich Mean Time or (GMT). Otherwise, values reflect local time.
Work data ID prefix added to each log message.
String added to each log message if work data is not available.
The event level specifies the level of detail captured for the log group. You configure the event level by specifying the log group and then specifying the associated event level. The event level you specify is the highest event level displayed for the log group. You can configure the log stream to display log items from levels at and below a particular event level.
Be careful when using event logging because it consumes server resources while capturing events, and consumes disk space to store the log files. We recommend that event logging be used primarily for troubleshooting purposes, and that you limit the amount of information captured in a log stream to control the consumption of server resources and disk space. Limiting the amount of information in the log stream also makes it easier to interpret the information in the log files. For example, you might configure one log stream to capture only configuration-related events by setting the Configuration log group event level to Detail, and setting all other log group event levels to Error.
In general, each event level includes less verbose event types. For example, if you configure an event level of Warning, then warnings and errors are logged to the specified log stream. The event levels in order of increasing verbosity are shown in Table 2.
Table 2: SIC Event Levels
No events will be logged for the log group.
An error as an event that may cause the system to operate incorrectly. Examples include exceptions being thrown, an inability to continue processing a transaction, or configuration errors that cause a component to fail to start.
Errors and warnings are logged. Warnings are less severe but more verbose than errors, in that a warning should be logged when the system was able to handle an unexpected input or condition without any threat to the operation of the server. Examples of warnings include invalid packet contents or failures in contacting remote servers.
Errors, warnings, and standard messages are logged. Standard logging messages show events as a result of normal operation.
Messages in the log are shown at event levels error, warning, standard, and detail. Detail logging is intended to inform why and how the particular result indicated by standard logging was reached. Server components that perform significant processing on the transaction, such as determining validity of the packet contents, log details about decisions they made. All server components that route the transaction through different processing based on the nature of the transaction log their routing activity at this level. The detail log is allowed to refer to the contents of messages logged at the standard level; that is, it will never be read without the standard messages.
Messages in the log are shown at event levels error, warning, standard, detail, and debug. Debug logging is provided for engineering troubleshooting only.
Log groups specify the type of server functionality for which you want to log events. The log groups listed in Table 3 are available.
Table 3: SIC Log Groups
Reports events related to server administration, such as:
Reports events related to configuration.
Reports events related to transaction processing.
Displays content of a packet in an <attribute name>:<attribute value> format.
Displays content of a packet in its raw (octets) format.
Reports events related to the system, such as: