Juniper MEDIA FLOW CONTROLLER 2.0.4 -, Administrator'S Manual

Page: 221 / 400

System Log 221
Media Flow Controller Administrator’s Guide Configuring and Using Media Flow Controller Logs and Alarms
Log (streamlog)” on page 220 . The Trace Log can be used to track down problems with a
specific delivery situation; see “Trace Log (tracelog)” on page 223 for details.
Set counter thresholds with stats commands; set email notifications with email commands.
See logging for CLI details. See Table 15 , for severity level options.
The System Log keeps track of all system activities, similar to syslog on a UNIX system. The
System Log provides this information, in this order:
•
Date and time
•
Media Flow Controller hostname and process name (ends with colon)
•
Process date and time, in brackets (may be missing)
•
Event information, may include severity level in brackets, see Table 15 , for levels
Example:
Jan 6 00:10:00 test-vos httpd: [Wed Jan 06 00:10:00 2010] [notice] Apache configured -- resuming
normal operations
Jan 6 05:58:50 test-vos pm[4617]: [pm.NOTICE]: Output from nknlogd (Nokeena Log Manager):
Debuglog socket 17 closed
Jan 6 08:58:27 test-vos login: ROOT LOGIN ON ttyS0: user admin (System Administrator)
System Baseline and Health
Logging can provide more information than any one person can process during any given
moment; finding the required information can be a daunting task. This problem can be reduced
if you take time with each installed system to perform a baseline audit. A baseline audit
identifies normal activity for your system, normal system log entries, normal network traffic for
the system, and how the system reacts to certain conditions.
A baseline audit helps you get the most value from syslog data. In particular, a baseline audit
helps identify what the system was like before you became suspicious of its behavior and
helps you determine what has changed. On a day to day basis the system logs can be used to
determine network health and how well the network and computer systems are running. This
provides the ability to proactively solve issues before the user is aware.
Table 15 Syslog Severity Levels
Level Description
emergency An error from which the system cannot recover/system unusable
alert An error requiring immediate action for operation to continue
critical An unexpected condition/response causing an error for unknown reasons
error An unexpected condition/response causing an error for known reasons
warning An anomalous condition that can be ignored and functioning continue but may affect
operations
notice Normal but significant condition/response that could affect operations
info Normal but significant condition/response that does not affect operations
debug Messages generated by the debugging utility