Cabletron Systems CSX1000 用户手册
Workgroup Remote Access Switch 339
C
ONFIGURING
O
THER
A
DVANCED
O
PTIONS
Log Options
one version of UNIX to the next, the system allows you to set the entire priority value as an integer.
This integer will be prepended to all messages sent to this Syslog server.
This integer will be prepended to all messages sent to this Syslog server.
Note:
You do not have to configure a Syslog Server name. The first Syslog Server configure will
be automatically named Syslog1, the second Syslog2, and so on. Up to ten Syslog Servers
can be configured. (For storing CDR reports, you can select up to three of these servers.)
be automatically named Syslog1, the second Syslog2, and so on. Up to ten Syslog Servers
can be configured. (For storing CDR reports, you can select up to three of these servers.)
L
OG
O
PTIONS
B
ACKGROUND
I
NFORMATION
The Log Options feature expands the system’s log file capability and provides a consistent interface
to the device when working with logging or tracing the activity of a subsystem. Currently, the log
options feature only supports the call detail recording (CDR) subsystem.
to the device when working with logging or tracing the activity of a subsystem. Currently, the log
options feature only supports the call detail recording (CDR) subsystem.
The log reports that allow you to trace the activity of a subsystem can be directed to a specific
location. The reports can be directed to a local log file, or to a UNIX-style syslog server.
location. The reports can be directed to a local log file, or to a UNIX-style syslog server.
An offnode server also aids in management of a site with multiple systems, since all systems can
then send their log messages to a central log server. Note that in the case where multiple systems
are logging their CDR reports to a single syslog server, the system name is used to distinguish
which system logged which report. This makes it crucial that each system be assigned a system
name that is unique within its environment.
then send their log messages to a central log server. Note that in the case where multiple systems
are logging their CDR reports to a single syslog server, the system name is used to distinguish
which system logged which report. This makes it crucial that each system be assigned a system
name that is unique within its environment.
L
OCAL
L
OG
F
ILE
O
VERVIEW
The CDR local log file is a circular file stored in RAM. It contains a fixed number of records. After
the log is full, each new record overwrites the oldest record in the file. This is similar to the system
reports log file that is viewed with the
the log is full, each new record overwrites the oldest record in the file. This is similar to the system
reports log file that is viewed with the
dr
command.
Note:
The CDR local log is intended for diagnostic use and is not suitable for production use as
a CDR log.
a CDR log.
S
YSLOG
S
ERVER
O
VERVIEW
When an offnode syslog server is specified as the destination for log reports, this provides more
direct control over:
•
direct control over:
•
the allocation of disk space
•
the integrity of disk space (redundant, tape backup, UPS, etc.)
•
the ease of data retrieval
•
the management of a multi-node site; all nodes can send their log messages to a central log serv-
er
er
Offnode log servers must be accessible via the system’s LAN port; they cannot be accessed via the
WAN. In addition, it is recommended that the log servers either be located on the same LAN
segment as the system, or that a static route be defined for the log server. If a routing protocol such
as RIP is used to establish a route to the log server, the server will be unavailable for the first 90 to
180 seconds after loading the system — until the route is established. This will cause log messages
to be lost that are generated in the first 90 to 180 seconds of operation.
WAN. In addition, it is recommended that the log servers either be located on the same LAN
segment as the system, or that a static route be defined for the log server. If a routing protocol such
as RIP is used to establish a route to the log server, the server will be unavailable for the first 90 to
180 seconds after loading the system — until the route is established. This will cause log messages
to be lost that are generated in the first 90 to 180 seconds of operation.
When we use the term UNIX Syslog Server, we are, more precisely, referring to the “syslogd”
daemon running on a UNIX system. Syslogd reads and forwards messages to the appropriate log
daemon running on a UNIX system. Syslogd reads and forwards messages to the appropriate log