Log file monitoring is an option available in the performance
monitoring rules. This monitoring agent scans log files on managed
Windows devices for specific strings or expressions, and generates
alerts when they are found. This is useful if you want to be
alerted when a particular condition exists that can be traced
through a log file.
You can monitor a text file generated by any application,
including .htm and .xml files (however, Unicode files can’t be
monitored). After you specify which file to monitor and define
rules using regular expressions, the file will be monitored as long
as the logfile monitoring rule is contained in a ruleset that is in
effect on that device.
The first time text in the log file matches a regular
expression, an alert is generated. The alert is generated only once
for that file even if there are multiple matches. Later, if the
file changes so there is no longer a matching condition, then the
agent begins scanning for that regular expression again and will
generate an alert on the next occurrence of the match.
You can also scan log backup files that are created when a log
file becomes too large and older entries in the file are appended
to a different file (a “rolling” log file). However, "wrapped” log
files, which remove older entries within a single log file to make
room for new entries, are not supported.
For this monitoring option, you must specify the location and
exact name of the file on the managed device, and you specify the
search criteria with a regular expression. When a string in the
file matches the expression, an alert action is generated if you
have defined a Log file monitoring alert type in the
appropriate alerting ruleset.
You can include log file monitoring in any alerting ruleset you
have defined. The following procedure describes the five general
steps for setting up log file monitoring:
Create a log file monitoring rule in an alert
ruleset
Specify which log file to monitor on the managed
devices
Define the monitoring rules for that file, using
regular expressions
Select a severity level for the rule and name the
instance so it will be identified in alerts
Apply action and time rules and save the rule in the
alert ruleset
To set up a log file monitoring rule
Click Tools
> Configuration > Alerting.
Under Alert rulesets, select the ruleset you
want to edit, then click Edit on the toolbar.
In the left column of the Alert ruleset window
that opens, click Alerts. Under the Monitor folder in
the list of alerts, click Log file monitoring.
Click Tasks > New in the right column.
In the Log file monitoring dialog box, type a
name and description for the log file monitoring rule.
To change the frequency at which the item is
monitored, change the Polling interval settings.
Click Log file configuration to specify which
log files are monitored, what you are monitoring for, and how you
will be alerted.
Regular expressions are used to define what content in the log file
should be monitored. When the monitoring service finds a match for
the regular expression in the log file, it follows the alert rules
to notify you of the occurrence.
Click Manage. In the Regular expression
management dialog box, add a descriptive name and a regular
expression, then click Add. Repeat for each regular
expression you want to use for monitoring log files. When you have
added them all, click OK.
You can add as many regular expressions as you want in this dialog
box. Note that you need to create a new rule for each expression
that you want to search for, and each rule is applied to only one
log file. In other words, each rule includes one regular expression
and one log file.
Select a regular expression in the Regular
expression drop-down list.
Enter the path and complete filename of the log file
you want to monitor in the Log file path box. This must be a
specific filename, and only that filename will be monitored (for
example, c:\logs\error.txt)
If you want to include backup files for the log
file, enter the path and complete filename of the backup file in
the Backup log file path box (this step is optional). This
also needs to be a complete path and filename for a specific
file.
Type an Instance descriptive name. This
identifies the log file monitoring rule in the alert notifications
you receive.
Select the severity level you want to apply to this
alerting rule.
If you want to monitor only new entries in the log
file (beginning at the time the monitoring rule is deployed to the
device), click Monitor changes to log files. (This option is
typically used for log files so the agent doesn’t keep scanning the
same existing text.)
If you want to monitor all existing and all new entries in the log
file, click Monitor entire log file. (This option is
typically used to monitor other less dynamic files, such as
configuration files.)
Click OK to add the rule to the list of
logfile monitoring rules.
Repeat steps 4-15 to add other logfile monitoring
rules.
After you have created the logfile monitoring rules you want, you
need to add them to the ruleset. You can add multiple monitoring
rules and apply action and time rules to them, depending on how you
want to be notified when log file changes trigger
alerts.
With the rules listed under Log file
monitoring, click Rule > New in the right
column.
Three boxes or "wells" are displayed at the bottom of the
page.
Drag one or more rules into the Alerts
box.
Click Actions on the left column, then drag
one or more action rules into the Actions box. The actions
you add here will be applied to each rule that you added.
Click Time on the left column, then drag a
time rule into the Time box.
Click OK to add the new rules to the
ruleset.
To view the new logfile monitoring rules in the ruleset, click
Rules summary. Each rule is displayed on a separate line,
and you can edit individual rules or clone a rule and make copies
with different actions, time rules, or severity states. If you want
the alerting rule to affect device health, double-click the rule in
the Rules summary list and select the Health check
box.
After you have added the log file monitoring rules
to the ruleset, click Publish to save the changes to the
ruleset. The changes will be applied to individual devices the next
time you deploy the ruleset, or the next time the device's
inventory service runs.
Notes
Log file monitoring is supported
only for managed Windows devices.
Any time you edit or delete a
rule, you need to publish the alert ruleset that the rule appears
in. The changes you make will not apply to devices until the
ruleset has been published (or until you redeploy the ruleset with
a scheduled task).
For more information about modifying alert rulesets
and defining alert actions, see Configuring alert rulesets.
This feature maps log files into memory to use less
memory during a search. Runtime memory is allocated for this as
linear regular expression searches occur. Because Windows locks the
file when it is mapped into memory, you may encounter issues with
some applications.