 |
|
The
Sentry-go® Quick Monitor Console
|
|
|
The Quick Monitor Console is an optional
component that can be installed on one or more server or desktop machines.
It can be used to ...
- Display alerts generated by Sentry-go Quick Monitors in a central console window
- Show the status of each Quick Monitor
- Show the status of server connectivity
- Give access to a Quick Monitor's web reports
- Configure a Quick Monitor remotely
- Synchronise configuration settings from one monitor to another
- Export configuration settings from a local or remote monitor to a file
- Import configuration settings to a local or remote monitor from a file
- Stop, start or restart a local or remote Quick Monitor
Click
here for more information on the Quick Monitor Console.
|
|
|
Configuring the
Sentry-go® Quick Monitor
|
|
|
During installation, Setup will have copied and configured a
series of default settings, allowing you to start monitoring straight away
with a minimum of fuss. The first time the Monitor is run, it too makes
automatic checks on these settings and disables any features that are not
available, again without any manual intervention being required.
To view these parameters & edit their values, define new monitoring
tasks and further configure alerting options, simply run the Quick Web Monitor
Configuration Utility or click "Configure Server" from
the Quick Monitor Console. This is a quick and easy utility to
use, designed specifically to allow you to configure the monitor to suit
your needs in just a few moments. It is also fully described in the
sections that follow.
|
|
|
| Starting
the Quick Monitor Configuration Utility
|
|
|
The Sentry-go Quick Monitor
Configuration Utility is used to specify the settings that the monitor will use when monitoring the file system on the local server. To start the
utility,
simply run the application from the shortcut placed on the desktop or
the taskbar's Start/Programs/Sentry-go menu
...
Alternatively,
you can
launch the application "goQMFileCfg.exe" directly from
the directory in which the Quick Monitor was installed. By default, this
will be "C:\Program
Files\Sentry-go".
If you are configuring
the monitor using the Quick Monitor Console, you will automatically be
connected to the appropriate server machine - you do not need to logon as
described below.
Logging On
The Configuration Utility allows you to
view and edit the Quick Monitor settings on either the local or a remote
PC/Server. To ensure that all facilities are available to you,
especially when configuring a server remotely, you should run the
Configuration Utility under a user account that has Administrative
privileges. If your user ID has insufficient rights on the server, some
facilities may be unavailable. To enable all features within the
utility, it is recommended that you configure the monitor locally - by
running the utility on the same machine that is running the monitor.
In order to run the utility, you simply need to enter or select
the server on which the monitor is running and your Sentry-go
Administrator's password for that monitor on the target machine. This
password is set during installation, but can be changed by selecting the
"Password ..." button on the login window ...
By default, the Utility will connect you to the local PC or Server. To connect
to a remote machine, simply enter
the Windows name of the required machine and click OK. To select the PC/Server
from a list, simply click the "..." button. The utility's main window contains a number of tabs. Each
defines a specific area of the Quick Monitor and is discussed in more
detail in the sections that follow.
|
|
|
| Monitoring
Files, Directories & Network Shares
|
|
|
The Sentry-go Quick Monitor allows you to easily keep track of changes to files,
directories & directory structures as well as the availability of network
shares. To configure file & directory checking, start the Configuration
Utility and
select the "Files & Directories" tab. To monitor changes made to
the contents of text files using keyword detection, click the
Event Logs/Files tab instead.
The following window will be
displayed, allowing you to add any number of file/directory checks you wish to
perform ...
The main list contains all the files
and/or directories currently being monitored by Sentry-go. To disable an entry,
simply uncheck it in the list.
To specify the response for a particular check,
simply highlight the items in the list to edit the following options at the
bottom of the window. The entries are set independently for each file or
directory check ...
| Run
this check every |
This
option allows you to control how often the check should be run ...
- Using default interval.
Select this option if you want to run the check every X
minutes. When selected, the default monitoring interval (set
at the bottom of the window) will be used.
- Every Hour at XX:YY
Select this option if you want to run the check every hour
at the time specified. In this case, the check will be
performed YY minutes past each hour.
- Every Day at XX:YY
Select this option if you want to run the check daily at
the time specified.
- Every <Day> at XX:YY
Select this option if you want to run the check weekly, on
the day & at the time specified.
To run the check on multiple days (but not daily), repeat
the check for and select the appropriate day.
|
| Respond
after this no. errors |
This value
can be used to specify the number of errors that must occur in succession
before the Quick Monitor takes the indicated action.
By default, this value is set to 1, indicating that an alert will
be raised each time the check fails. If the value was set to 3,
three successive errors must occur before the response is run.
|
| Response
to run |
This option indicates how the Quick
Monitor should perform in the event an error is detected.
- No Response - Notify Only.
Select this option if you simply wish to alert one or more
System Administrators about the error by
e-mail, network messaging or both. In this case, no
corrective action will be attempted by the Quick Monitor.
If this option is selected, one or both of the following actions
will be taken when an alert is triggered ...
- System Administrators
defined on the "Alerting" tab are
notified if the Alert level of the error is equal to the
alert level that the user wishes to receive - or higher
if the Administrator is receiving higher alerts as well.
By default, the alert level will be set to 1. To set an
alternate level (between 1 and 10), simply
select the appropriate value as the "Treat
as" value below.
- Files & scripts
defined on the "Alerting" tab are invoked
via the Quick Monitor's Alert Engine if the Alert level of the error
is equal to the alert level that the script is to be
actioned for - or higher if the script is being run for
higher alerts as well. By default, the alert level will
be set to 1. To set an alternate level (between 1 and
10), simply
select the appropriate value as the "Treat
as" value below.
This option is particularly useful if you wish to handle
the error in a customised way (e.g. to send the error to 3rd party
application or utility) or to notify one or more Administrators using
alternative methods such as SMS or pager.
For more information on the Alert Engine and associated
scripts, click here.
- No Response - Log Only.
If logging is enabled via the "Logging" tab,
either to file, database or both, then details of all alerts
will be written to the appropriate location. If this option
is set, no notification or action will be triggered.
Instead, the alert will simply be written to the log.
- Reboot the Server.
Select this option if the Quick Monitor should reboot the
server in response to this failure. In this case, no
corrective action will be attempted by the Quick Monitor.
Before a reboot occurs, notifications, indicating that a reboot is about to
occur will automatically
be sent to the appropriate System Administrators based on
the Alert level of the error. A short delay will precede any
reboot, to ensure the above Administrators receive their
notification.
If, after the server is rebooted, the same alert is
triggered again, the same Administrators will be notified but
the reboot cancelled. This prevents continual
reboots from being performed.
- Run a Command.
This option allows the Quick Monitor to take automatic
corrective action to the appropriate test. If selected, the "Command to Run"
option is enabled, allowing you to specify a command, batch
file or executable you wish to run in response to
the failure. For example, to clear
down temporary files to reclaim free disk space etc.
After the command has completed, the check that triggered
the alert will automatically be re-run. If it fails once again, the Quick Monitor
will automatically revert this option to "Notify the
System Administrator" and the appropriate users will be
notified as discussed above.
In order to be kept informed of any response the Quick
Monitor takes, you can also configure the following options
on the "Settings" tab ...
- Run without notifying
- Run & notify, but do
not call the Alert Engine
- Run & notify, then
invoke the Alert Engine
- Run a VBScript.
This option also allows the Quick Monitor to take automatic
corrective action to the appropriate test. If selected, the "Command to Run"
option is enabled, allowing you to specify a Windows
Scripting Host (WSH) or VBScript (VBS) file which should be
run in response to the failure.
The same notification behavior as described above will be
run for VBScripts as well.
|
| Command |
If the
above option allows a command or script to be run automatically,
that command or filename should be entered here. Any files must be
available and accessible on the local server and so the use of
local paths in filenames is recommended.
In order to pass error-specific
information on the command line, simply include any of the
in-built variables - e.g. <$$ERROR>, <$$SERVER>,
<$$TIMELOGGED> etc. Click
here for more information on these variable substitutions.
|
| After
response, re-check in (mins) |
If the
specified check fails and the configured does not resolve the
problem, this value can be used to defer any future check for
a given period of time. This can be used to give the Administrator time to
investigate the problem further before being notified again and
prevents continuous notifications being sent or automatic
resolutions attempted.
If the monitor's automatic response resolves the problem, this
value is ignored and monitoring continues as usual.
In most cases, an alert is only triggered if a new file
matching the given criteria is found. The monitor will not
repeatedly generate subsequent alerts for the same file unless the
conditions or properties of the file change. In this case, this
value can remain at 0, so that monitoring continues for other
files etc. where wildcards are used.
|
| Treat As
(Alert Level) |
The value
selected here indicates the Alert level for the corresponding
alert that is raised in the event the check fails. The Alert Level
is used by the Quick Monitor to determine which System
Administrators should be notified and/or Scripts run in response to
the triggered alert.
Click here for more information on Alert
Levels.
|
| Default
monitoring interval ... |
This
value specifies how often, in minutes the corresponding check
should be performed if the monitoring interval (above) is set to
"Using default interval". Setting this option to 0 disables all
checks on the tab that have a monitoring value of "Using
default interval".
|
Adding or Editing a Monitored
File/Directory
To add or edit a monitored file or
directory, start the Configuration utility, click the Files & Directories
tab, highlight any existing entry and click Edit or click the Add button to display the following window.
You can also copy a check by highlighting the appropriate entry and clicking
"copy" ...
From here you can set the following
options in order to configure the file or directory you wish to check ...
| Refer
to this Check As |
This
is the name by which you refer to the check. It is the name
displayed in reports and on alerts and should therefore be
descriptive of the check being made. This name must be unique.
|
| Monitor
this file or mask |
Select
this option if you wish to check one or more files. You can
monitor a specific file, a group of files (e.g. *.txt) within
the same directory or one or more files from the given directory
and all subdirectories below it.
When selected, simply enter the full name & path of the file
in the field below or click "..." to select a file
from Windows. This value can be ...
- The complete filename - e.g.
MyLog.txt
- A partial filename - e.g.
*.txt, My*.log etc.
- An entire mask - e.g. *.*, *
etc.
To check that a mask maps to the file(s) you wish to verify
works, simply access the command prompt, change to the
appropriate directory and type dir <mask>.
System environment variables may be used within
the file name entered - e.g. %WINDIR%. In addition, if the filename contains a
date, the following formats may be used ...
- $$YY to include the 2 character
year
- $$MM to include the 2
character month
- $$DD to include the 2
character day
- $$DD-n where n is a
number greater than 1. Allows you to include a date n-days
in the past. The associated month and/or year are
automatically adjusted as required.
- $$DD+n where n is a
number greater than 1. Allows you to include a date n-days
in the future. The associated month and/or year are
automatically adjusted as required.
- $$DD[-n] to include the
2 character day. The -n will not be altered.
- $$DD[+n] to include the
2 character day. The +n will not be altered.
If date variables are used, the
Quick Monitor will automatically generate the appropriate name
before test is performed, thus correcting the date when the time
passes midnight etc.
|
| Monitor
this directory |
Select
this option if you wish to check a directory, rather than one or
more files.
When selected, simply enter the full path of the directory you wish
to check in the field below. Alternatively, click the
"..." button to select an existing local or remote
directory from a list.
System environment variables may be used within
the directory name entered - e.g. %WINDIR%. In addition, if the
directory contains a
date, the following formats may be used ...
- $$YY to include the 2 character
year
- $$MM to include the 2
character month
- $$DD to include the 2
character day
- $$DD-n where n is a
number greater than 1. Allows you to include a date n-days
in the past. The associated month and/or year are
automatically adjusted as required.
- $$DD+n where n is a
number greater than 1. Allows you to include a date n-days
in the future. The associated month and/or year are
automatically adjusted as required.
- $$DD[-n] to include the
2 character day. The -n will not be altered.
- $$DD[+n] to include the
2 character day. The +n will not be altered.
If date variables are used, the
Quick Monitor will automatically generate the appropriate name
before test is performed, thus correcting the date when the time
passes midnight etc.
|
| Scan sub-directories
below the above path ... |
Check
this option if you want the Quick Monitor to continue to file
scan into all directories below the directory entered (i.e. scan
subdirectories) ...
- If this option is unchecked,
the monitor will check the contents of either the directory
entered, or the directory where the file (or mask) is
starting from.
- If checked, any directories
found below the path entered will also be scanned - as will their subdirectories
respectively until no more are found. For example, to scan
the entire C: drive for a file, enter "C:\" as the
directory, enter the filename of mask and check this
option.
When sub-directory checking is enabled, an alert will
be triggered when the first error condition is encountered -
e.g. when any file in the scan matches the conditions
set.
|
| Exclude files
with these masks |
This
option allows you to enter one or more filenames or partial
filenames separated by commas. If a matched filename equals or contains any of these strings, it will be ignored by the Quick
Monitor. A wildcard (* character) can be used to indicate how
filenames should be matched ...
- To exclude files ending in a
string, start the name with a "*" - e.g. to
exclude any file ending in '.bat', enter '*.bat'.
- To exclude files beginning with
a string, enter the string, followed by "*" - e.g.
to exclude all files beginning with "'L", enter
"L*".
- To exclude files containing a
given string (anywhere within the name), enter the string
with no wildcard character - e.g. to exclude any file
containing "test", enter "TEST".
- To exclude files beginning
with a string and ending with another string, enter a
wildcard in the middle of the name - e.g. to exclude any
text (.txt) file beginning with the word "TEST",
enter enter "TEST*.txt".
The file is excluded if the
name matches any of the checks listed in this field.
Excluded names refer to the
name of the file only - directory names are ignored. See also
below.
|
| Exclude
directories with these masks |
This
option allows you to enter one or more directory names or
partial directory names separated by commas. If a directory or
sub-directory that is being checked contains any of these
strings, it will be ignored by the Quick
Monitor. For example, if the entry is 'Develop, 2004' ...
- The following directories
& sub-directories would continue to be scanned ...
C:\Program Files\Production
C:\Program Files\Production\2005
- While these would be ignored
...
C:\Program Files\Production\2004
C:\Program Files\Development\2005
C:\Program Files\Development\2006
Wildcards should not be
used in this field.
|
Trigger
an Alert if ...
The values that follow allow you to configure when Sentry-go
should trigger an alert.
|
| The
file or directory size ... |
Select
this option if you wish to periodically monitor the total size of one
or more files or directories. When selected, simply enter the
threshold value, then select the comparison - "equal
to", "not equal to", "greater than" or
"less than") and the threshold unit (bytes, Kb, Mb
etc.) An alert will be generated if the above condition is met.
- To monitor the size of a file,
enter the appropriate directory name (under "Scan this
Directory") and the file's full name under "File
or Mask".
- To monitor the size of all
files of a particular type, enter the appropriate directory
name (under "Scan this Directory") and the mask
"File or Mask" - e.g. *.tmp.
To monitor the size of a directory, enter the appropriate
directory name and "*" for the File/Mask
When setting the threshold unit, always bear in mind the
check being made and use the most appropriate value. For
example, if you were checking the size of C:\WINNT and all
sub-directories, the value returned likely to be quite large.
Therefore a higher unit such as Mb or Gb should be used to avoid
overflow errors from occurring.
|
| Individual
file size(s) ... |
Select
this option if you wish to periodically monitor the size of
individual files within a directory or directories. When selected, simply enter the
threshold value, then select the comparison - "equal
to", "not equal to", "greater than" or
"less than") and the threshold unit (bytes, Kb, Mb
etc.) An alert will be generated if the above condition is met.
This check is identical to the
one above, except the size of each individual files is checked
and compared with the threshold, as opposed to the cumulative
size of all qualifying files.
When setting the threshold unit, always bear in mind the
check being made and use the most appropriate value. For
example, if you were checking the size of C:\WINNT and all
sub-directories, the value returned likely to be quite large.
Therefore a higher unit such as Mb or Gb should be used to avoid
overflow errors from occurring.
|
| The
no. files in the directory ... |
Select
this option if you wish to periodically monitor the number of
files in the specified directory, and optionally subdirectories.
When selected, simply enter the threshold value, then select the
comparison - "equal to", "not equal to",
"greater than" or "less than"). An alert
will be generated if the above condition is met.
|
| The
file is updated within the last ... |
Select
this option if you wish to periodically monitor the last update
time of a file and trigger an alert when it is updated within
the given timeframe. When selected, simply enter the threshold
value and it's corresponding unit - "minutes",
"hours", "days" or "weeks".
|
| The
file is not updated within the last ... |
Select
this option if you wish to periodically monitor the last update
time of a file and trigger an alert when it is not updated
within the given timeframe. When selected, simply enter the
threshold value and it's corresponding unit -
"minutes", "hours", "days" or
"weeks".
This check is particularly useful if you have an application
that updates a log file every X minutes and you wish to check
that it's running correctly etc.
|
| Don't
trigger an alert if no matching files found |
By
default, the above option will trigger an alert if a matching
file has not been updated within the given timeframe or no files
matching the mask are found. To ignore conditions where no
matching files are found, check this option.
|
| Less
than X file(s) created within the last ... |
Select
this option if you wish to periodically ensure that a certain
number of files have been created within a given timeframe. If
the specified number of files matching the entered mask have not
been created, an alert will be triggered.
This check is particularly useful if you have an application
that creates log files every X minutes and you wish to check
that it's running correctly etc.
When selected, you must also indicate the number of files that
are expected to be created (e.g. 1 or more), as well as the
timeframe in either "minutes", "hours" or "days".
|
| The
directory or file does not exist ... |
Select
this option if you wish to check that the given directory or
file exists. If it does not exist, an alert will be triggered.
|
| The directory does not exist, or a subdirectory is added or removed
... |
Select
this option if you wish to check that the given directory exists
and be notified if a subdirectory within it (and optionally
below it) is either added or removed.
Only directories are verified with this check. To verify
files as well, simply add a new check using the appropriate
option below - e.g. "A file is deleted from the
directory" etc.)
|
| The directory does not exist, or a subdirectory is
removed ... |
Select
this option if you wish to check that the given directory exists
and be notified if a subdirectory within it (and optionally
below it) is removed.
Only directories are verified with this check. To verify
files as well, simply add a new check using the appropriate
option below - e.g. "A file is deleted from the
directory" etc.)
|
| A
file is added to the directory ... |
Select
this option if you wish to monitor the contents of the given
directory and be alerted when a file is added to it.
|
| A
file is deleted from the directory ... |
Select
this option if you wish to monitor the contents of the given
directory and be alerted when a file is deleted from it.
|
| A
file is updated within the directory ... |
Select
this option if you wish to monitor the contents of the given
directory and be alerted when a file is updated within it.
|
File & Directory Monitoring Examples
To help explain the file/directory
monitoring features of Sentry-go, here are some example settings & options ...
| To check
this ... |
Select |
Select
& Enter ... |
Sub-
Dirs ? |
| |
|
|
|
| Size of the
C:\LOGS directory |
The
file or directory size ... |
Directory,
C:\LOGS |
N |
| Size of all .LOG
files in C:\LOGS |
The
file or directory size ... |
File,
C:\LOGS\*.LOG |
N |
| Size of all .LOG
file in C:\LOGS & below |
The
file or directory size ... |
File,
C:\LOGS\*.LOG |
Y |
| The total no. of
files on the C: drive |
The
no. files in the directory ... |
Directory,
C:\ |
Y |
| The total no. of
files in C:\TEMP |
The
no. files in the directory ... |
Directory,
C:\TEMP |
N |
| Check that no
SETUP.LOG file on the C: drive has been updated within the last 24 hours |
The
file is updated within the last ... |
File,
C:\SETUP.LOG |
Y |
| Check that the
file C:\BACKUPS\BACKUP.TXT was updated within the last 24 hours |
The
file is not updated within the last ... |
File,
C:\BACKUPS\BACKUP.TXT |
N |
| Check that any
file in the C:\BACKUPS directory was updated within the last 24 hours |
The
file is not updated within the last ... |
File,
C:\BACKUPS\* |
N |
| Check that any
.LOG file in the C:\BACKUPS directory was updated within the last 24
hours |
The
file is not updated within the last ... |
File,
C:\BACKUPS\*.LOG |
N |
| Check that at
least 1 new .LOG file is created in the C:\BACKUPS directory within the
scan period. |
Less than X (enter 1) files created within the last ... |
File,
C:\BACKUPS\*.LOG |
N |
| Check that the
file C:\APP\TestMsg.txt exists |
The
directory or file does not exist ... |
File,
C:\APP\TestMsg.txt |
N |
| Check to see if
the file SETUP.LOG has been created in C:\ (root only) |
A
file is added to the directory ... |
File,
C:\SETUP.LOG |
N |
| Check to see if
the file SETUP.LOG has been created anywhere on the C: drive |
A
file is added to the directory ... |
File,
C:\SETUP.LOG |
Y |
| Check that no .EXE
files have been deleted from D:\PROGRAM FILES\MyApp |
A
file is deleted from the directory |
File,
D:\PROGRAM FILES\MyApp\*.EXE |
N |
| Be alerted if the
file Dr Watson.log is updated under C:\WINNT |
A
file is updated within the directory ... |
File,
C:\WINNT\DrWatson.log |
N |
| Be alerted if a
directory is removed from below the path C:\DIRECTORY |
The directory does
not exist or a subdirectory is removed ... |
Directory,
C:\DIRECTORY |
N |
| Be alerted if a
directory is removed from below the path C:\DIRECTORY or any lower
subdirectory below that |
The directory does
not exist or a subdirectory is removed ... |
Directory,
C:\DIRECTORY |
Y |
| Be alerted if a
file is removed from below the path C:\DIRECTORY or any lower
subdirectory below that |
The entered file (or matching mask) is deleted from the
directory ... |
File,
C:\DIRECTORY\* |
Y |
Once you are happy with the settings
entered, click OK to update or add the monitored item.
Monitoring
Remote Drives/Shares
In addition to monitoring local drives, the
Quick Monitor can also monitor remote drives/shares. Simply enter the full
name of the server & share and optionally a file pathname. For example
...
- \\ServerName\ShareName
- \\ServerName\ShareName\Directory\File.log
|
|
|
|
Monitoring
Event Logs & Text-based Log Files
|
|
|
Many
server-based applications, as well as Windows itself write their errors to the Windows
Event Log. In addition, applications may also write messages to
their own text-based log files. However, because both of these are
typically located on the local machine,
proactively monitoring them can be difficult. With Sentry-go, monitoring
messages written to one or more Event Logs, text-based or memory mapped
files is both quick & easy to achieve.
To configure Event Log & log
File monitoring, start the Configuration
Utility and
select the "Event Logs
& Log Files" tab. The following window will be
displayed, allowing you to configure one or more event logs or log files ...
The main list contains all the Event Logs and log files that are currently
being monitored by Sentry-go. To disable an entry,
simply uncheck it in the list.
To specify the response for a particular check,
simply highlight the items in the list to edit the following options at the
bottom of the window. The entries are set independently for each event log and
log file ...
| Response
to run |
Select the action
that you wish to take in the event a failure is written to the
specified event log or log file ...
- No Response - Notify Only.
Select this option if you simply wish to alert one or more
System Administrators about the error by
e-mail, network messaging or both. In this case, no
corrective action will be attempted by the Quick Monitor.
If this option is selected, one or both of the following actions
will be taken when an alert is triggered ...
- System Administrators
defined on the "Alerting" tab are
notified if the Alert level of the error is equal to the
alert level that the user wishes to receive - or higher
if the Administrator is receiving higher alerts as well.
By default, the alert level will be set to 1. To set an
alternate level (between 1 and 10), simply
select the appropriate value as the "Treat
as" value below.
- Files & scripts
defined on the "Alerting" tab are invoked
via the Quick Monitor's Alert Engine if the Alert level of the error
is equal to the alert level that the script is to be
actioned for - or higher if the script is being run for
higher alerts as well. By default, the alert level will
be set to 1. To set an alternate level (between 1 and
10), simply
select the appropriate value as the "Treat
as" value below.
This option is particularly useful if you wish to handle
the error in a customised way (e.g. to send the error to 3rd party
application or utility) or to notify one or more Administrators using
alternative methods such as SMS or pager.
For more information on the Alert Engine and associated
scripts, click here.
- No Response - Log Only.
If logging is enabled via the "Logging" tab,
either to file, database or both, then details of all alerts
will be written to the appropriate location. If this option
is set, no notification or action will be triggered.
Instead, the alert will simply be written to the log.
- Reboot the Server.
Select this option if the Quick Monitor should reboot the
server in response to this failure. In this case, no
corrective action will be attempted by the Quick Monitor.
Before a reboot occurs, notifications, indicating that a reboot is about to
occur will automatically
be sent to the appropriate System Administrators based on
the Alert level of the error. A short delay will precede any
reboot, to ensure the above Administrators receive their
notification.
If, after the server is rebooted, the same alert is
triggered again, the same Administrators will be notified but
the reboot cancelled. This prevents continual
reboots from being performed.
- Run a Command.
This option allows the Quick Monitor to take automatic
corrective action to the appropriate test. If selected, the "Command to Run"
option is enabled, allowing you to specify a command, batch
file or executable you wish to run in response to
the failure. For example, to clear
down temporary files to reclaim free disk space etc.
After the command has completed, the check that triggered
the alert will automatically be re-run. If it fails once again, the Quick Monitor
will automatically revert this option to "Notify the
System Administrator" and the appropriate users will be
notified as discussed above.
In order to be kept informed of any response the Quick
Monitor takes, you can also configure the following options
on the "Settings" tab ...
- Run without notifying
- Run & notify, but do
not call the Alert Engine
- Run & notify, then
invoke the Alert Engine
- Run a VBScript.
This option also allows the Quick Monitor to take automatic
corrective action to the appropriate test. If selected, the "Command to Run"
option is enabled, allowing you to specify a Windows
Scripting Host (WSH) or VBScript (VBS) file which should be
run in response to the failure.
The same notification behavior as described above will be
run for VBScripts as well.
|
| Command |
If the
above option allows a command or script to be run automatically,
that command or filename should be entered here. Any files must be
available and accessible on the local server and so the use of
local paths in filenames is recommended.
In order to pass error-specific
information on the command line, simply include any of the
in-built variables - e.g. <$$ERROR>, <$$SERVER>,
<$$TIMELOGGED> etc. Click
here for more information on these variable substitutions.
|
| Treat As
(Alert Level) |
The value
selected here indicates the Alert level for the corresponding
alert that is raised in the event the check fails. The Alert Level
is used by the Quick Monitor to determine which System
Administrators should be notified and/or Scripts run in response to
the triggered alert.
Click here for more information on using and allocating Alert
Levels.
|
Adding or Editing a
Monitored Event Log or Text-based Log File
To add or edit a monitored file or
directory, start the Configuration utility, click the Event Log & Log Files tab,
highlight any existing entry and click Edit or click the Add button to display the following window ...
From here the following
options are available ...
| Refer
to this log file or check as ... |
This
is the unique name of the check being made. It is this name that
will be displayed on reports and when alerts are generated. It
is recommended that a short descriptive name be used for this
value.
|
| The
type of this file is ... |
The
value selected here indicates the type of file you'll be
monitoring. It can be set to one of the following ...
- Application Event Log.
Select this option if you wish to monitor entries written to
the Windows Application Event Log.
- System Event Log.
Select this option if you wish to monitor entries written to
the Windows System Event Log.
- Security Event Log.
Select this option if you wish to monitor entries
security-written entries written to the Windows Security
Event Log.
- Text-based Log File.
Select this option if you wish to monitor records written to
a named text-based log file. The name & path & of
the file should be specified in the field below.
If the file is located on a different file system such
as a Unix drive accessed through NFS, select "Memory-mapped Log
File" and not this option.
- Memory-mapped Log File.
Select this option if you wish to monitor records written to
a named memory-mapped file. The name & path & of the
file should be specified in the field below.
Memory-mapped
files allow applications to map all or part of a file on disk to
a specific range of addresses within the process's address
space. With memory mapped files, data is often cached and
written in blocks. Although this improves performance, it does
mean that monitoring must be handled differently to a standard
text-based file.
Select this option if the log file is a memory mapped file. For
standard text-based log files, you should select the
"Text-based log file" instead. If you are in doubt as to the type of log file being
produced, please refer to the system's documentation.
However, if the
file is located on a different file system such as a Unix
drive accessed through NFS, always select this option in
preference to "Text-based Log File".
- Windows Event Log.
Select this option if you wish to monitor a named Windows
Event Log. The name of the log should be entered below.
|
| The
Name and/or Path of this file is ... |
Where
applicable, enter the name and/or path of the log file you wish
to monitor.
A log file can be defined multiple times if required, in order
to specify separate actions and detect different sets of
keywords etc.
|
| Event
Type |
Optional.
Check this option and select the type of event you wish to
report -e.g. errors, warnings etc.
If not entered, all event types are used in the scan. Event
Types are only available when monitoring Event Logs.
|
| Event
Source |
Optional.
Check this option and enter the name of the event source (the
source name shown in Event Viewer for a particular message -
normally the name of the application generating the event) you
wish to report.
If not entered, event from all sources are used in the scan.
Event Source is only available when monitoring Event
Logs.
|
| Event
ID |
Optional.
Check this option and enter the ID of the message you wish to
check. Each event log message is generated with an ID that
uniquely defines it within the context of the source application
or system.
If not entered, all IDs are used in the scan. Event IDs
are only available when monitoring Event Logs.
|
Considerations for Other File Systems
In addition to monitoring changes made to
text files on Windows machines, it is also possible to monitor files
written to other file systems, if the file is accessible from the machine
running Sentry-go. For example, using NFS to map to a Unix directory.
However, in this case, you should be aware of the following ...
- The drive mapping must be available to
the user running Sentry-go, typically the local system user.
- When defining the file to Sentry-go,
select "Memory Mapped Log File", not "Text-based Log
File".
Defining
Keywords or Phrases
The Quick Monitor detects errors in Event Logs and/or log files using Sentry-go's keyword detection technology. Keywords or phrases can be
used either to detect an error, or to find errors that you do not wish
to monitor. Both are defined at the bottom of this window ...
| Add
Keyword |
Click
this button to add a keyword or phrase that you wish to monitor
to the list. If the keyword or phrase is found, an alert will be
triggered, unless excluded keywords are also found.
|
| Add
Exclude |
Click
this button to add a keyword or phrase that indicates that the
message should be ignored. If an excluded keyword is found, the
message is automatically ignored, regardless of other settings.
|
| Edit |
Click
Edit to edit an existing keyword or phrase listed.
|
| Delete |
Click
Delete to remove an existing keyword or phrase from the list.
|
| To
Trigger an Alert ... must be present |
This
option determines when an alert should be triggered &
keyword detection is defined ...
- All Phrases.
Select this option if all keywords listed must be
present in the message in order to generate an alert.
(Excluded keywords do not count in this check).
- Any Phrase.
Select this option to trigger an alert if one or more of the
keywords listed are found in the message.
|
Keywords or
Phrases that Trigger an Error
When defining a message, there is no
need to add complete error messages to this list - one or more keywords
is usually sufficient. By default, standard messages (and all event
log errors) are included when Setup installs the Quick Monitor.
The keywords used depends on the file being monitored ...
- In most cases, generic
keywords can be used such as "error", "failed",
"insufficient", "problem" etc.
- To
be notified of any message that contains the word "error",
simply add the word "error" to the included list
(without quotation marks).
- To
be notified of any message that contains the phrase "this is an
error", simply add that phrase (without quotes) to the included
list.
- To
be notified of any message that contains the phrases "this is
an error" and "database", use the [And] escape
sequence within the included list. In other words, you'd add "this is an error [And] database"
(without quotation marks).
|
|
|
|
Logging Alerts to a File or
Database
|
|
|
When an error or fault is detected by the Quick Monitor, you can log
details to a file or database - maybe for further processing, statistical analysis or later
reporting. To do this, start the Configuration utility and select the
"Logging" tab to display the following window
...
The
"Alert Logging" section allows you to define both file and/or
database logging easily and extremely flexibly. Logging
to a File To log any
triggered alert to a file, run the Configuration Utility, select the
"Logging Tab" and check the "Log Triggered Alerts to
a File" option. The following information can then be configured ...
| File
Record String |
This is
the mask for the record that will be written to the file. It can
be made up of one or more system variables, which will
be substituted before the record is written. The variables used
are the same as used by the Alert Engine - click here for
Sentry-go Quick Monitor Variable Substitutions.
For example, for comma-separated
records ...
<$$TIMELOGGED>, <$$ALERTLEVEL>,
<$$SERVER>, <$$TEST>, <$$ERROR>
|
| File
Path/Name |
This
field is used to specify the path/name of the log file you wish
to create. To choose a name/path, click the "..."
button to the right of the field.
Note that the path selected/entered must be relative to
the local server. If configuring a remote machine, navigate to
that machine and select a UNC path - e.g.
\\SERVER\Path.
|
Logging
to a Database This
option is similar to the above file log, except that triggered alerts are
inserted into an SQL database. Any database that supports ODBC can be
used. Simply install a suitable ODBC driver, create an ODBC entry that
uses it and reference it on this panel - see below. You can insert the
data into any database table, file or record using the "Insert
string" defined. This allows you to define the SQL statement,
substituting the actual alert values at run time. To
log alert information to a database table, simply check the "Log Triggered Alerts to
a Database option". The following information can then be configured ...
| SQL Insert String |
This is
the mask for the SQL statement (normally an INSERT) that will be
run in order to insert the alert into the database. It can
be made up of one or more system variables, which will
be substituted before the database connection is made and the
statement run. The variables used
are the same as used by the Alert Engine - click here for
Sentry-go Quick Monitor Variable Substitutions.
For example, ...
INSERT INTO SentrygoAlerts (TimeOccurred, AlertLevel, Server, CheckName, AlertText)
VALUES ('<$$TIMELOGGED>', <$$ALERTLEVEL>, '<$$SERVER>',
'<$$TEST>', '<$$ERROR>')
The statement itself is dependent on the target table definition
and the data types (e.g. string vs numeric etc.) See
"Creating an Alert Table" below for more information. |
| Connecting
DSN |
Select
the data source that the Quick Monitor should use in order to
connect to the target database/table.
Click the "..." button
to create a new DSN through ODBC. Note that this option is
only available if you are configuring the local server.
|
| User
ID |
This
value is used to specify the SQL Server User ID that is to be used with the
above ODBC
connection in order to logon to the database (if you are logging
to a SQL Server database). To use a Trusted SQL Server Connection, leave this and the password entry blank.
For more information on using trusted connections with
Sentry-go, click here.
|
| Password |
This is
the password associated with the above SQL Server User ID. If no
password is required, simply leave this entry blank.
|
Creating an Alert
Table You can create
your Alert table using any format you require in any ODBC-compliant
database such as SQL Server, Oracle, Microsoft Access - e.g. to fit in
with other systems etc.
Example databases/scripts
for Microsoft SQL Server and Microsoft Access can be found in the
"Alert Logging" folder on the installation disk or download
file.
You can also utilise another application's
database table if required. However, when selecting or defining a suitable
table or database file, it is worth bearing in mind the following
definitions for the substitution fields used by Sentry-go ...
| Variable |
Data type |
| <$$SERVER> |
CHAR/VARCHAR
(255) |
| <$$ERROR> |
VARCHAR
(<Length of Message>). It is recommended that either a
text or memo-type field, or a VARCHAR field which is trimmed
(cut to this length or less) is used within the SQL statement. |
| <$$TEST>
& <$$SOURCE> |
CHAR/VARCHAR
(255) |
| <$$INFO> |
CHAR/VARCHAR
(255) |
| <$$TITLE> |
CHAR/VARCHAR
(255) |
| <$$TIMELOGGED> |
CHAR/VARCHAR
(30) |
| <$$TIMENOW> |
CHAR/VARCHAR
(30) |
| <$$OSNAME> |
CHAR/VARCHAR
(255) |
| <$$SERVICEPACK> |
CHAR/VARCHAR
(255) |
| <$$ALERTLEVEL> |
Numeric
1-10 |
|
|
|
|
Sending
Alerts when Errors are Detected
|
|
|
Most of the time, the Sentry-go Quick Monitor will run
silently
in the background on the local server. If an error is detected, there are
typically two ways of reacting to it ...
-
In some cases, the first step may be obvious - e.g.
clear down temporary files or restart a failed service.
-
In other cases,
further investigation may be required. In this case, one or more
Administrators or support personnel need to be
informed ASAP.
The action you wish to take - be it an automated
response or triggered alert can easily be configured with Sentry-go.
Simply start the Configuration utility and select the "Alerting"
tab to display the following window ...
From here you can configure two types of
alerts, those directly sent from the monitor (via a standard e-mail or
network message) and those that are processed by the Alert Engine.
Alert Levels
Alert levels are a fundamental part of the
Quick Monitor alerting system and allow you to control who should be notified
and how they should be contacted when an error is detected.
Alert Levels are numeric values and range from 1 to 10. They are used to
categorise the type of alerts being sent - e.g. to a group of
Administrators, a team or via specific means such as sending a pager SMS
notification. Each monitored
event is allocated an alert level (by default this will always be 1) but
this can be changed when configuring the check. In addition to this, each
Administrator or Alert Script/file is also allocated one or more alert levels.
If the alert level being triggered matches the alert level of the
Administrator etc., the latter will be notified. To be alert a particular
user
to all events, simply ensure that user is allocated all Alert Levels. To receive
alerts of a specific level, select those levels only.
Notifying with Standard E-mails or
Network Messages
To alert one or more Administrators via
e-mail or network message, run the Configuration Utility, select the
"Alerting" tab and complete the entries at the top of the window
...
| Notify
Administrators |
The first
list on this tab displays all defined System Administrators that will be notified
either by e-mail or network messaging
To add a new Administrator, click the "Add
..." button (see below). To remove an entry, highlight it
and click Remove.
|
| E-Mail
Server (SMTP) Name |
This
value is used to specify the name of your SMTP (e-mail) server.
This is the server that the Quick Monitor will connect to when
it needs to notify system administrators (see below) by
e-mail.
|
| using
port |
This
value specifies the port on which the above SMTP is listening
for inbound requests. By default, SMTP servers listen on port 25
and this value should be used unless you know that your server
uses a different port or you have been advised otherwise by your
provider.
|
| Send E-mail
using this ID |
This is
the user ID that will be used to send any E-mail notifications.
It is recommended that this
value is set to "Sentry-go/QM@<YourDomain>"
where <YourDomain> is the domain under which the server
runs as some e-mail servers require a valid domain for
verification. Failing to set this value may result in e-mail
alerts not being sent.
|
| Network
Message Server |
This
value is used to specify the name of the Windows server that
should be used to send network messages (NET SEND). This server
must be running the Windows Messenger service and be available
to the Quick Monitor over the network. By default, this value is
blank, indicating that the local server should be used.
|
Adding
a New System Administrator (Notification)
When you click the Add button against the System
Administrator's
list, the following
window will be displayed, allowing you to define the new Administrator, their
contact details and Alert levels of interest etc. ...
The following
options are available from this window ...
| Please
enter the user's E-Mail Address/Network ID |
This
field is used to specify either the e-mail address or network
user ID that will be used to
notify this Administrator. The type of value entered is
determined by the selection made below.
If an e-mail address is
specified, it should be fully qualified with the full domain
name and must be accessible (or reachable)
from the SMTP server specified on the previous screen.
|
| Use
E-mail Address |
Select
this option if the user is to be notified/alerted by e-mail -
i.e. the above address is a valid e-mail address.
|
| Use
Network Messaging (NetSend) |
Select
this option if the user is to be notified/alerted by e-mail -
i.e. the above address is a valid e-mail address.
|
| When
should they be notified ? |
This list specifies
all the alert levels that this user should be alerted to. Please
select all that apply.
For more information on
alert levels, please click here.
|
| Daily
Summary |
In some
cases, the Quick Monitor will simply run in background - for
example, if the server is running well and no alerts are
triggered, no notifications will be sent. You may, however wish to be
informed that the server is running satisfactorily, or keep a
check on the total number of alerts sent from the monitor on a daily
basis.
To enable this, simply tick the Daily Summary check box. When
checked, this e-mail address will automatically receive a daily
summary report indicating the total number of alerts raised
within each primary category ...
- No. Monitor-Specific errors
- No. Disk Space errors
- No. Event Log &/or /Log File errors
- No. Performance-related errors
Alert Summaries are always
sent by e-mail. This option is therefore disabled if network
messaging has been specified above.
|
When complete, click
OK to return to the previous screen. The new Administrator will
be added to the list. Using the above configuration, alerts
can be sent by network message ...
or by e-mail to any e-mail
client - e.g. Outlook or other e-mail-enabled devices such as
"Mobile E-mail" from your mobile
phone ...
Testing
Notification Settings
To test the sending of an
e-mail or network message, run the Configuration Utility on the server
running the monitor, click the Alerting tab, highlight the appropriate
e-mail address or user ID in the top list and click "Test". The
following window will be displayed, showing progress as the test continues
& any errors that occur.
If an error does occur and the
information displayed doesn't highlight the exact fault (or it directs you
to the Sentry-go log file), click the "Log File ..." button at
the bottom of the window. This will display the Quick Monitor's log file.
Scroll to the bottom to view entries written by the above test.
Automatically
Resending Failed E-mails
By default, e-mails are sent
to the configured e-mail (SMTP) server as soon as the alert is triggered.
However, if this server is unavailable or cannot be contacted, the e-mail
alert will be lost (although other methods will continue to run and the
associated recent errors web report will still highlight the fault). To
change this behaviour, you can configure the Quick Monitor to periodically
attempt to resend an alert in the event the e-mail cannot be sent. To
enable or disable this, simply locate the following value in the local
Windows Registry and edit is as require ...
HKLM
\SOFTWARE
\3Ds
\Sentry-go <Quick Monitor Name>
\CurrentVersion
\Contacts
\DelayedSMTPCheckInterval |
This
value controls whether the Quick Monitor should attempt to
resend failed e-mail notifications and, if so, how often it
should retry ...
- If the above value is not
present or set to 0 (the default), no resends will be
attempted.
- If the above value is set to a
numeric value of 1 or more, any failed e-mails will be
resent every X minutes (where X is the number specified)
until either the e-mail is sent or the alert is overwritten.
|
Warning: Take care when
editing the Windows Registry and do not change any other values unless
asked to do so, or you fully understand the consequences. Making a mistake
here may result in your server or Windows O/S becoming unusable.
Alerting
through the Alert Engine (to a SysLog server, custom e-mail, SMS gateway,
batch file or script etc.)
In addition to defining automatic responses and notifying Administrators
using e-mail or network messaging, the Quick Monitor can also be configured
to invoke the Alert Engine in order to run a batch file, script or
preformatted file. Unlike standard commands or scripts which are
run "as is", invoking the Alert Engine allows values specific to
the error that triggered the alert to be injected (substituted) directly
into the script or batch file at runtime. It is designed
primarily to allow you to alert in any way you like - e.g. pass the error
details to a 3rd party application, call a 3rd
party utility in order to send an SMS or pager message etc.
What
is an SMS Gateway ?
An SMS gateway is an external
service provided by a third party organisation that allows you to forward
information in a predefined format and have them send the details to one or
more mobile phones via the Short Message Service (SMS). Typically, details
can be sent via a standard SMTP e-mail or HTTP. Why
use an SMS Gateway ? Unless
you want to alert an Administrator via SMS, there is no need to use this
service. However, an SMS message is an ideal way of alerting you to a
significant or important fault. By using an external service, rather than
sending the message direct, you have the following advantages ...
-
Access & configuration
is easy
-
No need to set up or
maintain external modem connections
-
You can use your own
provider or shop around for a low cost provider that can interface
with your mobile network
The Alert Script templates
provided with the Quick Monitor contain logic for a number of SMS Gateway
providers
including ...
-
BigFoot
-
ClickATell
-
IntelliSoftware
Although 3Ds (UK) Limited
do not specifically recommend any one solution, these have been tested
successfully with Sentry-go software and, at the time of writing provide a
low cost, affordable solution, If you are looking for an alternate
provider, we recommend that you look for a company based on the mobile
networtks they connect to - some charge higher costs for foreign networks.
How the Alert Engine Works
An Alert Engine script is a file - either a batch, script or preformatted
file that can contain one or more special
"place-markers" as discussed below. When the Engine is invoked, the
Quick Monitor performs the following actions ...
- A monitored error occurs & the Quick
Monitor triggers an alert.
- If the error's alert level is equal to
or lower than this value, the Alert Engine will be invoked.
- When invoked, the Alert Engine first copies
the file to a temporary location.
- The newly copied file is then compiled (rewritten), replacing any place-markers with error-specific values.
- If the file is defined as a batch file
it is then run through the Windows command interpreter & then (the
temporary file is) deleted.
- If the file is defined as a script, it
is then run through the Script interpreter & then (the temporary
file is) deleted.
- If the file is a preformatted file, the
information is used to invoke the appropriate interface. Once
complete, the temporary file is deleted.
Choosing
an Alert File Type
The Alert Engine allows you to define one of 5 types of Alert Engine file.
| To
do this ... |
Use
this ... |
Description
... |
| Run one
or more commands that can be run from the command line, external
utilities etc. in succession (no programming logic required) |
Batch Alert
file. |
This is a
standard batch file containing one or more Windows commands, external
command-line utilities or other batch files. Place markers may be coded
within the file - e.g. as parameters to commands to include
error-specific information. |
| Run one
or more commands, access COM objects & scripting components,
perform programming logic (e.g. logic based on specific
conditions) |
Alert
Script. |
This is
a VBScript or Windows Scripting Host file that contains one or more
script-based commands. Scripts can be used in place of batch files when
more programmatic control is required. Place markers may be coded within
the file - e.g. as parameters to script commands to include
error-specific information. |
| Log
alert details to a SysLog server |
SysLog
file. |
This is
a preformatted file that the Alert Engine uses to log error information
to a central SysLog server. The file controls the server name, port,
severity & facility code as well as the error message itself. |
| Send a
custom format e-mail, containing your own or specific text (as
well as alert-specific information), interface with an external
system (e.g. SMS/pager gateway) that accepts pre-formatted e-mails
as input etc. |
SMTP file. |
This is a
preformatted file that the Alert Engine uses to send a custom format
e-mail to one or more users, or external systems such as an SMS gateway.
Many systems use e-mails as a way of retrieving information from
external sources and this is one of the easiest ways of interfacing the
details to such a system.
You can send a standard e-mail notification to one or more users
by defining them directly under "Notify Administrators" on the
Alert tab. |
| Interface
with an external system (e.g. SMS/pager gateway) that accepts
pre-formatted HTTP (web) requests as input etc. |
HTTP file. |
This is a
preformatted file that the Alert Engine uses to send error information
to an external system using an HTTP (web server) interface. Using a URL,
you can connect to and send information to the external web server -
such as an SMS gateway. |
To configure the Alert Engine and determine which scripts would be run when,
select the "Alerting" tab. The lower half of the window allows
you to define the scripts that the Alert Engine will run ...
From here you can define each
script and determine when they should be invoked.
| These
files & scripts will be run by the Alert Engine ... |
The files
& scripts listed here will be run by the Alert Engine when
the appropriate alert level (or, if configured a higher level)
is triggered. To Add or remove a file from this list click the
appropriate "Add ..." or "Remove" button.
-
Batch files are run through the standard Windows command line
interpreter.
- Windows Scripting Host (WSH) or VBScript are run through the CScript
interpreter.
- SysLog, SMTP and HTTP files are preformatted and contain a
number of specific commands that tell the Alert Engine how to
process the information.
|
Adding a New Alert Engine File
When you click the Add button against the Alert Engine file/script list, the following
window will be displayed ...
The following
options are available on this window ...
| Script
Location |
This
field is used to specify the full path and name of the batch
file or script that is to be run through the Alert Engine. Command line
parameters can also be included here if required.
When editing an existing file, this will be the full path
& name to that file on the server being monitored. If you're
configuring the local server, you can click the "Edit
..." button to edit the file.
When defining a new file, simply enter the full path and
name of the file, relative to the local server. If you're
configuring the local server, you can click the "Edit
..." button to create & then edit the file. For
preformatted files, a template will be provided, allowing you to
quickly & easily supply your own information.
|
| What
type of file does this represent ? |
This
options allows you select the type of Alert Engine script you
are defining - click here
for more information ...
-
Run as a
standard Windows Batch file.
Use this option to define a standard batch file containing
one or more Windows commands, external command-line
utilities or other batch files as well as error-specific
information. Batch file are simpler than scripts & are
best used when you wish to run a series of commands that
need little or no programmatic control.
Batch files are run through the Windows command interpreter
(CMD).
-
Run as
VBScript or Windows Scripting Host
file. Use this option to define a VBScript or Windows
Scripting Host file. Scripts provide greater programmatic
control over batch files, but are more complex to develop,
requiring programming knowledge.
Script files are run through the CScript interpreter (e.g. a
.vbs
file).
-
Send a SysLog file.
Use this option when you want to log the alert to a SysLog
server.
-
Send an SMTP
e-mail file.
Use this option to send a custom/preformatted e-mail to one
or more users or external systems. This can be a formatted
e-mail containing your own text as well as error-specific
information that can be sent to one or more users, or an
external system such as an SMS gateway. Such a system can
then be used to forward the alert on to a mobile phone or
pager etc.
You can send a standard e-mail notification to one or
more users by defining them directly under "Notify
Administrators" on the Alert tab.
-
Send an HTTP
web file.
Use this option to send error/alert information to an
external system using an HTTP (web) interface. As with the
SMTP option above, the HTTP protocol can often be used to
forward information on to an external system such as an SMS
gateway.
|
| Edit
... |
This
button is available if you are updating the configuration on the
local server. Click this button to edit an existing file or
provide a template when a new Alert file is being created.
|
| When
should it run ? |
This list specifies
all the alert levels that this script or file should be run for.
Please select all that apply.
For more information on
alert levels, please click here.
|
Editing an Alert Engine File
Alert Engine files are text files and can therefore be edited using an
text editor such as Notepad. However, when configuring the local server,
you can also use the in-built editor which is often more convenient and
allows new files to be based on predefined templates, thus avoiding
mistakes and speeding up the process of creating the file. To create a
new file or edit an existing one, ensure the file name/path is entered
and click the "Edit ..." button on the above window. The
following will be displayed ...
From here you can edit the file
as required. Click OK to save, or Cancel to abort any changes made. Predefined
Alert File Templates
When you select an Alert SMTP,
SysLog or HTTP format File, a template will be provided for you. In some
cases this will provide a number of example files & providers that you
can use. It will also show a generic template that you can easily modify
for your own provider or server. Simply edit the specific data values and
remove any lines you no longer require.
When editing a template that has multiple examples, always
remove any unwanted data lines from the original example. If you don't,
the last entry found will be used and your file may not work
correctly.
Running
a Batch File or Script through the Alert Engine
Batch file or scripts are standard text
file run through the appropriate interpreter. In this example, a short VBScript is used
to export alert information to an in-house system, through an external
utility called LogInfo.exe.
In particular, it shows how to include the strings "<$$ERROR>"
and "<$$SERVER>" which are then expanded at runtime by the Alert
Engine to include the actual error message and server name prior to
running the script. Batch files are similar, except that standard Windows
commands & external command-line utilities are used in place of
script-based commands.
'
' Sentry-go Quick Monitor Alert Engine Script to log alert information
' via an external LogInfo.exe routine.
'
' See associated Quick Monitor documentation for more details on how to
' specify the information to send via this file.
'
' ------ Local declarations ------
Dim strTextToSend
' ------ Set up parameters ------
strTextToSend = "QM Error - < $$ERROR> on
<$$SERVER>"
Set Shell = CreateObject("WScript.Shell")
' ------ Create & execute the target command ------
Shell.Run """c:\program files\Inhouse\LogInfo.exe"""
""" & strTextToSend & """,,True
' ------ Cleanup ------
Set Shell = Nothing
' ------ End of Script ------ |
Sending a Custom SMTP (e-mail) through the Alert Engine
When the Alert File is defined as an SMTP file, the preformatted commands
can easily be used to interface with an external SMTP server - e.g. to
send a custom e-mail or to interface with an external system such as an SMS
gateway. The following
commands, which are generally used in the format
<Command>:<Value> should be included in this type of file ...
| Command |
Meaning |
| # |
Any
commands or lines prefixed with a # are counted as comments by the
Alert Engine and ignored. Blank lines are also ignored.
|
| QM-SERVER:
<TargetServer> |
Required.
This command is used to identify the target SMTP (e-mail) server -
either by name or IP address.
|
QM-PORT:
<Port No>
|
Optional.
This is the port number on which the SMTP server is listening for inbound requests.
Unless otherwise stated by your provider or Administrator, this is normally
25.
|
QM-SENDFROM: <User>@<YourDomain>
|
Required. This is the e-mail address of the user sending the mail. In general, this should
be in the format <Someone>@<YourDomain> as the server may check the originating domain before
proceeding..
|
| QM-SENDTO: <User>@<Domain>,
<User>@<Domain>, <User>@<Domain> etc. |
Required. This is the e-mail address of one or more users that should receive the mail. Separate
multiple entries with a comma.
|
QM-SUBJECT:
<Title>
|
Required. This is the string that will form the subject for the e-mail. It can contain any
valid text or place markers.
|
| <Alert Text> |
The last parameter (with no command prefix) is the alert you wish to log. It can span multiple
lines and contain one or more place markers.
|
For example ...
#
# SMTP Alert File
#
# This file allows you to send alert details to an SMTP server.
# Simply edit the appropriate details below and configure the Alert
# Engine to call this file for the appropriate alert levels. Any
# place markers found (e.g. <$$ERROR>) will automatically be expanded
# by the Alert Engine at runtime.
#
QM-SERVER: MySMTPServer
QM-SENDFROM: Sentry-go@Company.com
QM-SENDTO: Support@Company.com, SMS@Provider.com
QM-SUBJECT: Critical Error Detected!
# The following (last)
entry is the message we want to log. This
# will be expanded to contain the actual error/alert information
# at runtime.
The error <$$ERROR> occurred on server <$$SERVER> at <$$TIMELOGGED>
|
Logging an Error to a SysLog Server
through the Alert Engine
When the Alert File is defined as a SysLog file, the preformatted commands
can easily be used to interface with an external SysLog server. The
following commands, which are generally used in the format
<Command>:<Value> should be included in this type of file ...
| Command |
Meaning |
| # |
Any
commands or lines prefixed with a # are counted as comments by the
Alert Engine and ignored. Blank lines are also ignored.
|
| QM-SERVER:
<TargetServer> |
Required.
This command is used to identify the target SysLog server - either
by name or IP address.
|
QM-PORT:
<Port No>
|
Required. This is the port number on which the server is listening for inbound HTTP requests.
Unless otherwise stated by your provider, this is normally 514.
|
| QM-SYSLOGSEVERITY:
<Lvl> |
Optional. Indicates which SysLog severity you wish to log the alert under.
If not specified, the default severity level is 1 (alert).
|
QM-SYSLOGFACILITY:
<Lvl>
|
Optional. Indicates which SysLog facility you wish to log the alert under.
If not specified, the default facility is 14 (alert message).
|
| <Alert Text> |
The last parameter (with no command prefix) is the alert you wish to log. It can span multiple
lines, but will be truncated if the length exceeds the maximum size allowed by the SysLog standard.
|
For example ...
#
# SysLog Alert File
#
# This file allows you to send alert details to a SysLog server.
# Simply edit the appropriate
details below and configure the Alert
# Engine to call this file for the appropriate alert levels. Any
# place markers found (e.g. <$$ERROR>) will automatically be expanded
# by the Alert Engine at runtime.
#
QM-SERVER: MySysLogServer
QM-PORT: 514
QM-SYSLOGSEVERITY: 1
QM-SYSLOGFACILITY: 1
# The following (last) entry is the message we want to log. This
# will be expanded to contain the actual error/alert information
# at runtime.
The error <$$ERROR> occurred on server <$$SERVER> at <$$TIMELOGGED>
|
Sending an Error via an HTTP Web Server through
the Alert Engine
When the Alert File is defined as an HTTP file, the preformatted commands
can easily be used to interface with an external web (HTTP) server (and
optionally on to an external system such as an SMS gateway). The following
commands, which are generally used in the format
<Command>:<Value> should be included in this type of file ...
| Command |
Meaning |
| # |
Any
commands or lines prefixed with a # are counted as comments by the
Alert Engine and ignored. Blank lines are also ignored.
|
| QM-HTTPSERVER:
<TargetServer> |
Required. Indicates the name or IP address of the target web/HTTP server. This name is
dependent on your provider .
|
| QM-PORT:
<Port No.> |
This is the port number on which the server is listening for inbound HTTP requests. Unless
otherwise stated by your provider, this is normally 80.
|
| QM-METHOD: GET |
Required. Indicates which HTTP call type the Alert Engine should use.
Currently, this must be set to GET - indicating that an HTTP Get
request is to be performed. This may be extended in a future
version of the Quick Monitor Alert Engine.
|
| QM-URL:
<URL>?<Params> |
Required. This is the URL that is to be called in order to make the GET request. It may contain
one or more parameters as required. Parameters must be URL encoded
- e.g. %20 in place of spaces, no double quotes etc.
|
| QM-SUCCESSTEXT:
<Text> |
Optional. If included, it indicates some text that the server will return to
indicate that the call was successful - e.g. OK. If the text is not returned (or no text is returned), the command will be considered to have failed.
If both QM-ERRORTEXT and QM-SUCCESSTEXT
are included and neither string is found, it is assumed
that the command failed.
|
| QM-ERRORTEXT:
<Text> |
Optional. If included, it indicates some text that the server will return to
indicate that the call failed - e.g. Error. If the text is not returned (or no text is
returned), the command will be considered to have worked successfully.
If both QM-ERRORTEXT and QM-SUCCESSTEXT are included and neither string is found, it is assumed that the command failed.
|
For example ...
#
# HTTP Alert File
#
# This file allows you to send alert details to an HTTP (Web) server.
# Simply edit the appropriate details below and configure the Alert
# Engine to call this file for the appropriate alert levels. Any
# place markers found (e.g. <$$ERROR>) will automatically be expanded
# by the Alert Engine at runtime.
#
QM-HTTPServer: www.YourProvidersSite.com
QM-Port: 80
QM-METHOD: GET
# The following should appear on a single line
QM-URL: http://www.YourProvider.com/TheirPage.htm?
Param1=Value&Text=<$$ERROR>%20on%20server%20<$$SERVER>
QM-SuccessText: OK
QM-ErrorText: ERR
|
Testing Alert Scripts
To test an alert script or file,
run the Configuration Utility on the server running the monitor, click the
Alerting tab, highlight the appropriate script in the lower list and click
"Test". The following window will be displayed, showing progress
as the test continues & any errors that occur.
If an error does occur and the
information displayed doesn't highlight the exact fault (or it directs you
to the Sentry-go log file), click the "Log File ..." button at
the bottom of the window. This will display the Quick Monitor's log file.
Scroll to the bottom to view entries written by the above test.
Handling Alert Engine Script or File Errors
In the event the Alert Engine is unable to complete the request, more
information will be logged to the Quick Monitor's error log file. You can
access this from the Programs menu short cut (or access the file directly
at <Path of Sentry-go Quick Monitor>\goQuickMon.log).
Sentry-go
place markers
Any file or script that is
processed by the
Alert Engine can contain one or more special place markers. These strings,
which are always in the format "<place marker-Name>" allow details specific to the error triggering the alert to
be included within the response. For example, you can include the error
message, title, server name and time logged to be included in, say, an SMS
or pager message or export to another application or database.
To include such information, simply add the appropriate place marker below
within
the command or file and should be treated as a string - e.g. strMessage =
strMessage & " <$$ERROR> on <$$TIMELOGGED>". The
following place markers are common to all Quick Server Monitors ...
| Place
Markers |
Translates to |
| <$$SERVER> |
The name of the local
server |
| <$$ERROR> |
The error message (from
a log file) or the error string raised by the Quick Monitor |
| <$$INFO> |
Returns
specific information about the error - e.g. the name of the
service that failed.
When set, the value is dependent on the alert being
triggered. |
| <$$TEST> or <$$TITLE> |
The name or title of the
test that was performed and ultimately raised the alert |
| <$$TIMELOGGED> |
The time the error was
raised in the format dd/mm/yy hh:mm.ss |
| <$$TIMELOGGED-MDY> |
The time the error was
raised in the format mm/dd/yy hh:mm.ss |
| <$$SOURCE> |
The source/application
name of an event log error or the name of the test etc. |
| <$$TIMENOW> |
The current time in the format
dd/mm/yy hh:mm.ss |
| <$$TIMENOW-MDY> |
The current time in
the format mm/dd/yy hh:mm.ss |
| <$$OSNAME> |
The type of the server's
Operating System - e.g. Windows NT, Windows 2000 etc. |
| <$$SERVICEPACK> |
The latest service pack
string, if applicable |
Additionally, for the File Monitor the following may also be
set. Some are dependent on the error triggered ...
| Place
Marker |
Applies
to |
Translates
to |
| <$$FILENAME> |
File
& Directory |
The
name of the file or mask that was checked & ultimately triggered
the alert |
| <$$CHANGEDFILES> |
File
& Directory |
A
comma-separated list (max. 255 characters) of file(s) that were
affected by the check. |
| <$$EVENTLOG> |
Event
Log |
The name of the Event log where the error was detected |
| <$$EVENTLOGCHECK> |
Event
Log |
The
unique name of the Event Log check (the test name) that triggered
the alert. |
| <$$LOGFILE> |
Text-based
Log File |
The name of the Log File where the error was detected |
| <$$FILEPATH> |
Text-based
Log File |
The name/path of the file being monitored |
| <$$LOGFILECHECK> |
Text-based
Log File |
The
unique name of the Log File check (the test name) that triggered the
alert. |
|
|
|
| Web Reporting
& Monitor Settings
|
|
|
In addition to monitoring & alerting functionality, Sentry-go Quick
Server Monitors also contain an embedded web server, allowing you access
information direct from the server/monitor using a desktop web browser
such as Microsoft's Internet Explorer or Netscape's
Navigator.
To enable or disable Web Reporting
as well as control
other Quick Monitor settings, run the Configuration utility and click the "Settings" tab.
The following window will be displayed ... 
Configuring
Quick Monitor Settings & the Embedded Web Server
The upper section of this window allows
you to control settings relating to the Quick Monitor itself, including the
embedded web server ...
| Wait this
no. seconds after Windows starts before monitoring |
The Quick Monitor runs as a
Windows service and is typically started automatically with other
services. When the server is started, these services and resources can
take differing times to start and may be launched in different
sequences. This, in turn, can cause false alerts to be triggered by
monitoring software which detects failures in services that have yet to
start.
To guard against this problem, you
can set this value, which is the time that must pass
before monitoring commences - i.e. to allow other services to start. The
value entered here is the number of seconds after Windows startup that
must elapse before monitoring commences.
|
| Cache this
no. alerts locally |
This value
specifies how many alerts will be saved at any one time on the local
machine. The more alerts cached, the more will be shown on the
Recent Alerts report. The most recent alerts are saved.
A minimum of 50 alerts can be saved. The higher this value,
the more disk space will be required on the disk containing the
Sentry-go directory.
|
| The web
server listens on port ... |
This
read-only/read-write value shows the port on which the local embedded web
server is listening for requests. It is typically 80 or above. The
value shown is the value that should be used when connecting to the
web server directly from the browser using this format (you can omit
the Port No. if this value is 80, the default port) ...
http://<Server Name>:<Port No>
When connected to the
local machine, the listen port can be updated if required. When
changing this value, the following should be noted ...
-
Always ensure no
other software, or Quick Monitor is using the port you wish to
assign. If two systems attempt to use the same port, the second
will fail and connectivity will be lost. Care should therefore
be taken when updating this value.
-
This value can only
be changed when configuring the local server. The value is
read-only when configuring remotely.
-
Existing web
sessions using the original port will need to be
refreshed/reconfigured to use the new port/URL.
|
| Display
dates in the following format |
The setting
of this option determines the format of dates displayed by the Quick
Monitor. It will be defaulted to the appropriate value during Setup
but can be changed to one of the following values here ...
- Day/Month/Year
Dates are presented in the UK date format - dd/mm/yyyy.
- Month/Day/Year
Dates are presented in the US date format - mm/dd/yyyy.
This setting also affects the
date format used within place-marker variables, such as <$$TIMELOGGED>,
with the exception of those using a named format - e.g. <$$TIMELOGGED-MDY>.
|
| Enable Web
Reports from this Quick Monitor |
Check this
option to enable the embedded web server and access information from
your web browser. To prevent web access from this Quick Monitor,
uncheck this option.
|
| Allow the
Quick Monitor to Automatically Adjust Login Privileges |
The Sentry-go
Quick Monitor runs as a Windows service. As such, if a different
login is required (e.g. to run remote commands via a virtual web
session), that login must be granted "Login as a Service"
access right in order to function.
If this option is selected, the Quick Monitor will automatically
grant the above access right when required. If it is left unchecked,
the System Administrator must grant this access right (or grant
access to a group which has the right) manually before the user can
login.
Note: Once granted, the permission is retained by the user.
Permissions are never revoked automatically by the Quick Monitor.
|
| Restart
the Quick Monitor Automatically on Applying New Settings |
When one or
more settings are updated and you wish the Quick Monitor to use
them, the monitor must be stopped & restarted. You can either do
this manually at a later date or the Configuration Utility can
prompt you when Apply or OK is clicked and changes have been made.
However, to restart the monitor automatically without prompting,
simply select this option.
|
| When
Sentry-go successfully resolves a problem automatically, what would
you like to do ...? |
When the
Quick Monitor takes automatic action in response to a problem (e.g.
runs a command or script, restarts a service etc.), you can be
alerted to the action in a number of ways. This allows you to be
kept informed of all problems, even if the monitor has corrected the
fault automatically ...
- Run without notifying.
Select this option to run automatic actions without notifying
any Administrators that the action has taken place.
- Notify Administrators, but do
not run the Alert Engine.
Select this option to notify associated Administrators by e-mail
or network message when automatic actions are run. All
Administrators configured to receive the associated alert (i.e.
based on the Alert Level) will be notified, but no Alert Engine
scripts will be run.
This option allows you to be notified by standard methods,
without incurring any additional costs for using pager
notification services etc., which can be reserved for
unrecoverable errors only.
- Notify Administrators and run
the Alert Engine.
Select this option to notify associated Administrators by e-mail
or network message and run associated Alert Engine scripts when
automatic actions are invoked. All Administrators configured to
receive the associated alert (i.e. based on the Alert Level)
will be notified, and all Alert Engine scripts (again based on
the Alert Level) will be run.
|
Configuring
Web Server Security
When you access Sentry-go Web
Reports from your browser, you have access to the Monitoring & Alert Engine,
as well as real-time reports and remote command sessions.*
For additional security and to protect access, especially when you wish to open
reports over the internet, you can specify a number of specific a number of
additional security options such as an Administrator's password & allowing
access to specific IP addresses only. You can
also specify a session timeout, which automatically logs out the user after X
minutes of inactivity.
* Remote Command Sessions
can be disabled individually during installation for additional security if
required.
To configure these options, click the "Web
Security" button
to display the following window ...
The following options are available
...
| Current
Password |
Before you
can change any Administrator setting, you must enter the existing
password on this field. if no password is currently define, leave
this field blank.
|
| New
Password |
To change the
Administrator's password, enter the current password (if any) above,
and the new password here. If you do not want to change the
password, leave this entry blank.
|
| Confirm
Password |
If you have
entered a new password, you must confirm the value by re-entering it
here.
|
| When is
login required ? |
The options
here determine when the login page is displayed by the web server.
If the login page is shown, the user must enter the Administrators
password before being allowed to continue ...
- Before accessing any web
reports from this server.
Select this option for maximum security. If selected, the
user must login before displaying any report (except the home
page and Configuration Verification Report).
The login page will be shown for all primary reports
unless the user is already logged in. To timeout sessions
automatically, see below.
- Before accessing non-read-only
reports.
Select this option to request user login for any report that
allows changes to be made on the server - e.g. Virtual
Command Sessions etc.
The login page will be shown for all primary reports
unless the user is already logged in. To timeout sessions
automatically, see below.
- Never (not recommended)
Select this option to disable logins and session timeouts.
This option is not recommended, especially if you allow access
from the internet.
When selected, no login will be required for any report
generated by the Quick Monitor except Virtual Command Sessions
which requires a Windows User/Password to be entered.
|
| Sessions
time out after ... |
This option
controls how long a session remains active when no requests are
made. The higher the value specified, the longer an existing logged
on session remains (and the user not re-prompted for their password
again) even if no requests are made.
This option is not used if logins are not enforced (i.e.
"Never" is selected above).
|
| Client
Access Restrictions |
By default, a
browser from any client PC with connectivity to the server can
access the Quick Monitor's web server. However, you can limit access
to specific IP addresses or a range of IP-addresses by listing them
here ...
- To allow access to a specific
address, enter it's IP address in the format a.b.c.d - e.g. 189.57.21.5
- To allow access to a group of
addresses, enter a partial address (the numbers that are common)
- e.g. 189.57.21
Use this option if your company uses DHCP where clients
can receive different addresses within given range.
If no entries are listed, any
client can access the web server. To disable all access, unselect
the "Enable Quick Monitor Web Reports" from the Settings
tab.
|
|
|
|
|
Applying New
Settings
|
|
|
The Sentry-go Quick Web Monitor reads many
of it's configuration options when it is started. If changes are made to
it's settings using the Configuration Utility, then simply stop and
start the service in order for them to take effect. You can do this in
one of three ways ...
-
By manually restarting the
service.
-
When you click Apply within the Configuration
utility *.
-
When you exit the Configuration
utility *.
* By setting the "Restart Quick Monitor
Automatically on Applying New Settings", you can control whether
the above prompt is displayed or not.
Restarting the Quick Monitor
To restart the Quick Monitor
manually, follow these steps ...
- On Windows NT, run the Services
applet from Windows Control Panel. On Windows 2000 and above, open the
"Administrative Tools" folder within Control Panel, and run Services form there. A window similar to that shown below will
be displayed ...
- By default, the list shows all the
available services on the local machine, so scroll down to highlight
the appropriate monitor.
- On Windows NT,
click Stop, then, when the service has stopped, click start.
- On
Windows 2000 and above, right click/restart to stop & start the monitor.
- Verify that the service starts by
checking the status within the Services window.
- If an error occurs or the status
doesn't change to "Started", refer to the monitor's log file
for more information. By default, this is a text file called "goQuickMon.log"
located in the Sentry-go home directory, defined during Setup.
|
|
|
|
Retrieving
a Daily Summary Report
|
|
|
Assuming that the local server/file server is working well - which is obviously the aim, the
Sentry-go Quick Monitor will remain
silent. When a fault is detected,
the appropriate alert will be triggered automatically. In order to
ensure the overall health of the system, you can receive a
"Health Report" or "Daily Summary" on the previous day's server errors - e.g. to
confirm that no problems were detected, or to total those that were.
The daily summary is automatically
generated
if any defined System Administrators have requested the report. To receive
the report, select the "Alerting" tab and Add the appropriate
Administrator's e-mail address via the "Add ..." Button. On the
associated Add dialog, simply check the "Send Daily Summary to this
user" option.
Summary information is sent daily, usually between midnight and 1 o'clock
in the morning. Server Totals are cached in memory. If the Quick Monitor is stopped or
reconfigured (and therefore restarted), the totals so far for that period will be reset.
|
|
|
|
HTML/Web Reporting
|
|
|
In addition to monitoring and alerting, there may be times when you wish to access
captured or logged information, or view the data being verified by the
monitoring engine in real-time. With Sentry-go, you can easily access all
this information, and more directly from your desktop browser.
System Requirements
The Sentry-go Quick Monitor includes an
integrated web server built-in, allowing you to display HTML reports directly from
the monitor itself, without the need to install, configure or gain access
to an external web server.
Web-based reporting is enabled through the
"Settings" tab within the Quick Monitor's Configuration Utility. All
you need is a web browser - we recommend you use Microsoft's Internet Explorer
v5.x (or above) or Netscape Navigator v6.x (or above) browsers.
Securing Web Reports & Web Server
Access
Unlike generic web servers such as IIS or
Apache, the integrated web server is dedicated to providing reports
directly from the Quick Monitor. In terms of security, this means that
only the reports documented here can be accessed - unknown scripts cannot
be executed and downloads not performed, thus ensuring a level of security
out of the box. However, in some cases, you may wish to restrict access to
reports and/or the web server further and so the following options are
provided ...
- Some or all reports can be password
protected, requiring the user to logon to the server before accessing
the information. Logged-on sessions can also automatically be timed
out after a configurable period of inactivity.
- Access to the Quick Monitor's web server
can be restricted to one or more specific IP addresses, or a range of
addresses - e.g. addresses on the same subnet etc. If this option is
enabled, only browsers logging in from clients with these IP addresses
will be allowed access to the server.
- The Virtual Command Session
require an additional login when run. This must be a valid Windows ID
(and password) on the local server.
- The Virtual Command Session can be
specifically disabled by the Administrator.
The web server can be enabled or disabled
through the "Settings" tab on the
Configuration Utility. Web security options are configured by clicking the
"Web Security" button - please click here
for more information.
|
|
|
|
Displaying
Web Reports
|
|
|
You can easily access any HTML report in one of
three ways - by entering the report's URL into your web browser, clicking
on a desktop shortcut or menu or using the Sentry-go Web Launcher.
The monitor's web server must be enabled in order to access these
on-line reports.
Using a
Shortcut
If you are running from the machine running
the Sentry-go monitor, Setup will have created a set of shortcuts (icons)
both on the desktop and the Start/Programs/Sentry-go menu. These shortcuts
can also be copied to other machines such as your own desktop, thus
allowing access to monitored values directly from there. Simply double click the appropriate icon to launch the default web browser
and connect it to the appropriate Sentry-go Quick Monitor.
Using the
Configuration Utility
If you are running the Quick Monitor
Configuration utility, you have immediate access to web reports. Simply
click the button at the bottom left of the window to launch your default
browser and access the home page ...
Using the Report's URL
Accessing an HTML Report in this way is as
simple as entering a URL into your web browser, with the exception that you may also need to specify a port address along with the page itself.
All web servers listen for client requests on a
"port". By default, a web server's default port is port 80. However, if this port is already being used on the local machine - e.g. by another web server or
another Quick Monitor, Setup will automatically allocate a new number - e.g. 81, 82 etc.
Finding the
Port Address
If you do not know the port address, run
the Configuration Utility and select the
"Settings" tab. The Web Server's listen port address is the
value shown against the "The Web Server listens on port"
entry.
- If the web server is listening on the
default port (80), do
not need to include the port number in the URL. By default, web browsers will attempt to connect
on this port, so you simply need to enter the
following URL into your web browser ...
http://<Server Name>/QMHome.htm - e.g. http://MyServer/QMHome.htm.
- If the web server is listening on a
different port, you can access the Quick Monitor's Home Page by entering the following
URL into your browser's address field, substituting the appropriate port number as required ...
http://<Server Name>:<Port Number>/QMHome.htm - e.g.
http://MyServer:83/QMHome.htm.
Using
the Sentry-go Web Launcher
Although you can access any report directly from the browser or shortcut
icon, you can also connect to any Quick Monitor on any available machine
using the Sentry-go Web Launcher. This utility allows you to quickly & easily start your
default browser
and connect it to the home page of any local or networked Quick
Monitor.
Running the Sentry-go Web Launcher
To run the Sentry-go Web Launcher, simply logon to a server where the
Sentry-go Quick Monitor is installed and double click the "goQMConnect.exe"
icon from Windows Explorer or the Sentry-go menu if installed. You may also wish to place this in your Startup
folder so that it automatically starts each time you logon to your
desktop. When run, the following window will be displayed ...
To launch your local web browser and
connect it automatically to the Quick Monitor's home page, simply enter
the Server's name or address and click OK. To select the server from a
list, click the "..." button to the right of the field. By
clicking OK, the browser will be launched and the above window closed. To
continue to run this application after the browser is run, click the
Apply button instead of "OK".
Searching for Quick Monitors
In order to locate the Quick Monitor's Web
Server, the utility automatically scans a range of addresses. By default,
this range is between 80 and 100 but if the Quick Monitor listen address
is outside these this, you can increase it by selecting the Options
button. The following window will then allow you to modify the range
scanned ...
|
|
|
|
The
Quick Monitor Home Page
|
|
|
The Quick Monitor's Home Page lists all available reports on the given server (for all Quick Monitors
installed) and provides links to
them. If multiple Quick Monitors
have been installed in different locations on the local server, only reports for
the Quick Monitors installed in the same directory as the connected
monitor will be displayed.
For each report, a link and brief
description will be shown. To access a specific report, simply click on the appropriate link for the given monitor.
| To
access this report ... |
| From the desktop |
Select the menu shortcut,
desktop shortcut or run the
Sentry-go Web launcher. |
| Enter this URL in your browser |
http://<Server Name>:<Port>/QMHome.htm |
|
|
|
|
Password
Protecting Web Reports
|
|
|
For added security, you can password protect
some or all Sentry-go web reports. When a logon is requested, you must enter the Sentry-go password
which is the "Administrators" password specified during Setup
and configured using the Configuration Utility's "Settings" tab
...
The
Quick Monitor Home Page never requires a password. For added security, the
"Virtual Command
Session" report, which allows you to execute remote commands on the server
running the Quick Monitor can also be disabled during installation.
|
|
|
| Recent
Alerts Report
|
|
|
The Recent Alerts report gives you immediate access to the last X errors that
have been triggered from the Quick Monitor on the specified server. The actual
number available is dependent on the number being cached by Sentry-go - this
is configured on
the Settings tab of the Configuration Utility. From this report, it is easy to see
...
-
The errors that have recently
been detected on the server.
-
Spot trends that might have
occurred
-
View other potentially
related issues that have arisen on the server for problem solving.
-
Ensure fixes have corrected a problem.
You
can also access alerts that you may not have been notified about - e.g. because
the associated alert level does not match your alert settings, or you weren't logged on when a
network message was sent etc. An example of a similar report
is shown here ...
| To
access this report ... |
| From the Quick Monitor Home
Page |
Click the Recent Alerts link |
| Enter this URL in your
browser |
http://<Server Name>:<Port>/FileErrors.sgp |
| Logon Used within Report |
N/A |
|
|
|
| Current
Status Report
|
|
|
The current status report allows you to connect directly to the Sentry-go
engine and display the values it is currently monitoring and comparing to the defined
thresholds. An example of a similar report is shown here ...
From here, the last known status
of all files & directories being monitored is displayed, along with the
most recent Event Log & log files messages recorded.
| To
access this report ... |
| From the Quick Monitor
Home Page |
Click the Current Status
link |
| Enter this URL in your
browser |
http://<Server Name>:<Port>/FileStatus.sgp. |
| Logon Used within Report |
N/A |
|
|
|
| Virtual
Session Report (Remote Command Execution)
|
|
|
The Virtual Session page allows you to connect to and run commands directly
on the remote server. It is designed to allow one or more batch commands
to be executed on the server without the need for additional 3rd party
software or a high speed network connection.
With
a Virtual Session, you can ...
The Virtual Session should not
be used for ...
Recommendations ...
It is recommended that
commands are batched and run as a complete unit, rather than issuing many
separate calls. Not only is this quicker, but commands run separately are
effectively run as independent processes, and therefore may not behave as
expected.
For example, if you want to copy a file from the server to your PC, but
cannot access a suitable server-side share, you can map a share from the
server to your PC, copy the file, then delete the mapping through a
Virtual Session. If this is performed as three separate commands (i.e.
clicking "Run Command(s) Now" for the map, copy and un-map
separately, the mapping may not be available during later sessions.
However, if all three commands are issues within the same session
(recommended), the request is not only performed more quickly but ensures the underlying
mappings work correctly ...
net
use X: \\MyPC\TempShare
copy c:\Logs\MyLog.log X:
net use X: /d
When accessed, a web page
similar to this will be displayed ...
The page is split into two
main sections. The first section allows you to authenticate with the
server whilst the second section allows you to enter one or more commands,
configure timeouts and display the results of any previous commands run.
To run one or more commands, you must first logon to the server by
entering the requested details. You can do this either by specifying login
information at same time as the commands you wish to run, or before
entering any commands.
Remote Server Login
The following fields allow you
to logon to the remote server ...
| You
are attached to server |
This
is the name of the server to which you are currently attached or
connected.
|
| User
ID for the remote server |
Enter
your Windows User ID for the server. You can either authenticate
to the domain (entered below) or the remote server
directly.
|
| Password |
This
field allows you to specify the password for the above user. For
security, authentication takes place directly on the target
server.
|
Execute Remote Commands
The following fields allow you
to specify, execute and display results from the remote server ...
| Previous
Return Code |
This
is the code returned from the last series of commands that run.
It may be set to ...
- N/A - no command has been run
or no information was available from the last command.
- Timed out - the remote command
ran but did not complete within the given timeout.
- Success - The last command ran
successfully and returned a result of 0.
- <Code> - The last
command ran successfully and returned a non-zero return
code. Depending on the command, this may indicate success or
failure.
|
| Previous
Results |
This
is the output generated from the last command. It can be
standard output (i.e. the result) or an error message, or both.
If multiple commands were run, output from each will be
displayed here.
|
| Command(s)
to Run |
This
field allows you to specify one or more batch commands that you
wish to execute on the remote server. Each command should be
entered on a new line, just like you would entered commands in a
console window.
When run, all the commands entered here will be batched into a
single file and executed on the remote server.
|
| Commands
time out after |
This
value specifies how long (in seconds) the Quick Monitor will
wait before the command(s) are timed out. If a timeout occurs,
an error will be reported and the batch file terminated on the
remote server.
The default value is 30. To wait until completion (i.e. no
timeout), enter 0 for this value.
|
| Don't
wait for completion |
If
you wish to execute remote commands but not wait for them to
complete, select this option. if this option is selected, no
output will be returned and job cancellation will not be
available.
|
| Show
output lines |
Enter
the number of lines you wish to display in the output window.
The value will take effect the next time you submit a request.
|
| Show
command lines |
Enter
the number of lines you wish to display for the command window.
The value will take effect the next time you submit a request.
|
Logging On & Running
Commands
To logon, run entered commands
or both, simply select the "Logon and Run Command(s) Now"
button. If a timeout value (or no timeout value) has been specified, the
commands will be submitted and the page redisplayed when a result (or
timeout) is returned. If the "Don't wait for Completion" option
has been checked, the page will redisplay after immediately after the
command is submitted on the remote server.
Cancelling Running Commands
In some cases, you may wish to
cancel a command that is already running. To do this, simply click the
"Cancel Running Command(s)" button. If the command can be cancelled,
the Quick Monitor will terminate the remote process and redisplay the web
page.
Note that terminating (or timing out) a process should be avoided
wherever possible & may not work in some cases - e.g. if a separate
process was spawned on the server. Depending on the commands being
cancelled, unpredictable results may occur if running commands are
abnormally terminated.
| To
access this report ... |
| From the Quick Monitor
Home Page |
Click the Virtual
Command Session link |
| Enter this URL in your
browser |
http://<Server Name>:<Port>/QMRExec.htm. |
| Logon Used within Report |
Windows login information
entered |
|
|
|
| Verify
Configuration Report
|
|
|
This report allows you to verify the Quick Monitor's current
configuration and highlights any errors that were encountered since the
last monitor's startup. It also allows you to trigger a test alert and
verify that alerting mechanisms, e-mail addresses and alert engine scripts
are performing correctly without having to wait for an alert to be
generated. A page similar to this will be displayed ...
Verifying the Monitor's Configuration
The report allows you to verify the current configuration being used by
the Quick Monitor. The monitoring engine itself handles configuration errors in the following way,
which are highlighted on this page ...
-
Each time the Quick
Monitor is started, all defined configuration values are loaded and checked.
-
If an error occurs - e.g.
a configured service is not installed on the server, an alert is
triggered and the check disabled. An entry is also written to this
report.
Note that if the
Quick Monitor's web server is not enabled, configuration errors are sent
as notifications to the appropriate System Administrator(s).
Verifying the Alerting
Mechanism
The lower section of the
report allows you to select an alerting option and trigger a test alert.
When run, the Quick Monitor automatically triggers an alert of the
appropriate level and ...
-
Sends network
notifications to all users configured for this alert level.
-
Sends e-mail notifications
to all users configured for this alert level.
-
Runs any configured Alert
Engine scripts or files for this alert level.
If, after the alert is
triggered you do not receive the appropriate notification, or the expected
scripts are not run, please check the appropriate entries within the
Configuration Utility. For more information on Alert Levels and
notifications, click here.
| To
access this report ... |
| From the Quick Monitor
Home Page |
Click the Verify
Configuration link |
| Enter this URL in your
browser |
http://<Server Name>:<Port>/QMConfigCheck.sgp. |
| Logon Used within Report |
N/A |
|
|
|
Customising
Web Reports with your own Headers & Footers
|
|
|
The reports detailed above come pre-formatted when you access the web
server from your desktop browser. However, there may be times when you
wish to customise the layout of these reports in terms of the header
and/or footer - for example, to integrate Sentry-go reporting into your
own support web site etc. The shaded areas below shows these header &
footer areas in a typical report ...
To customise these areas, you
simply provide your own HTML files that are merged into the generated
reports at runtime, replacing the default header/footer provided.
Independent Headers &
Footers Independent
headers & footers allow you to define a header and/or footer that is
applied to a particular web report. If both an independent header/footer
and a common one is defined, the former will always be used. When defined,
the HTML file will be used to generate the areas shown above. If required,
you should also add the report's title to the custom header file. To do
this, simply follow these steps ...
-
Locate the Quick Monitor's
HTML directory. This is normally a folder called "HTML"
within the directory where the monitor is installed - e.g. C:\Program
Files\Sentry-go\HTML. Your HTML files should be placed here.
-
Locate the Quick Monitor's
Images directory. This is normally a folder called "Images"
within the above HTML directory - e.g. C:\Program
Files\Sentry-go\HTML\Images. Your graphics files should be placed
here.
-
Create your header and/or
footer files for each report you wish to customise. These files ...
- Should be standard HTML files
- Should not include <HTML>, <BODY>, <Head> tags
etc.
- Should be called <Name of Report>-Header.htm and/ or <Name of
Report>-Footer.htm etc. ...
- So, to override the login window, the 2 files would be called
QMLogon-Header.htm & QMLogon-Footer.htm
- To override the Web monitor's status report, the 2 files
would be called WebStatus-Header.htm &
WebErrors-Footer.htm etc.
-
Place graphics in the
Images directory identified above.
-
Set the source for any
graphics included in the above headers/footers to
"images/<Name of File>" ...
- e.g. <img border="0" src="images/MyGraphic.jpg" width="650"
height="100">
-
For example ...
<!--
Example Custom HTML Header File - By giving this file the
appropriate name and placing it in the Quick Monitor's HTML
directory, it will automatically replace the report's standard
header block
//-->
<p align="center"><img border="0"
src="images/CompanyHeader.jpg"
width="450" height="150"></p>
<p align="center"><font face="Arial Black" size="4"><i><u>
Company Support System - Recent Server Errors!</u></i></font></p>
|
Common Header & Footer A
common header & footer allow you to define a single header and/or
footer file that is applied to all reports generated by the Quick Monitor.
If both an common header/footer and an independent one is defined, the
latter will always be used. When defined, the HTML file will be used to
generate the areas shown above. In addition, the report's title will also
be generated automatically below the contents of the header file. To do
this, simply follow these steps ...
-
Locate the Quick Monitor's
HTML directory. This is normally a folder called "HTML"
within the directory where the monitor is installed - e.g. C:\Program
Files\Sentry-go\HTML. Your HTML files should be placed here.
-
Locate the Quick Monitor's
Images directory. This is normally a folder called "Images"
within the above HTML directory - e.g. C:\Program
Files\Sentry-go\HTML\Images. Your graphics files should be placed
here.
-
Create your header and/or
footer files for each report you wish to customise. These files ...
- Should be standard HTML files
- Should not include <HTML>, <BODY>, <Head> tags
etc.
- Should be called QM-Header.htm and/ or QM-Footer.htm etc. ...
-
Place graphics in the
Images directory identified above.
-
Set the source for any
graphics included in the above headers/footers to
"images/<Name of File>" ...
- e.g. <img border="0" src="images/MyGraphic.jpg" width="650"
height="100">
|
|
|
|
Uninstalling the Sentry-go
Quick File Monitor
|
|
|
In
the event you
wish to remove the Sentry-go Quick Monitor from the local PC, simply run
the "Add/Remove Programs" applet from the Windows
Control Panel. A window similar to the following will be displayed. From
here, scroll down to the appropriate Sentry-go Quick Monitor entry
and click the appropriate remove button ...
Setup will prompt you to
confirm the software's removal ...
-
Click "Yes" to remove the
Quick Monitor software. Because some components are shared across
the installation, only specific files & shortcuts will be removed
- see below.
-
Upon completion,
Setup checks to see if any other Sentry-go software is installed on the
machine. If it isn't, you will be prompted to remove shared Sentry-go
settings and configuration options. This allows all shared settings and
files to be removed from the PC.
In most cases, if you are sure that all components were
installed in the same Sentry-go directory and all have been removed
(which this prompt implies), it is
recommended that all files and configuration options are
removed from the local machine by selecting "Yes" to this
prompt.
|
|
