SICM How to ...
- ... monitor using ping
- ... monitor using snmp
- ... add a snmp template
- ... create an e-mail alert
- ... split one device across multiple menus
- ... monitor a frame relay device
- ... change the alignment of the web output
- ... stop the images from caching on IIS
- ... send multiple e-mails
- ... ignore invalid input values
- ... run multiple instances of SICM
A collection of information, tips and answers to questions to improve the operation of SICM
Two methods are available for monitoring a device using ping:
- To use the ping functionality built into SICM, in config.xml:
- Delete or comment out
<ping_file>path\ping_file</ping_file>
- Set
<timeout>5</timeout> as appropriate
- Set
<ping_count>4</ping_count> as appropriate
- To use ping provided by the Operating System, in config.xml:
- Set
<ping_file>path\ping_file</ping_file> as appropriate
- Set
<timeout>5</timeout> as appropriate
- Set
<ping_count>4</ping_count> as appropriate
- Add the devices to the target file
- Add
<template template_id="ping">
</template>
to the target file
- Stop, then run SICM in test mode to validate
- Start SICM
- View the results after allowing 2 * frequency time to pass
- For further information, refer to: SICM Configuration and Operation Guide
- Ensure the snmp agent on the device is active. Use
Getif on Win32 or
net-snmp on Linux to verify this.
- Determine what you intend to monitor. Add new templates if required.
- Add the device to the target file
- Stop, then run SICM in test mode to validate
- Start SICM
- View the results after allowing 2 * frequency time to pass
- For further information, refer to: SICM Configuration and Operation Guide
- Determine what snmp OID you intend to monitor. Use an snmp agent on an appropriate device as
well as Getif on Win32 or
net-snmp on Linux to verify the OID.
- Add new or modify a template as required. The template will include, for example:
<metric oid="add_oid_here" data_source_type="GAUGE"
plot="AREA" color="00BB00">add_description_here</metric>
- For testing, add the device to the target file
- Stop, then run SICM in test mode to validate
- Start SICM
- View the results after allowing 2 * frequency time to pass
- For further information, refer to: SICM Configuration and Operation Guide
To send e-mail alerts when a query fails or does not meet a user defined criteria:
- Configure <mail_server> and <mail_from> in the configuration file (config.xml)
- Configure <interface_alert> in the target file (target.xml)
- Optionally configure criteria in <metric_alert> in the target file (target.xml)
Note1: <interface_alert> - is triggered when the target fails to respond to
a (ping, snmp or script) query. Configuration is only required for e-mail alerting.
Note2: <metric_alert> - is triggered when a query result does not meet a
criteria. Configuration is required to set the criteria.
- For further information, refer to: SICM Configuration and Operation Guide
Question:
Is it possible to have the same target mentioned in different menus, using
different templates and/or interfaces?
Example:
- Switch1 int 1 in the menu RouterConnections
- Switch1 int 49 and 50 in the menu FiberConnections
- Switch1 int 10 till 20 in the menu ServerConnections
Answer:
This can be achieved by setting 3 'target_id's for the one target. The ip_address
and community are the same for all 3 'target_id's:
- target_id = "Switch1r" for Switch1 RouterConnections,
- target_id = "Switch1f" for Switch1 FibreConnections and
- target_id = "Switch1s" for Switch1 ServerConnections
For this, the targets file would be:
<target target_id="Switch1r" ...>
<target_desc>switch1 RouterConnections</target_desc>
<menu>RouterConnections</menu>
<template ...<interface>
<interface_id>1</interface_id>
... </interface>
</template>
</target>
<target target_id="Switch1f" ...>
<target_desc>switch1 FibreConnections</target_desc>
<menu>FiberConnections</menu>
<template ...<interface>
<interface_id>49</interface_id>
... </interface>
<interface>
<interface_id>50</interface_id>
... </interface>
</template>
</target>
<target target_id="Switch1s" ...>
<target_desc>switch1 ServerConnections</target_desc>
<menu>ServerConnections</menu>
<template ...<interface>
<interface_id>10</interface_id>
... </interface>
<interface>
<interface_id>20</interface_id>
... </interface>
</template>
</target>
Question:
I need to monitor frame relay services.
Can I monitor interfaces AND subinterfaces in SICM?
Here is an example of a snmpwalk from a cisco router:
- 1.3.6.1.2.1.10.32.2.1.4.4.85 : Counter: 0
- 1.3.6.1.2.1.10.32.2.1.4.4.90 : Counter: 0
- 1.3.6.1.2.1.10.32.2.1.4.6.101 : Counter: 0
- 1.3.6.1.2.1.10.32.2.1.4.6.105 : Counter: 0
Answer:
SICM can monitor interfaces and subinterfaces.
Simply add the interface and subinterface into the <interface_id> in the targets file.
Example:
<interface_id>4.85</interface_id>
Refer to the template "framerelay_errors" in the templates.xml file where a RFC1315 Frame Relay FECN/BECM
template is configured.
Question:
Is it possible to have the website aligned in the center?
Answer:
SICM can be aligned in the browser to either left or center
Question:
The autorefresh doesn't seem to work correctly (I need to press F5 to get a refresh). Is there something
missing in my setup?
Answer:
To stop IIS caching the rrd images (allows the images in browsers to refresh
each 5 minutes)
- Goto Internet Information Services using the MMC: Administrative Tools >> Internet
Services Manager
- In the 'Default Web Site', right click the 'sicm' folder >> Properties >> HTTP Headers
- Select 'Enable Content Expiration' and 'Expire Immediately'
Note: To date, this issue has only been observed in IIS
To send e-mail alerts to multiple recipients, separate each e-mail
address in targets.xml by a ; or , (no spaces)
Example: <interface_alert>2:sicm@example.com;second_sicm@example.com</interface_alert>
Question:
Each time a target restarts the SICM graphs show a value beyond that valid, eg: iis_web_hits returns
600+ million hits/min when IIS is restarted. Is there anyway to ignore the invalid values?
Answer:
If <input_min> and/or <input_max> are defined in the targets file, any value outside
the defined range will be ignored.
- For further information, refer to: SICM Configuration and Operation Guide
Question:
Can I run multiple instances of SICM?
Answer:
Yes, multiple instances can be run. To configure this:
- Install 2 (or more) instances of SICM in differing directories, eg: /sicm1 and
/sicm2
- Ensure each <web_dir> is unique.
- Ensure each <targetfile> is unique.
- Ensure each <templatefile> is unique.
- Ensure each <alert_file> is unique.
- Ensure each <log_file> is unique.
- If running as a service on win32, ensure the service names are unique, eg: sicm1
and sicm2
|