Server - Custom Checks

Here you will find detailed information about Custom Checks for Server monitored objects.

Custom checks are chosen, adapted and deployed by you for one or more monitored objects and they are designed to allow quick and easy fulfilling of business monitoring requirements. Avantra custom checks are a mix of no-code, low-code and pro-code extensions to the Avantra AIOps platform. Each monitored object type has a number of custom check capabilities.

All custom checks within Avantra have a number of standard or common attributes. For more information please review our page on Custom Checks.

DISK_IO

Checks IO on given file systems for the amount of IO reads/writes and IO throughput per second.

Description

Use this custom check to monitor disk IO values on given file systems.

Reference Data
Managed Object

Dynamic Group or Static Group of Servers

Depends on

AGENTALIVE

Configuration
Time window to measure

Define the amount of time used to calculate the I/O throughput average values. The higher the time you specify, the better occasional peaks will be “flattened”, i.e. may less frequently cause Warning or Critical check status.

The default value is set to 10 minutes, which is a reasonable setting. Together with the default cycle time of 5 minutes, the check result is updated every 5 minutes and will always use data of the last two check cycles.

Please note that values that are not multiples of the check cycle are rounded to the nearest multiple of the check cycle value.

File Systems

Define the file systems you want this Custom Check to monitor. You may use * as a file system to make a single entry for all file filesystems. If you make multiple specifications and a file system matches the same, the threshold values are merged together for each file system and the lowest of each will be the effective one.

You may, for instance, define generic thresholds using a file system *. Then you might want to use lower thresholds for a special file system, e.g. /database.

FILESYSTEMS

Monitor individual file systems on a host.

Description

This custom check works exactly the same as the built-in Check FILESYSTEMS but checks only customer-specified file systems. This custom check is intended for users, who want to check certain file systems independently from others, e.gto provide individual notifications to owners of a particular file system. This custom check type is not intended to replace the native FILESYSTEMS check. However, differently from the native check, it can be applied to all types of monitored objects, e.g. SAP Instances and SAP Systems.

Reference Data
Managed Object

Dynamic Group or Static Group of Servers, SAP Instances, SAP Systems, stand-alone Databases

Depends on

AGENTALIVE

Configuration
File System

Specify the file systems to be included. Wildcards * and ? are supported.

Usage

Specify the thresholds, see as well FSUsageWarn and FSUsageWarn. For Unix, you can as well define thresholds for the inodes, see FSNodeWarn and FSNodeCrit.

Nodes

Specify the thresholds for monitoring the inodes on Unix, see as well FSNodeWarn and FSNodeCrit.

FILE_CONTENT

Use the content of a text file as Check Result.

Description

The FILE_CONTENT custom check uses the Avantra Agent to check if a given file exists, is up to date and confirms the contents of the file. If it does not exist, the check status is set to Critical. If it exists the file’s last modification time is verified and can define a Critical threshold. If the file is reasonably new, the content of the file is read. The first line must contain the keywords UNKNOWN, OK, WARNING, CRITICAL, or the corresponding numerical values -1, 0, 1, 2 and this value is returned as check status. The remaining lines of the file form the check message.

Reference Data
Managed Object

Dynamic Group or Static Group of Servers, SAP Instances, SAP Systems, stand-alone Databases

Depends on

AGENTALIVE

Configuration
Check File

Define the absolute path name of the file monitored in this field. File name substitution is provided in case wildcard characters (like * or ?) are used.

For the sake of stability and sanity of the monitoring process wildcards are only allowed to be used in the file name part of the Check File Definition, but not in the directory part. If you want to check for the existence of multiple files in separate directories, you may use the RUN_PROG Custom Check and tools like find (on Unix-like operating systems) or dir /s (on Microsoft Windows operating systems).

Critical Alert if File is older than

Define a Critical threshold.

FILE_EXISTS

Check if a certain file exists.

Description

The Avantra Agent checks if a given file exists. You can define the check status to report in case the file does or does not exist, e.g. an Ok if the file exists and a Critical otherwise.

Reference Data
Managed Object

Dynamic Group or Static Group of Servers, SAP Instances, SAP Systems, stand-alone Databases

Depends on

AGENTALIVE

Configuration
Check File

Define the absolute path name of the file monitored in this field. File name substitution is provided in case wildcard characters (like * or ?) are used.

For the sake of stability and sanity of the monitoring process wildcards are only allowed to be used in the file name part of the Check File Definition, but not in the directory part. If you want to check for the existence of multiple files in separate directories, you may use the RUN_PROG Custom Check and tools like find (on Unix-like operating systems) or dir /s (on Microsoft Windows operating systems).

Check Status if File Exists

Define the check status to be returned if the file exists.

Else

Define the check status to be returned if the file does not exist.

FILE_UPTODATE

Check if a certain file is reasonably new.

Description

The FILE_UPTODATE custom check uses the Avantra Agent to check if a given file both exists and verifies the last modification date. If the file does not exist, the check status is set to Critical. If it exists the file’s last modification time is verified. You can define Warning and Critical thresholds.

Reference Data
Managed Object

Dynamic Group or Static Group of Servers, SAP Instances, SAP Systems, stand-alone Databases

Depends on

AGENTALIVE

Configuration
Check File

Define the absolute path name of the file monitored in this field. File name substitution is provided in case wildcard characters (like * or ?) are used.

For the sake of stability and sanity of the monitoring process wildcards are only allowed to be used in the file name part of the Check File Definition, but not in the directory part. If you want to check for the existence of multiple files in separate directories, you may use the RUN_PROG Custom Check and tools like find (on Unix-like operating systems) or dir /s (on Microsoft Windows operating systems).

Warning Alert if File is older than

Define the Warning threshold.

Critical Alert if File is older than

Define a Critical threshold.

HTTP_RESPONSE

Verify if websites can be accessed.

Description

The HTTP_RESPONSE Custom Check allows you to:

  • verify if a certain website is reachable (i.e. if a certain URL can be accessed and an HTTP 200 OK response is returned),

  • measure web page response times against Warning and Critical thresholds,

  • verify if the retrieved web page contains a certain content.

The HTTP_RESPONSE Custom Check supports the following HTTP/1.1 methods:

HEAD

Performs an HTTP HEAD request, i.e. only the header of the response will be retrieved.

This is usually sufficient if you only want to know if there is any response for a given URL

GET

Performs an HTTP GET request. Basic Authentication is supported as well

POST

Performs an HTTP POST request. HTML Form data can be provided URL-encoded

The HTTP_RESPONSE Custom Check also supports the HTTPS protocol scheme (more exactly: HTTP over TLS).

In case the {avantra-agent} cannot connect to the site, or the site returns a status code different from the 200 OK response, the Check turns to Warning until the number of failed consecutive connection attempts reaches the value defined in _Status turns to CRITICAL after n successive connection failures. Afterwards, the Check turns to Critical.

If the site returns a 200 OK response and there is a value defined in Content must contain string, the {avantra-agent} verifies if the defined string is contained in the HTTP response. If so, the Check returns the check status defined in _Check status if the string is found otherwise the {Check} turns to the value defined in _Check status if the string is not found.

If there are response time thresholds defined in the fields Warning alert if response time exceeds and Critical alert if response time exceeds, the response time of the site is compared to these values and the Check returns a corresponding Ok, Warning, or Critical check status.

You can also check if the SSL certificate for the server is valid and generate a {Warning} or Critical check status if it is about to expire. Of course, this works only if you use HTTPS in the _URL.

Reference Data
Managed Object

Dynamic Group or Static Group of Servers, SAP Instances, SAP Systems, stand-alone Databases

Depends on

AGENTALIVE

Configuration
URL

The URL of the site you want to verify.

Evaluate JavaScript

Set this flag if you want to interpret JavaScript eventually returned with the response.

Only set this flag if you really require JavaScript evaluation. It may be expensive in terms of execution time and memory usage.
Proxy URL

If you need to define a proxy server in order to access the site defined in the URL field you may fill in this value.

User Agent String

By default, the Avantra Agent uses the string agent/[version] in order to identify as a user agent. Some web servers may restrict access to certain browsers, like the Microsoft Internet Explorer or Mozilla Firefox, or they behave differently using one of these browsers.

If you experience problems with pages you can open with your browser, but an HTTP_RESPONSE Custom Check returns e.g. an HTTP Server Error 5xx code, you may try to set a different value for the user agent in this field. For more information about user agent strings, visit this site: http://user-agent-string.info/.

Method

Choose HEAD or GET or POST.

Basic Authentication Credentials

In case you have chosen GET in the Method field you may fill in a user name and a password in this field if Basic Authentication is required by the requested URL.

Post Form Data

In case you have chosen POST in the Method field, you have to fill in the form data in URL-encoded format.

Time Out

You may define a value for the connection timeout. The default is 10 seconds. Please note: This is a timeout for the sub-process spawned to perform the connection attempt. The timeout for the HTTP request itself will always be 80% of the defined value.

Status turns to CRITICAL after n successive connection failures

If filled with a value > 1, all consecutive failed connection attempts will return a Warning check status until the threshold is exceeded. The Check turns to Critical afterwards.

Content must contain string

You can filter the response for one or multiple strings by filling in data in this field.

If you fill in multiple strings (separated by blanks), they are treated like combined by a Logical OR, i.e. it is sufficient if one of the strings is found in the result. Therefore, in order to search for a string containing blanks, use the wildcard * instead of the blank. The same must be used to require multiple strings (in that order) to be contained in the result.

The HTTP header is also considered part of the result, so this option is also available for the HEAD method, although of limited use.

You can define the {check-status} returned in case the string is or is not found using the fields _Check status if the string is found and Check status if the string is not found.

Check status if the string is not found

If there is a value defined in the Content must contain string and the string is not found in the response, this check status is returned.

Check status if the string is found

If there is a value defined in the Content must contain string and the string is found in the response, this check status is returned.

Warning alert if response time exceeds

Fill in a Warning threshold for the response time.

Critical alert if response time exceeds

Fill in a Critical threshold for the response time.

Warning Status if SSL certificate expires within

Fill in a Warning threshold if the check shall verify SSL certificate expiry time.

Warning Status if SSL certificate expires within

Fill in a Critical threshold if the check shall verify SSL certificate expiry time.

Status if SSL certificate is invalid

Select the check status to return in case an SSL certificate is invalid.

JMX_APP_SERVER

Monitor generic Java application server providing JMX (Java Management Extensions) interface.

Description

The JMX_APP_SERVER Custom Check enables you to monitor generic Java application servers which provide a JMX interface that can be accessed via TCP/IP. Common examples are Apache Tomcat, JBoss, Websphere, GlassFish. Due to its generic nature, this Custom Check requires some configuration, but it can monitor almost any data you like, for example, heap memory usage or threads pool usage. First of all, you need to enable remote access to the JMX interface of your application server. For Apache Tomcat, for example, the easiest way is to add it to the environment.

Example 1. Enabling JMX for Apache Tomcat

Add the line

set "CATALINA_OPTS=-Dcom.sun.management.jmxremote -Dcom.sun.management.jmxremote.port=8123
-Dcom.sun.management.jmxremote.ssl=false
-Dcom.sun.management.jmxremote.authenticate=false"

to Tomcat’s setenv.sh/bat file and restart the application server.

Once you have enabled remote access to JMX you need to get the object name, attributes, etc. you want to monitor. To do this, you can use the jconsole tool (Java Monitoring & Management Console) provided by the SDK. This tool shows you all the MBeans you can access. Please make sure that you use the remote port to access JMX, like the Avantra Agent does, instead of connecting to the local process.

Please note that Avantra support cannot assist you in how to enable JMX on our application server, or which MBeans you should monitor.
Reference Data
Managed Object

Dynamic Group or Static Group of Servers, SAP Instances, SAP Systems, stand-alone Databases

Depends on

AGENTALIVE

Configuration
Host

Hostname or IP address of the JMX application server. If the agent runs on the same host you might use localhost as hostname. Please make sure, that your application server is bound to the hostname or IP address you specify here.

Port

The TCP/IP port of the application server’s JMX remote access.

User

Name of the user in case required by the remote JMX access.

Password

Password corresponding to User in case required by the remote JMX access.

Object Name

Object Name of the MBean to be monitored. We recommend that you copy this value from jconsole and paste it here.

Attribute

The attribute to be monitored of the MBean with Object Name above. Again we recommend that you paste it here.

MBean Attribute Evaluation

The expression defines how to calculate a value to be monitored - based on the attributes or Composite Data attribute values. Supported are basic calculations (+ - / *), but no paratheses.

WARNING Threshold

Specify a Warning condition and a threshold for the MBean Attribute Evaluation

A value that matches the condition will issue a Warning Check Result if the condition is fulfilled and a Critical condition is not fulfilled.

CRITICAL Threshold

Specify a Critical condition and a threshold for the MBean Attribute Evaluation

A value that matches the condition will issue a Critical Check Result.

Message

Specify the custom check result message. If you leave this field blank, a standard message giving the result of the MBean Attribute Evaluation will be displayed.

If you want to display a more human-readable result message, type your desired message and use $1 as a placeholder for the MBean Attribute Evaluation result.

RUN_JS

Run a script and evaluate SAP RFC functions, database queries, OS commands, or Cloud Services.

Description

RUN_JS is a very powerful Custom Check type, which combines several different ways to retrieve data and evaluate the results using JavaScript. RUN_JS can easily access ready-to-use objects, for example, established RFC connections to SAP ABAP systems, established database connections, an OS object to run commands, or make HTTP requests to Cloud Services. This custom check type can be used to generate comprehensive check results, including sortable tables, that can even combine different sources of data retrieval.

run js concept

The RUN_JS custom check is a very powerful tool. Data retrieval by RFC, SQL, OS, or HTTP is executed using the permissions you have provided for the corresponding user. You can gain access to sensitive data if RFC, database, or OS user permissions are not restricted. Use this custom check at your own risk. Avantra shall not be liable for any direct, indirect, special, incidental, or consequential damages arising from the use of this custom check. Nor can we provide any support for designing data retrieval and writing evaluation scripts.

The RUN_JS custom check supports ECMAScript 2019 for Avantra Agents 20.11 and higher and ECMAScript 5.1 for older versions.

Starting with Avantra 20.11.5, this custom check type depends on the global Avantra server setting Security.EnableOSCodeExec.

The check evaluation logic is provided by a (usually small) JavaScript program. This script can access objects which are provided by the Avantra Agent, so the programmer can focus on the check logic and not worry about establishing connections, etc.

To script the evaluation logic, users of this check type need to be familiar with the built-in variables and API to use them, as described in section API Documentation below.

The RUN_JS custom check can be created for any monitored object type, however, certain built-in variables are only available in the context of the corresponding system type, i. e. an RFC connection to an ABAP system is only available in the context of SAP systems and SAP instances.

Apart from the examples provided in the API Documentation below, see also solution document RUN_JS Custom Check Examples, which contains ready-to-use examples for download.

You need a Java Virtual Machine 1.8.0_65 or higher in order to make use of all functions of this Check.
Configuration
Script

The JavaScript program providing the check data retrieval and evaluation logic. See section API Documentation.

Report Execution Time

If this option is enabled, the time to run the script is appended to the check result text. This might be useful to check if the performance of a script is good.

Execute Test Run

Starting with Avantra Version 21.11, you can run tests for newly created RUN_JS checks on the selected Agent directly from the Avantra WebUI. This feature does not require saving or activating the check, i.e. it can be tested before affecting the execution of the check on other systems.

To test your check code:

After modifying your script in the Script field, click the Execute test run button under the check name at the top of the dialogue box. Select one or multiple system to test the check on and click the Test button to run your script directly on those selected systems.

Although this is a testing tool, the script will execute on the systems in the same way it does in a real check execution. If the check has any side effects then these will also happen in test mode.

This feature can be disabled by the Security.RUNJSEnableTestExecution setting in your Avantra Server.

LOG_ADAPTER

Monitor textual log files and the Microsoft Windows EventLog.

Description

The LOG_ADAPTER Custom Check allows you to monitor textual log files as well as Microsoft Windows EventLogs. It is based on pre-defined or self-defined Log Adapters, which provide log entries of arbitrary log sources as an array of standardized sets of key-value pairs. Using those key-value pairs as Macros in conditional Expressions, the LOG_ADAPTER Custom Check allows you to define conditions to trigger Check Results. The LOG_ADAPTER Custom Check supports logs spread over multiple log files as well as cyclic log files.

Reference Data
Managed Object

Dynamic Group or Static Group of Servers, SAP Instances, SAP Systems, stand-alone Databases

Depends on

AGENTALIVE

Configuration
Log Adapter Type

The type of a Log Adapter.

Log Adapter

Select your desired Log Adapter. As soon as you have selected a Log Adapter, the available Macros are loaded and the Expression Editors in various fields are initialized.

A click on the icon provides additional data about the Macros provided by the selected Log Adapter.

Avantra automatically detects the encoding of the file. If it fails, use the Log File Encoding setting to specify the file encoding manually.
Log File Path Pattern

If you have selected File Adapter in Log Adapter Type you have to fill in this value, which indicates the path and file name of the target log files.

The LOG_ADAPTER Custom Check supports logs spread over multiple log files as well as cyclic log files. In order to identify the log files belonging to a scenario, the LOG_ADAPTER Custom Check features a flexible log file path pattern interpreter.

In order to best use the log adapter, you must understand how log files are written by your application.

Typical applications write log files until a specific size is reached and then start writing to a new log file. What happens to the old log file depends on your application and is not standardized. Most applications write log files with one of the following procedures. Use the corresponding values for the Log File Path Pattern.

The current log file where the application writes into it has always the same and unique name. When this file reaches the size limit, the log file is renamed (or moved) and gets a new name. The new name typically contains an ascending number or a timestamp to uniquely identify this file. After that, a new file is created with the same name again and the application writes into the new log file.

Define the Log File Path with the exact name of the log file. Do not use wildcards. Do NOT define the Archived Log File Path. Avantra finds old (or rotated) files with the unique file id (inode).

The current log file where the application writes into has a unique name. When the log file reaches the size limit, a new log file is created with a new unique name and the application writes into the new log file. The old log file is not touched anymore. With this approach, each log file has a unique name which is typically an ascending number or a timestamp. Since the file name is not known in advance, you must use wildcards to identify this file (see below).

Define the Log File Path using wildcards to identify the file. Define the path as exact as possible. If old files have a different name pattern, define also the Archived Log File Path (with wildcards).

Similar to the first. but the log file is not renamed (or moved) but copied to a new file (or new location). A new file is created with the same name and the application starts to write into the new file. When a file is copied it gets a new name and a new file id (inodes).

Define the Log File Path with the name of the file. You can use wildcards if necessary. You must also define Archived Log File Path to find the old file. Use wildcards if necessary. Please make sure that the path defined in Log File Path does not match rotated files and the path defined in Archived Log File Path does not match the current file.

Your application does not have a limit to the log file size. It always writes to the same log file. (If the log file is getting too big, this is another issue that is not covered in LOG_ADAPTER).

Define the Log File Path with the exact name of the log file. You can use wildcards if necessary.

If you do not know exactly how your application writes log files, perform the following:

  • Define the Log File Path with the name of the file. You can use wildcards if necessary.

  • Also, define the Archived Log File Path with the name of the archived file. You can use wildcards if necessary.

You may use the following utilities and shortcuts to define general file patterns:

  • Use the to view and insert available Custom Check Macros to generalize your path pattern for different system environments.

  • Use wildcard characters for single (?) and multiple (\*) characters to match multiple log file names. These wildcard characters are only supported in the file names, not in the directory parts.

  • If you use wildcards, the LOG_ADAPTER Custom Check will determine the latest file matching the wildcards in every read operation. It will only monitor this single file. If your wildcard definition matches multiple files that are not written in chronological order, the LOG_ADAPTER Custom Check may read data from different files over time, depending on which file was updated last. The result is most likely not what you expect, and the check will return a Warning`Context lost`Check Result from time to time.

  • Use the slash key (/) instead of backslash (\) as file separator on Microsoft Windows and Unix operating systems for platform-independent path definitions.

Archived Log File Path

See Log File Path Pattern.

Log Adapter Processing

This setting defines which entries of the log files are to be processed. Log entries written since last check cycle processes all log entries that have been added after the last read of the file. Log entries written in last x minutes reads all entries that have been added during the past Log Entry Timespan minutes.

Using Log entries written in last x minutes requires that you have defined a Log Adapter that contains timestamps in a way that the Avantra Agent is able to read and understand. If your log file does not provide reasonable timestamp information, choose Log entries written since last check cycle instead.

This parameter, together with the execution cycle, has a direct impact on the duration a LOG_ADAPTER Custom Check alert is visible in the RealTime Monitoring. The time you indicate in Check Result is visible in the RealTime Monitoring.

When you select Log entries written since last check cycle, the alert will be visible for Log entries written since last check cycle * execution cycle minutes.

Log Entry Timespan

Fill in the period of time the {avantra-agent} should look back while processing the log file if Log entries written in last x minutes is selected in _Log Adapter Processing>.

Warning Condition

Define a conditional expression that indicates log entries considered as Warning.

Critical Condition

Define a conditional expression that indicates log entries considered as Critical.

A Critical match will override Warning matches in terms of the check status. The check message will also be overridden unless you select All matching log entries in Create Message For.

Match Counter Threshold

In the Warning Trigger section, this value defines how many entries must be matched to the condition before the check status will be set as Warning. I.e. if there are only 2 entries matching and this value is set to 3, then the check status will be OK. If there are 3 or more entries matching, then the check status will be set as Warning.

In the Critical Trigger section, this value defines how many entries must be matched to the condition before the check status will be set as Critical. I.e. if there are only 2 entries matching and this value is set to 3, then the check status will be OK. If there are 3 or more entries matching, then the check status will be set as Critical.

Resolution Condition

Define a conditional expression that indicates log entries that resolve issues found using Warning Condition or Critical Condition.

Create Message For

You can specify the format of the check message shown in RealTime Monitoring or used in Notifications.

The message format has a static header that indicates which and how many of the trigger conditions have matched, followed by optional information on the matching log entries.

You may choose between the listed formats. For the detailed layout of the display of log entries, see also Message Format.

No log entry

No details of matched log entries are shown.

First matching log entry

Details of the first matching log entry are shown.

Last matching log entry

Details of the last matching log entry are shown.

All matching log entries

Details of all matching log entries are shown.

Message Format

Define the display of a matched log entry in the check message. You may enter arbitrary text tokens in the field, and you may use all Macros available for the selected Log Adapter (using the icon) to build a meaningful description text.

NETWORK_IO

Checks IO on given network interfaces

Description

The NETWORK_IO Custom Check allows you to monitor network IO on given network interfaces for send and receive throughput, errors, utilization. You can specify warning and critical thresholds for all monitored IO types.

Reference Data
Managed Object

Dynamic Group or Static Group of Servers

Depends on

AGENTALIVE

Configuration
Time window to measure

Define the amount of time used to calculate the I/O throughput average values. The higher the time you specify, the better occasional peaks will be "flattened", i.e. may less frequently cause Warning or Critical check status.

The default value is set to 10 minutes, which is a reasonable setting. Together with the default cycle time of 5 minutes, the check result is updated every 5 minutes and will always use data of the last two check cycles. Please note that values which are not multiples of the check cycle are rounded to the nearest multiple of the check cycle value.

Network Interfaces

Define the network interfaces you want to monitor. You may use * as network interface name to make a single entry for all network interfaces. * is a good value for an initial setup, because all network interfaces are then displayed in the check results table including description (if available) and IP address.

If you make multiple specifications and an interface matches the same, the threshold values are merged together for each interface and the lowest of each will be the effective one.

You may, for instance, define generic thresholds using a network interface *. Then you might want to use lower thresholds for a special interface, e.g. eth4.

NW_RESPONSE

Monitor availability and response time of remote hosts.

Description

The NW_RESPONSE Custom Check is designed to monitor the availability and response time of a remote host. It sends a configurable number of ICMP echo request packets to a given host, receives the ICMP echo replies, and calculates packet loss as well as packet round-trip-time, i.e. the check virtually does what the ping operating system command does.

You have the option to define packet loss thresholds; if there is no packet loss, the network round-trip-time can be compared to corresponding response time thresholds in order to generate Warning or Critical Check Results.

It is characteristic of the ICMP protocol to require administrative permissions in order to generate ICMP echo requests. On Microsoft Windows operating systems the permissions the Avantra Agent runs with (as a service) are usually sufficient. On Unix-like operating systems the Avantra Agent would need to run as user root to execute the NW_RESPONSE Custom Check. This is generally not recommended. If you want to use this feature, it is recommended to run it on a Microsoft Windows host.

Reference Data
Managed Object

Dynamic Group or Static Group of Servers, SAP Instances, SAP Systems, stand-alone Databases

Depends on

AGENTALIVE

Configuration
Host Name or IP Address

Fill in the host name or the IP address of the remote host.

Packet Loss Warning Threshold

Define a Warning threshold for packet loss.

Packet Loss Critical Threshold

Define a Critical threshold for packet loss.

Average RTT Warning Threshold

Define a Warning threshold for the average round-trip time.

Average RTT Warning Threshold

Define a Critical threshold for the average round-trip time.

Timeout

Define a time out for the Check. The check status will be Critical if the time out is exceeded.

Number of Packets

Define the number of ICMP packets to send per Check

Packet Size

Define the size of ICMP echo request packets

PROCESS

Watch if processes are running.

Description

This Custom Check can monitor whether or not a configurable number of processes sharing the same process name are running.

The Check is available on Unix-like operating systems and on Microsoft Windows operating systems. However, it might more useful on a Unix host.

You can define minimum and maximum Warning and Critical thresholds for the number of instances of the process in question.

Reference Data
Managed Object

Dynamic Group or Static Group of Servers, SAP Instances, SAP Systems, stand-alone Databases

Depends on

AGENTALIVE

Configuration
Process Name

Enter the name of the process you want to monitor.

  • On Unix-like operating systems, fill in the process’ simple command name, i.e. the process’ executable name without any arguments. The simple command name is the value displayed by the -o comm option in the POSIX version of the ps command. See also man ps on your system.

  • Also, on Unix-like operating systems, if a simple command name is not sufficient (like in Java applications), you can also define a name to match (a substring of) the process’ command line call. The command line call is the value displayed by the -o args option in the POSIX version of the ps command. See also man ps on your system. Please note that this value may be limited to 256 characters on some systems.

    If you prefer this option, make sure you set the flag Match process name in process table…​.

  • On Microsoft Windows operating systems enter the process name (as shown in the Processes tab of the Task Manager), but without the trailing .exe.

Critical alert if number of processes is lower than

If the number of running processes falls below this threshold, a Critical check status is returned.

Warning alert if number of processes is lower than

If the number of running processes falls below this value, but does not exceed the value defined in Critical alert if number of processes is lower than, a Warning check status is returned.

If the field is left blank, no Warning check status is returned at all.

If the value in this field is less than the one defined in Critical alert if number of processes is lower than, the latter value will be ignored. This method can be used to receive only Warning but no Critical Check Results.

Warning alert if number of processes is higher than

If the number of running processes exceeds this threshold, but is still below the value defined in Critical alert if number of processes is higher than, a Warning check status is returned.

If the field is left blank, no Warning Check Results are returned at all.

If the value in this field is higher than the one defined in Critical alert if number of processes is higher than, the latter value will be ignored. This method can be used to receive only Warning but no Critical Check Results.

Critical alert if number of processes is higher than

If the number of running processes exceeds this threshold, a Critical check status is returned. If the field is left blank, no Critical Check Results will be returned.

RUN_PROG

Integrate your own monitoring scripts or programs into Avantra

Description

The RUN_PROG Custom Check allows you to integrate your own monitoring scripts or programs into Avantra. When you define a RUN_PROG Custom Check, you can upload your script or program to the Avantra Database. It is transferred to the Avantra Agent together with the agent configuration. The Avantra Agent executes the program during the Basic Check Cycle, the exit state of the program is considered as check status, while the program output is used as the check message. You can also configure the RUN_PROG Custom Check to find and collect performance data in the program output.

There are two options for this Custom Check: you can either simply execute a script or program that is already installed on the target System (more exactly: on the hosting Server), such as an operating system command. Or, what you probably want in most cases, is to store the script/program in the Avantra Database and deploy and run it on the target Systems. In the latter case, the program is transferred to the Avantra Agent, saved to the local filesystem, and executed from there. It is also verified prior to each execution that the script/program on the local filesystem has not been altered. Otherwise, it will be replaced by the version stored in the Avantra Database.

Starting with Avantra 20.11.5, this custom check type depends on the global Avantra server setting Security.EnableOSCodeExec.

Once the Custom Check is deleted, the program is removed from the filesystem (if Delete program after check deletion is set). Before we start describing how to set up and use the RUN_PROG Custom Check, we would like to highlight some important facts:

  • Scripts/programs must not be too large. 50 to 60 KB (compressed size) will be ok (which in fact can be a very comprehensive script).

    Please note that the program is attached to the System's configuration, so it is transferred over the network whenever the configuration changes. If you desire to execute larger programs, you can deploy the programs yourself to the target Systems and configure this Check accordingly.

  • Only single scripts/programs can be transferred, there is no pack/unpack mechanism for multiple scripts/programs. You may work around this in the same way as described in the previous item.

  • It is recommended to avoid blanks in the Custom Check name as well as in the script/program name. While the Avantra Agent does its best to cope with blanks, you need to provide proper quoting in the command line definition. Avoiding blanks makes the definition easier and more robust. (The Custom Check name by default will be used as the name of the directory the script/program is stored into.)

  • The script/program is executed in the security context of the Avantra Agent, i.e. on Unix-like operating systems, it is running with the permissions of the user running the Avantra Agent, on Microsoft Windows operating systems it is running as LocalSystem user, unless you have changed the service’s Log on properties.

  • Please observe the following coding guidelines for scripts/programs:

    1. Keep the output message short. One or two lines (80 characters each) shall suffice.

    2. Return a proper exit state: 0 will create an Ok Check Result, 1 a Warning, and 2 or higher a Critical Check Result.

      If you want the Check to return an Unknown check status, use the same exit code you define in Exit code for Unknown check status. You may use the Unknown return code if the program has not been called properly, or if your program is not able to determine the check result.

    3. If you execute system commands, always use the full path, or use an internal PATH variable in your scripts in order to be sure all commands can be executed! This is especially important if scripts are deployed on several different platforms.

    4. If you use temporary files, make sure you clean them up properly!

    5. The script/program typically should not run longer than 10 seconds. It is recommended that the script/program itself takes care to exit properly if a runtime threshold is exceeded. Otherwise, Avantra Agent will stop the program after the number of seconds defined in the Timeout field. Using timeouts is important especially when the script/program accesses network resources.

    6. Keep your scripts/programs small! Small programs are transferred faster to the Avantra Agent and probably execute more quickly! If you need to execute longer running programs, consider using the FILE_CONTENT Custom Check.

Once you have deployed the script or program you have also the option to remove it again. Use the Delete Program From Database button.

Reference Data
Managed Object

Dynamic Group or Static Group of Servers, SAP Instances, SAP Systems, stand-alone Databases

Depends on

AGENTALIVE

Configuration
Command line call

Define the full command line call to be executed by the Avantra Agent. It is recommended to use absolute path names whenever possible.

Example 2. Example for Run Program Custom Check Command Line Call Definition
/bin/sh "%%PROG_DIR%%/proggy.sh" -param 1 -param 2

on Unix-like operating systems or

"C:\Program Files\system32\cmd.exe" "%%PROG_DIR%%\proggy.cmd" /param 1 /param 2

on Microsoft Windows operating systems.

You have to use quotes whenever a blank is contained in the program or directory name. Please note that also the resolved macros may contain blanks, so it is recommended to use quotes around macros as well.

You may use newlines in this field for ease of reading and writing. Note that they are stripped (not replaced by blanks) prior to execution.

Working directory

Define a working directory of the executed script or program.

Environment variables

Define environment variables for the executed script or program. Please note that multiple entries must be separated by commas or semicolons. The format is name ‘=' value { (';' | ',’) name '=’ value}

Example 1:

timeout=300;mode=test

Example 2:

file=/path/to/your/file;url=http://host/path

Time out

Define a timeout for the executed script or program.

There is a hard limit for this value defined by the CheckCycleTime Monitoring Parameter. In other words, the effective time out is the minimum of the value defined in this field and the CheckCycleTime Monitoring Parameter.

Exit code for unknown check status

Use this field to define the exit code of your program that shall indicate an Unknown check status.

The default value, which is -1 for historical reasons, is not supported on Unix-like operating systems, so you may want to adapt this value accordingly if you plan to operate with Unknown check status.

Collect Performance Values

If you want the {custom-check} to collect performance data (which must be written to console output by the program), check _Select Performance Resource Type and choose a previously defined custom performance resource type from the list. See Configuring Custom Performance Data Resource Types on how to define custom performance resource types.

The syntax supported follows the plugin output specification of a common open source monitoring product.

If the program does not return or returns erroneous performance data the Custom Check will report a Warning status, with a message e. g. 'Result is supposed to return performance data but nothing was found' or 'Missing or invalid performance data'.

Return Timeout as

Determine whether to return the check status as Ok, Warning, Critical, Unknown if the timeout specified in the Time out setting is exceeded.

Omit the command line in check result

Check the box to not display the command line in the check result.

Program

Use the Upload Program button in order to select a script or program to deploy.

Original File Name

Read-only field containing the name of the script or program once it is deployed. The name is displayed as a link that allows you to download the script or program.

This field is only available if a script or program has been uploaded.

Checksum

Read-only field containing the MD5http://en.wikipedia.org/wiki/Checksum[checksum] of the script or program once it is deployed. This value is used to detect whether the file copied to the target System's filesystem has not been modified.

This field is only available if a script or program has been uploaded.

Upload Date

Read-only field containing the date the script or program was uploaded.

This field is only available if a script or program has been uploaded.

Program Directory

You may define where the script/program is supposed to reside in the target System's filesystem by filling in the appropriate directory.

If you fill in this value, please note that the user running the Avantra Agent needs read, write, and execute permissions on this directory!

This field is only available if a script or program has been uploaded.

Delete program after check deletion

If you set this flag, the Avantra Agent will take care to remove the script/program once the Custom Check is deleted. The Program Directory is also removed if it is empty.

This field is only available if a script or program has been uploaded.

Service Name to record Availabilities

If you enter a value, the Custom Check will record the Service Availability for use in e.g. Service Level Reports.

Report Warning as Available

If Service Availability is recorded, a Warning check status is usually considered Down. Set this flag if you want the Warning check status to be considered Up instead.

WIN_SERVICE

Monitor if Microsoft Windows Services are running.

Description

The WIN_SERVICE Custom Check allows you to monitor if a certain Service on a Microsoft Windows host is or is not running. This Custom Check has no effect on Unix-like operating systems.

Reference Data
Managed Object

Dynamic Group or Static Group of Servers, SAP Instances, SAP Systems, stand-alone Databases

Depends on

AGENTALIVE

Configuration
Service Name

Define the display name or the technical service name in this field. You can find the technical Service Name in the field Service Name of the General tab in the Properties of the respective service.

Use the Services snap-in of the Microsoft Management Console. The display name is consequently shown in the Display Name field at the same tab. You must not use the executable name.

This field may contain Custom Check Macros.

Status If Service Is Running

Select the check status in case the service is running.

Else

Select the check status in case the service exists but is not running.

Status if service does not exist

Select the check status in case the service does not exist.