The Network Connectivity Monitoring (net_connect) probe monitors the following parameters of remote hosts:
Host connectivity using
(ICMP ECHO) commands
Packet loss, jitter, and latency for the connection to the hosts
TCP connections to a list of user-defined services
This section describes the history of the revisions for net_connect probe.
Support case(s) may not be viewable to all customers.
Enhanced profiles that enable you to configure metrics, baselines, alarm thresholds, alarms - including Time Over Threshold alarms - and custom alarm and close alarm messages, all within a single MCS profile. For more information, see Configuring Alarm Thresholds in MCS
(October 2020) New MCS template v3.43 with:
Added "Check Interval" in Network Connectivity (Enhanced) template. Specifies the time interval that the probe waits before it executes the profile.
Added "Packets to send" in Network Connectivity (Enhanced) template. Specifies the number of packets to send on each interval.
Added "Delay between Packets (in ms)" in Network Connectivity (Enhanced) template. Specifies the delay (in milliseconds) between successive packets.
Added "QoS Source” in Network Connectivity (Enhanced) template. Allows you to select a source that is included in the QoS message.
If you want to create fresh MCS profiles on this version. You must have at least net_connect 3.37 package in your archive or net_connect 3.37 must be deployed on robot.
To update 'QoS Interval' or 'Check Interval' in Network Connectivity templates, you must adjust the values for 'Packets to send' and 'Delay between Packets (in ms)' fields. These fields must be updated in the Network Connectivity (Enhanced) template as well. Probe sends the number of packets with delay that is specified within the interval. For the short intervals, large number of packets to send with delay might cause issues. Also you may get irrelevant values for metrics, as probe might not be able to process all the packets within the specified time interval.
Template Migration support:
net_connect Enhanced MCS template 3.42 to net_connect Enhanced MCS template 3.43
Fixed an issue where net_connect probe MCS template does not have individual alarms that is based on the Target.
Support case: 32054133
(January 2020) The net_connect MCS (version 3.42) has been updated with the following features -
Introduced the multiple profile support for Network Connectivity and Service Monitor Enhanced template.
Template Migration from 3.40 (Enhanced) to 3.42 (Enhanced)
Template Migration from 1.27 (Legacy) to 3.40 (Enhanced)
Fixed the issue where the Network Connectivity (Enhanced) MCS template does not allow custom profile names, support case number 20072782.
Fixed the issue where the Service Monitor MCS template (Enhanced) gets deleted at times in the SSRV2PackageTemplate table, support case number 20157970.
(August 2019) The net_connect MCS (version 3.40) has been updated to add support for monitoring custom ports in net_connect. Service Monitor (Enhanced) child template has been added under the Network Connectivity (Enhanced) profile to monitor custom service ports. Enhanced profiles that enable you to configure metrics, baselines, alarm thresholds, alarms - including Time Over Threshold alarms - and custom alarm and close alarm messages, all within a single MCS profile.
MCS Network Connectivity (Enhanced) profile yields connection host to (protocol) failed. Support case 01366510
Network Connectivity (Enhanced) Template needs Monitoring custom ports. Support case 20012352
Network Connectivity (Enhanced) template (net_connect) doesn't set active/alarm. Support case 20019610
(April 2019) The net_connect MCS (version 3.38 onwards) has been updated to address the common vulnerabilities and exposures by updating the jackson-databind libraries. For more information and CVE numbers, see Addressing Jackson Vulnerabilities
Fixed an issue where generic qos (like jitter, latency, etc), were generated for services which had the same source, target, and device ID.
Fixed an issue where the net_connect MCS template failed to deploy successfully.
Support case 01165106
Fixed an issue where Oracle services (port check) were not available in the MCS.
Support case 01098084
Updated this probe as part of removing dependency on the end-of-life (EOL) Microsoft Visual C++ Redistributable in CA UIM 9.0.2. CA UIM 9.0.2 now uses Microsoft Visual C++ Redistributable for Visual Studio 2017. If you are installing this probe version on machines running Windows operating system, we recommend you to download the appropriate Microsoft Visual C++ Redistributable for Visual Studio 2017 package from the Nimsoft archive (x86/x64) to stop the machine from auto-restarting:
When deploying a probe that is dependent on the Microsoft Visual C++ Redistributable for Visual Studio 2017 package version 1.0 (vs2017_vcredist_x86 1.0 and vs2017_vcredist_x64 1.0), Windows operating system restart occurs on the robot where the probe is deployed. In the VS2017 package version 1.0, there is no "norestart" option specified in the Post Install command. Therefore, the Post Install might trigger OS reboot as part of the package installation. For a workaround, see the KB Article kb000120656
Updated this probe as part of removing known security vulnerabilities in CA UIM 9.0.2 by using the upgraded OpenSSL components.
(February 2019) Added support for monitoring Windows 2019.
The probe now updates the log file to indicate that an existing profile in the configuration file has no keys, and is invalid. The probe skips such profiles and continues to monitor other profiles.
Support case number 732132
Unable to ping an IPv6 address.
Unable to update a profile when it is searched using the
Find host in list of Profile
Support case number 00644688
Added the ability to clear alarms when you edit or delete profiles created on this and later versions of the probe. For more information, see
Support case number 434814
When using the Ping Sweep feature, the probe did not generate the total number of IP addresses as specified on the UI.
Support case number 408147
The probe incorrectly changed the QoS source when you enabled or disabled the Dynamic IP Monitoring feature. For more information, see
Support case number 360406
The probe did not generate QoS messages and alarms for Packet Loss monitoring when you override the global QoS interval in your profile. For more information, see the AC or IM Configuration articles.
Support case number 322915
Added support to use the robot name as the identification method in alarms.
The probe did not retrieve the
of the specified
of the monitored host.
Support case number 00267015
The probe did not save the changes in Infrastructure Manager when managing services through Bulk Configuration.
Support case number
Set the default state of the
monitor to ON to display data on the Unified Dashboard automatically.
Added support to monitor hosts with dynamic IP address.
When viewing data on USM, the net_connect probe DEV files conflicted with the interface_traffic probe DEV files and displayed random device information. If both the hostname and IP address are provided for a monitored host, the net_connect probe used only the hostname to display data.
Salesforce cases 00165874, 70007073
You can create a key in the probe to display data on USM, using the host IP address. For more information, see the Known Issues and Workarounds
When monitoring only the services of a host, and not the ICMP connectivity of a host, the probe did not generate any QoS data.
Salesforce case 70002909, 70004940
The probe did not resolve the $group variable if it was used in a tree structure.
Salesforce case 70002994
The probe was deleting the monitoring profiles created using the drag-and-drop feature.
Salesforce case 70002993
The probe can now be migrated to standard static alarm thresholds using the threshold_migrator probe.
Threshold Configuration Migration
section for more information.
User was not able to set the default host parameters in the Infrastructure Manager version of the probe.
Salesforce case 00169100
Added three new fields (Max Ping Threads, Max Service Threads, and Max PacketLoss Threads) under Setup Properties > Advanced > Performance Properties section in IM GUI, and under net_connect > General Configuration section in AC GUI to provide flexibility to the user in configuring separate number of threads for ping, services, and packet loss feature.
Probe will now send single ping for connectivity check and response time calculation instead of separate pings for both.
Added Log Size field to define the size of the log file.
User was not able to monitor the IPv6 host.
Salesforce case 00154639
Multiple alarms were received and the alarm message contained the text: 'jitter is above threshold limit' even when no threshold was defined.
User was not able to change the message for jitter alarms. User can now edit already existing (default) jitter alarms, but cannot create new jitter (OK and Fail) alarms using Message Pool Manager in IM GUI. These two default jitter messages will be sent by the probe for all the profiles. User cannot configure custom jitter alarm using Host or Service properties (The probe does not support this functionality). User can also edit message text of default jitter alarms, which will reflect in the Alarm Console. It is recommended not to delete any default jitter alarm, as these are the only 'jitter OK' and 'jitter Fail' alarms sent by the probe. User will not be able to add these jitter alarms again from probe GUI.
Probe stopped working with services that did not have Active key.
Salesforce case 00141024
After re-applying default settings to a profile, a newly created message did not get saved in the configuration file.
Salesforce case 00146089
Runtime error occurred when a service monitor was edited.
Salesforce case 00148572
After the value of the Failed intervals field was changed once, the value did not get changed if it was changed again.
If a service was added using drag-and-drop function, some default sections were not saved in the configuration file.
Salesforce case 00142415
Fixed a defect where probe was constantly in error state and not generating PID.
Fixed a defect where the QoS Target associated with old profiles was not changing to Profile Name if the probe version was upgraded.
Fixed a defect where the probe stopped rotating the log file if the log size was set higher than 2 GB.
For an inaccessible service on a host machine, the probe was not delaying the retry attempt for the time specified in the Delay Between Retries field.
User was not able to change the severity of the alarm Failed to execute profiles in scheduled time interva
. This alarm is now configurable through GUI.
The probe failed to restart when the Monitor ICMP connectivity (ping) check box was unchecked and the probe was restarted.
Fixed the issue where the probe was generating the faulty "connection failed" alarms and "NULL" QoS for devices without publicly defined hostname, such as switches. Now, the probe uses the IP address for devices that do not have a publicly defined hostname and generate appropriate alarms and QoS.
Fixed an issue where jitter and packet-loss alarms were not clearing.
Fixed an issue where the probe sends 0 as the response time for all pings instead of the actual response time.
Fixed an issue where the net_connect probe sends QOS value 0 for devices that do not ping.
Fixed an issue where the net_connect probe sends duplicate QOS when packet-loss monitoring is enabled.
Fixed issue where Jitter and Packet Loss alarms not clearing.
Fixed issue where net_connect probe sends 0 as the value for response time for all ping instead of the actual response time.
Fixed issue where net_connect probe sends QOS value 0 for non pingable device.
Fixed issue where net_connect probe sends duplicate QOS when packet loss monitoring is enabled.
Added support for cfg having large (25000) profiles.
Reduced time of load for large profiles.
Fixed Metric ID for alarm
Fixed an issue of CPU usage and clock drift (QOS with negative response time).
Added entries in CiManager for unit of Packet latency and Packet Jitter.
Fixed an issue where $contactInfo value is missing
Fixed the timeout issue. Default timeout changed from 10 to 2 sec.
Fixed an issue where in parent-child relationship probe is processing the child even when the parent node is down.
Fixed an issue where a configured service on a host was not running.
Fixed ping fail alerts.
Fixed fail to schedule alerts.
Fixed: Probe GUI crashes when user close in between abort process in ping sweep.
Fixed: Child Profiles not getting deleted from right pane.
Revamped probe for scalability
The issue of failing to execute in scheduled time interval has been fixed.
The higher CPU issue fixed.
Revamped the probe for scalability for False Alarms, that is probe reporting +ve ping as -ve ping due to time out at higher number of profiles.
Revamped the probe scalability for failing to execute profile at scheduled time interval.
Revamped the probe scalability and resolved some issues relating to crash of the probe.
Also changed the config parameters default value, that is max_socket = 100 (global level) and timeout = 10 Sec. (at profile level).
Fixed High CPU Utilization issue.
Fixed a probe crash on restart and empty callbacks call from probe utility.
Fixed an issue with the timer functionality.
Fixed an issue with QOS messages not coming for the profile.
Fixed an issue for multiple Alarms coming for a profile.
Fixed an issue to sort profile on basis of IP address. Supported sorting for both IPv4 and IPv6 address.
Added new callback to return active profiles and their services.
Wild card or regex is supported in profile name.
Thread Model change for Profile execution.
Fixed an issue where change in QoS source for one profile also changes other profiles.
Resolved a restart problem.
Fixed issue when probe is not using packet size setup in ping functionality.
Implemented Source Override when even ping is disabled.
Fixed crash issue when profile is active and monitor icmp check box is unchecked.
Implementation of IPv6 compatibility.
Implementation of IPv6 compatibility.
Fixed an issue where the probe was reporting incorrect jitter values in alarm and QoS.
Fixed a crash occurring due to insufficient wait time used by the probe to wait on profile & service threads graceful exit.
The probe now supports a new raw configurable attribute named thread_timeout for setting the thread timeout value in seconds.
The probe looks for the new attribute under setup section of configuration file. By default the probe uses 30 as timeout value to wait on the threads to gracefully exit on probe stop/restart situations.
Added option for changing QoS source.
Do not change the source unless you have good reasons. The Robot address is the "correct" source for this probe.
Added support for configuring packet loss alarms & QoS separately.
Added packet jitter value to the values PDS sent with alarms.
The key name for this value in the alarm is packet_jitter.
Added support for expanding group value for chained host profiles.
Added an alarm when profile execution exceeds regular profile execution interval.
Fixed the SOC xml Junit test issue.
Fix: GUI bug when using bulk configurator to configure services. If text field 'Delay between retries' is empty, configuration of services would silently fail.
Changed default ICMP packet size to 32 bytes, default delay between packets to 30 seconds, default packets to send in packet loss to 10.
Added alarm on jitter.
Added option in UI to configure service threads.
Updated bulk configuration UI for missing configuration parameters.
Texts in Service-oriented configuration (SOC) now being presented properly.
Added option to continue checking services, even if "Monitor ICMP connectivity" fails.
Added option to send QoS for Latency and Jitter with packet loss monitoring, when delay between packet is set.
Added new check boxes for Latency and Jitter in SOC GUI.
Added fixes for web-based Service Oriented Configuration.
Added functionality to copy the Result of the show status window to clipboard.
Added support for internationalization.
Added support for reading alarm tokens from cfg.
Support for Web-based Service Oriented Configuration (SOC).
Initial support for internationalized alarm texts.
Changed thread pooling implementation to schedule all the profiles from queue.
Added minor bug fix in GUI.
Probe version released with NMS version 4.30.
Added support for localized alarm messages.
Added a log message on negative response time. Also stopped sending NULL QoS when negative response time is calculated due to host time issues.
Added fix to provide number of retries before sending alarm with delay between retries.
Added fix to properly clear service alarms.
Redesigned bulk configuration window.
Added option in GUI to set default service parameters.
Added fix to remove error message in logs in ssh service by sending SSH identification string.
Added fix to use new API when both IP and hostname are available to avoid name-lookup.
Added fix in probe to send proper hostname in callback.
Added fix in GUI to update hostname and ip address on test.
Added fix in probe to break the loops on restart/stop.
Added fix in probe to initialize the buffers, before checking the response while monitoring the services.
Added fix in probe to change "Bind to network interface" on restart.
Added fix for crash in GUI when profile was renamed and deleted.
Commented code for group expansion.
Added code to add contact info field in bulk configuration.
Added a field for delay between packets in profile form.
Fixed constant QoS value.
Added a fix to not clear the alarm in every interval before sending the challenge response failure alarm.
Fixed memory leak (introduced in version 2.23).
Added a fix for intermittent probe restart when running a service request.
Updated callback to test the response using IP address when both hostname and IP is specified.
Added a fix to resolve correct IP address when hostname is specified in profile.
Fix for not always using correct IP address when hostname is specified in profile. (Fix available in version 2.15 and from version 2.23, not versions in between).
Updated to latest NIS database layout.
Fixed problem with profiles that never got rescheduled.
Added a fix to schedule the profiles which failed to execute in previous run.
Added a fix for crash on restart or while stopping the probe.
Added fix to stop proper timers.
Added fix to schedule all the profiles from queue.
Fixed problems with threadpool and multi-core cpu.
Fixed problems with "blank" challenge response input.
Fix: If a connection test fails, the QoS for packet loss is NULL not 100(%).
Added support for the following platforms:
Fixed problem with the challenge response timeout not being read. Note the following issue:
The Timeout field on the service properties dialog requires a value. If empty, the GUI will crash!
Fixed problem with abnormal CPU usage of Linux.
Performance improvements done to handle 5000+ profiles.
Fix bulk configurator to accept less than 1 second timeout on service alarm and host ping timeout.
Update supported platform list.
Version 2.03 of the net_connect probe requires version >=3.x of the robot to run on the Windows platform.
Port challenge response feature added.
Ability to paste into the IP Address field added.
Contact info field added to host properties, to be used in alarm messages.
ICMP packet loss feature added.
Added ability to add Services in the Bulk configurator.
Added ability to enter IP addresses with leading zeros.
Added search button to find IP or Name.
The Port scan windows can be resized.
Bug fixes: default value of ip address 0.0.0.0 removed.
Update UNIX platform list.
Added payload data to ICMP packet.
Fix: GUI could cause a runtime error if you entered an integer value as name for a new profile.
Fix: ping test response times should be correct now
Fix: drag-n-drop bug that would allow you to drag a host from GUI and drop it in another program, it would disappear from net_connect GUI
Fix: IP-addresses should be easier to edit.
Feature: Bulk configuration tool.
Feature: "Group" variable can be used in alarm messages.
Feature: Ability to generate alert when ping latency threshold exceeds a customizable limit.
Feature: When adding a profile with a name that already exists, program will suggest a new name.
Feature: The new host dialog has improved code to detect and suggest parent group/folder.
Feature: Simple port scan utility.
Added more standard port numbers to probe definition.
Fixed name-resolution problem in UI.
Added ping sweep functionality.
Added support for overriding the global QoS setting per profile.
Fixed minor UI issues.
Added support for profile groups in UI.
Threshold Configuration Migration
The probe (version 3.11 or later) threshold configurations can be migrated
to standard static alarm thresholds using the threshold_migrator 2.10 version or later with CA UIM 8.31 or later.
The changes in the probe after migration are:
The Infrastructure Manager GUI of the probe will not be available and the probe will only be configured on Admin Console.
Probe specific alarm configurations in the probe monitors will be replaced by Static Alarm and Time Over Threshold configurations.
The alarms will be sent by the
The QOS_NET_CONNECT, which monitors the ping response, will not be migrated
Probe Specific Hardware Requirements
The net_connect probe must be installed on systems with the following minimum resources:
Memory: 2-4GB of RAM. The OOB configuration of the probe requires 256 MB of RAM
CPU: 3GHz dual-core processor, 32-bit, or 64-bit
Probe Specific Software Requirements
The net_connect probe requires the following software environment:
CA Unified Infrastructure Management (CA UIM) 8.1 or later
Robot 7.62 or later (recommended)
Java JRE 6 or later (required for Admin Console)
The net_connect probe requires the following software environment to migrate with threshold_migrator probe:
CA Unified Infrastructure Management 8.31 or later
Robot 7.8 or later (recommended)
Java JRE version 7 or later
Probe Provisioning Manager (PPM) probe version 3.21 or later
Baseline Engine (baseline_engine) version 2.60 or later
Consider the following points when you upgrade to version 3.31:
When you edit or delete a profile created on versions earlier than 3.31, the probe does not clear any previously generated alarms.
The probe changes the QoS Source from
of the profiles where Dynamic IP Monitoring feature is enabled. Navigate to each profile to revert it to Profile Name.
Known Issues and Workarounds
The probe has the following known issues:
The integration feature with Enterprise Console is no longer applicable as Enterprise Console is a legacy CA product and is not supported.
In the IPv6 environment,
You can only monitor remote AIX systems.
Probe does not accept the hostname using the ping sweep feature.
To display data on USM using the host IP address, create a key,
in the probe Raw Configuration under
section. Set the key value to
and the probe creates DEV files based on the host IP address.