The TCP monitor tracks and records network-related TCP performance parameters. This monitor tracks socket-related data and helps pinpoint TCP-related problems.
This monitor uses a third party package for SNMP. To collect data from this monitor, run the SNMP agent on the system to monitor.
Listed below are some advantages of this monitor over other TCP monitors:
- This monitor can monitor socket-related data of remote systems.
- ProactiveNet Agent installation is not required on the system to monitor.
- This monitor is platform-independent.
- Two additional statistics attributes, FINWAIT and CLOSEWAIT, are added to this monitor, which can be retrieved through SNMP interface.
Monitored attributes for TCP monitors
The following table lists the monitored attributes of a TCP monitor:
| Attribute | Description |
|---|---|
| Min Retransmission Timeout (#) | Minimum value permitted by TCP implementation for retransmission timeout. It is not an average for that poll period. It is a normalized value. Only one poll is required for sample. |
| Max Retransmission Timeout (#) | Maximum value permitted by TCP implementation for retransmission timeout. It is not an average for that poll period. It is a normalized value. Only one poll is required for sample. |
| Connections Established (#) | Total connections established. It is not an average for that poll period. It is a normalized value. Only one poll is required for sample. |
| Connections Active (per sec) | Number of times TCP connections have made a direct transition to the SYN_SENT state from the CLOSED state. It is the rate at which the value changes for that poll period. |
| Connections Passive (per sec) | Number of times TCP connections have made a direct transition to the SYN_RCVD state from the LISTEN state. It is the rate at which the value changes for that poll period. |
| Connections Failure (per sec) | Number of times TCP connections have made a direct transition to the CLOSED state from either the SYN_SENT state or the SYN_RCVD state. It is the rate at which the value changes for that poll period. |
| Connections Reset (per sec) | Number of times TCP connections have made a direct transition to the CLOSED state from either the ESTABLISHED state or the CLOSE_WAIT state. It is the rate at which the value changes for that poll period. |
| Segments Received (per sec) | Total number of segments received, including those received in error. It is the rate at which the value changes for that poll period. |
| Segments Sent (per sec) | Total number of segments sent, including those on current connections but excluding those containing only retransmitted octets. It is the rate at which the value changes for that poll period. |
| Segments Retransmitted (per sec) | Total number of segments transmitted again, that is, the number of TCP segments transmitted containing one or more previously transmitted octets. It is the rate at which the value changes for that poll period. |
| Error Segments Received (per sec) | Total number of segments received in error (Example, bad TCP checksums). It is the rate at which the value changes for that poll period. |
| Resets Sent (per sec) | Number of TCP segments sent containing the RST flag. It is the rate at which the value changes for that poll period. |
| Connections in FIN_WAIT (#) | Information about the state of the TCP connection. Represents TCP connections either in finWait1 state or finWait2 state. This is not an average for the poll period. It is a Normalized value. Only one poll is required for sample period. |
| Connections in CLOSE_WAIT (#) | Information about the state of the TCP connection. Represents TCP connections in closeWait state. This is not an average for the poll period. It is a Normalized value. Only one poll is required for sample period. |
Configuration information for TCP Monitors
The following table describes the information required during configuration of the TCP Monitor.
| Input fields on the Configuration tab | |
|---|---|
| Instance Name | Specify a name for the monitor instance. This is displayed on folders, graphs, lists, etc. |
| Tag | Specify a tag value to search or group the objects. You can enter multiple name-value pairs, text, or include both (name-value pair and text). This gives you the option to group objects in different forms. Click the + icon to select a default tag class. The Enter Tag Details screen opens. Select Tag Class from the drop-down list and enter a tag value. Click Add to append tag class and value. Click Finish. Note The tag specified by you is not considered as a default class. You cannot enter special characters &"/<>' in the tag field.
|
| Source Agent | Select the name or IP address of the BMC ProactiveNet server or BMC ProactiveNet agent to host the new monitor instance. |
| Target IP/Host Name | Specify the name/IP address of the machine hosting the service. BMC ProactiveNet recommends that the target IP and the IP address of the device where the monitor is being created be the same. |
| Port | Specify the port on which SNMP Agent is running |
| Community String | Specify the community string for the SNMP device. This string can be used to restrict SNMP access to a device. Most devices use 'public' as a community string. |
| SNMP Version | From the drop-down list, select the SNMP version to be monitored. |
| System-populated fields | |
| Max Connections | Limit on the total number of TCP connections the entity can support. |
| Control tab | |
| Collect Data | Select/clear check box to enable/disable data collection for the monitored component. By default, data collection is enabled (check box selected). |
| Statistics Poll Interval | Specify the time interval between two consecutive statistical data polls. Default is 5 minutes. |
| Statistics Poll Timeout | Specify the time-out period for each statistical data poll. If no data is available at the end of this interval, system records zero availability and proceeds with the next scheduled poll. Default is 20 seconds. |
| Configuration Poll Interval | Specify the time interval between two consecutive configuration data polls. Default is 24 hours (One day). |
| Configuration Poll Timeout | Specify the time-out period for each configuration data poll. If no data is available at the end of this period, the system records zero availability and proceeds with the next scheduled poll. Default is 30 seconds. |
| Group tab | |
| Group Membership | This enables you to add a device and its monitors to one or more groups. Optionally, you may associate the device with a group. A group is used to logically organize the devices and monitors by department, geographic area, or service. Assigning a device to a group also enables the system administrator to restrict or allow access to certain devices. To associate the device with a group, select the group name here. |
Note
This monitor does not support Availability attribute.
Active connections, Passive connections, and Established connections are the primary health indicators for this monitor.
NO_DATA is generated if SNMP agent goes down.
Public as community string with read only privileges is required for creating TCP monitor.
Additional configuration and setup issues
The following table presents details about Additional configuration and setup issues for the TCP Monitor.
| Issue | Details |
|---|---|
| Monitor configuration | No additional configuration requirements for this monitor. Maintain the default configuration of BMC ProactiveNet. |
| External setup | To create a Generic TCP monitor on Linux Advanced Server 3.0, modify snmpd.conf. Default entries in snmpd.conf are:
|
| External dependencies | No external dependencies for this monitor. |
| BMC ProactiveNet agent dependencies | BMC ProactiveNet agent installation on the monitored host is not mandatory. Monitor instance can be created on any BMC ProactiveNet agent. |
Overview
Content Tools
Apps
Reporter Replacement
com.bmc.atlassian.bmc-util-plugin-2.section.label
Tasks