Agent name identification

Identifying the agent name

WhaTap basically uses the data collected from application servers to identify each server while collecting monitoring data. Basic information uses the application server as a unique identifier by a combination of application server's type, IP address, and such. If necessary, use the name set by the user or change the pattern to a unique value. The agent name must be unique.

The reason why the data extracted from the application server is used, is like the following. When a communication loss caused by the stop of the application server, network disconnection, or agent problem between the collection server and the agent, is restored, the continuity of the data from the reconnected agent is maintained.

Caution

• The agent names must be unique on a project basis.
• If the agent ID and name have been changed, previous data is not connected.

The default patterns used to identify the application server are as follows:

Default pattern

NODE-{ip2}-{ip3}

Cluster environment

NODE{cluster}-{ip2}-{ip3}

Variable of the agent name

Variable	Description
{type}	Name of the application type (Node)
{ip0}	The first byte of the IPv4 address is used. (e.g. 10 in 10.11.12.13)
{ip1}	The second byte of the IPv4 address is used. (e.g. 11 in 10.11.12.13)
{ip2}	The third byte of the IPv4 address is used. (e.g. 12 in 10.11.12.13)
{ip3}	The fourth byte of the IPv4 address is used. (e.g. 13 in 10.11.12.13)
{pid}	Application process ID
{hostname}	Host name
{cluster}	Cluster ID when multiple Node.js files are running on a server

Changing the agent name pattern

You can change the WhaTap agent name through an environment variable. Add the following code to the first line of the application's main module.

process.env.WHATAP_NAME = "NODE-{ip2}-{ip3}";
var whatap=require('whatap').NodeAgent;
...

Agent nameing on the server

Names are automatically assigned by the server rather than determining them based on the agent environment. It is used when a Node.js server operates in the container or PaaS environment.

whatap.conf

auto_oname_enabled=true
auto_oname_prefix=nodejs

Set the .env(app.js)

process.env.WHATAP_LICENSE='x46n3226be1ah-z2rsecfcvlq2ph-z11bc81gfhqpgg';
process.env.WHATAP_SERVER_HOST='52.78.209.94/52.78.224.235';

process.env.auto_oname_enabled=true;
process.env.auto_oname_prefix='mynode';

var whatap=require('whatap').NodeAgent;
...

Note

• To set in the env, it must be declared before require('whatap').
• Because the Heroku environment changes the IP address each time the server is restarted, it is changed to a new name.

The auto_oname_prefix option is the prefix of the agent name. The WhaTap server gives an agent name by combining the name and serial number set in the auto_oname_prefix option.

Assigned agent name example, mynode1

Agent name setting options

For more information about the agent name setting in Node.js, see the following.

• auto_oname_enabled Boolean

  Default false

  It enables the function to automatically obtain the agent name to register from the server.

• auto_oname_prefix String

  Default agent

  When an agent name is automatically granted from the server, the prefix of the agent name (business name as usual) is used. The prefix + serial number (from 1) is assigned.

• auto_oname_reset Int

  Default 0

  Modify it to obtain a new agent name from the server. Once set, the agent name is maintained until Node.js is restarted. For a reset, modify the value of auto_oname_reset. If a different value is assigned, the setting is applied.

• oname_port_postfix_enabled Boolean

  Default false

  You can create an agent name (ONAME) by adding a port number.

  e.g. NODE-14-103-3000

• whatap.okind String

  You can configure monitoring units by grouping multiple agents into "type-based" groups. This option allows you to monitor and manage multiple agents in groups, analyze performance metrics, and set the alerts.

  Note

  You can check it as a group option classified as By kind in the Dashboard and Analysis, Statistics menus.

• whatap.onode String

  You can configure monitoring units by grouping multiple agents into "node-based" groups. This option allows you to monitor and manage multiple agents in groups, analyze performance metrics, and set the alerts.

  Note

  You can check it as a group option classified for nodes in the Dashboard and Analysis, Statistics menus.