Browse Prior Art Database

A Solution to self-destroy an agent application Disclosure Number: IPCOM000236917D
Publication Date: 2014-May-22
Document File: 6 page(s) / 82K

Publishing Venue

The Prior Art Database


Out-of-date and useless agent is kept in the machine after completing service lifecycle or encountering connection issues. The solution introduced in this paper resolved this problem, and keeps agent machine clean once agent becomes useless.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 53% of the total text.

Page 01 of 6

A Solution to self

A Solution to self-

Agent application is widely used in Client-Server infrastructure management applications. When a device needs to be managed to be managed by an application, the 1st step is to deploy an agent on that device. Both server and clients work well during the normal communication period, but imagine that, some day, this device doesn't need to be managed by server application, admin will normally remove it from server side. Another case is, agent loses communication with server for a long time (several months) because of network issue, and another admin would like to manage this device again with different agent version. Then what's happened? Thereare many issues caused by previous deployed agent:

. the out-of-date agent still locates in device as rubbish which occupied disk space

. the info of out-of-date agent may cause potential security risk

. the out-of-date agent still occupy the default service port, and cause some problem in future deployment

. the different agent version may cause future installation issue and need more effort to clean it up

The solution which described in this paper is focus on how agent can do self-destroying.

When agent is in out-of-service state, it actively communicates with application server, confirming whether it completes service life-cycle or not. If yes, it can destroy itselfin order to keep agent machine clean. Figure 1 describes main ideas of this solution.

Figure 2 introduces how this solution works. Application admin needs to set up some conditions for self-destroy function, max_idle_time, max_try, and timeout. max_idle_time means the max time which agent stay in idle state. max-try means the max times which agent try to connect with application server. Timeout means the time which agent can wait for server's response.

A process is created to keep checking its status, if its idle time is longer than max_idle_time, it start trying to connect with application server. Once agent is defined as useless, process will clean up installation configuration, occupied ports, and completely destroy the agent and close itself.

Figure 3 describes how to define the agent is useless and can be removed. Firstly, agent t...