25 Nov/13 - Tool: User Parameter & Table Agent Duplicator(s)
One of the features of TQManager is the ability to integrate external data into the collection process and It offers two types of integration depending on where the external data will be stored in the “production” database:
User Parameter Agent: Data collected with this method will be stored into the “1-minute” aggregate as entries. The TQ aggregation process will then aggregate the data into the “10-minute”, “1-hour” and “8-hour” aggregates.
User Table Agent: Data collected with this method will be stored into a specific table file that has been created by the user for this purpose. It is part of the “production” database but the table file is completely customized by the user. In this case a table definition file should be made and introduced into the DB via the command “tqtblprb -d production -i -f ” prior to launching the agent. Aggregation of the data is done on the same table file.
Customers are using these two agent types in order to introduce more performance statistics like for example “corrected free memory due to ZFS”, “poolstat” and others into the production database.
Basically these agents are added manually via the agent WEB interface, this is OK but when several servers need to be equipped with this extra agent the manual process/approach could become too time consuming. As such we came up with a more automated mechanism, for UX/Linux, so that the extra agent(s) can be deployed via your software distribution tooling.