[Fusioninventory-devel] Why to execute the Inventory task in all cases?

[Guillaume Rousse]

Le 03/05/2013 09:47, Tomás Abad a écrit :
>     Hello to everybody,
>
>     Now that I'm testing the new FusionInventory Agent v2.3.0, nowadays
> in development, creating some NetInventory and NetDiscovery tasks in
> FusionInventory for GLPI, I have seen, taking a look to the
> FusionInventory Agent registry file, that the Inventory task is always
> executed, independently of the task requested (NetDiscovery or
> NetInventory, in this case).
>
>     I think this behaviour is a waste of time and resources (It's an
> opinion). I suppose there will be a good reason to do that but, why
> don't execute only the requested task?
Quoting the documentation:
When executing, the agent tries to run every available task for every 
configured target.

So that's not just the local inventory task that is always executed, but 
every available one. Only the local inventory task can be executed 
without external parameters, tough, meaning other are generally quickly 
short-circuited.

Yes, that's silly. Hence the need for a better, self-explaining 
execution model, and not just a better communication protocol. For 
instance, using distinct task-dedicated executables, such as 
fusioninventory-inventory, fusioninventory-esx, etc... when you're only 
interested in a specific task.
-- 
Guillaume