ad-hoc ssh agent + control persist

[Ángel González]

On 23/09/13 16:26, Francois Durand wrote:
> On Mon, Sep 23, 2013 at 11:04 AM, Darren Tucker<dtucker at zip.com.au>  wrote:
>
>> On Sun, Sep 22, 2013 at 07:10:22PM +0200, Francois Durand wrote:
>> [...]
>>> One deployment needs several ssh connections. For obvious performance
>>> reasons using control persistance (ControlPersist=60 for instance) is
>>> highly desirable.
>> [...]
>>> Any ideas on this ?
>> use "ssh -Oexit" and "ssh-agent -k" to explicitly stop the connection
>> and agent as the last steps of your script?
>>
>
> If I understand correctly, this completely nullify the control persistance
> benefit, and not using control persistance in the first place would achieve
> the same result, wouldn't it ?

No. This way you have control persistance until you run that, where as 
not having
control persistance you never have it.

This assumes that you are able to detect when your deployment is 
finished for
performing those cleanup steps.