question about agent deployment / agent.properties

[bcm1]

Hi,

I need to create some sort of simplified agent deployment mechanism.
basically I am in a multi vendor environment.
I don't have access to all the box where the agent are to be installed and I can't use ssh / key mechanism.
But still I want to reduce human intervention to it's minimum.

So I have created my own tgz including an agent.properites.
agent.setup.camIP=x.x.x.x
agent.setup.camSSLPort=7443
agent.setup.camSecure=yes
agent.setup.camLogin=hqadmin
agent.setup.camPword=xxxxxxxx
agent.setup.agentIP=*default*
agent.setup.agentPort=*default*
agent.setup.resetupTokens=yes

Please note I have changed resetupTokensto yes.

But that's what I am unsure of. It seems to work. install the tgz and just start the init script and all good, but will the renegociate a new token each time ?

Is there a way to have some sort of pre generated token that I could embed in the tgz ?
say genereate the token on the server and then copy it to whereever it should go ?

Thanks
Ben

[BradFelmey]

Although this doesn't directly answer your question, I believe that agent deployment is one of the upcoming sekrit features of HQ. 😉

[bcm1]

Hi Brad,
Thanks for your answer, but as you say, it does not really answer my question.
is there some documentation on agent token generation ? and how to manually generate one ?
is it possible ?
what does this parameter re negotiate token in the agent property really do ?
thanks anyway.
Cheers
Ben

[MarieMcGarry · Accepted solution]

Hey Ben,

I'm a new technical writer at Hyperic. I'm working on new/improved installation/config documentation, and am scouring the user forum so I can figure out what's missing from the current docs.

Don't know whether this'll help, but here's my understanding of the token mechanism.

The token is generated the first time the Agent starts up, regardless of the setting of agent.setup.resetupTokens. Setting and leaving this property to "yes" will cause a new token to be generated each time the Agent is started. This would be useful if, for some reason, a token got corrupted or deleted.

I'm not aware of any way to generate a token manually.

There's a doc page that attempts to address the topic of reducing user interaction required for agent deployment here: http://support.hyperic.com/display/DOC/HQ+Agent+Deployment. I haven't vetted that content yet, so consider it "FWIW".

[bcm1]

That's a bloody good news !
The hyperic doc is good but definitely does not cover every aspect of the configuration/installation
Thanks for your answer it helps.
although having some sort of pre-token generation would be preferable in some case.
that will probably goes into feature request.
Cheers
Ben