The message sent from the core to the drivers is built by VirtualMachineManager::format_message. This message is then received by the driver, see for instance one_vmm_exec.rb poll method, and decoded (VirtualMachineDriver.rb).
I guess you can safely add a new xml element to the message if you need to. Let me know if this works out. Regards -- Carlos Martín, MSc Project Engineer OpenNebula - The Open-source Solution for Data Center Virtualization www.OpenNebula.org | [email protected] | @OpenNebula<http://twitter.com/opennebula><[email protected]> On Mon, Mar 25, 2013 at 1:47 PM, Tobias Zillner <[email protected]>wrote: > It depends which commad i am executing. Dump memory is resource intesive > but commds like read the process list shoudn't cost to much performance. > > The problem is that the usershould be able to choose wich command is > executed. So i must add a possibility to get this command to the driver. Or > am i thinking wrong? > > Regards, > > Tobias > > *Gesendet:* Montag, 25. März 2013 um 13:19 Uhr > *Von:* "Carlos Martín Sánchez" <[email protected]> > *An:* "Tobias Zillner" <[email protected]> > *Cc:* users <[email protected]> > *Betreff:* Re: [one-users] vmm driver result in shell > Is your command very resource-intensive? > > I'm thinking that maybe you could just add your commands to the periodic > VM monitorization. Or add a flag somewhere to indicate if the operation > should be done in the next poll action... If you avoid to change the > OpenNebula code it will be easier to distribute, use with newer versions, > etc. > -- > Carlos Martín, MSc > Project Engineer > OpenNebula - The Open-source Solution for Data Center Virtualization > www.OpenNebula.org | [email protected] | > @OpenNebula<http://twitter.com/opennebula> > > On Mon, Mar 25, 2013 at 12:59 PM, Tobias Zillner <[email protected]>wrote: >> >> Hi, >> >> thank you for your quick answer! >> >> I am executing a volatility command on the driver. Results are some kind >> of lists or process information which is show directly on the screen/shell. >> >> At the moment i added functionality to one.vm.action to process my >> commands. Which part would be best to save the results in a document >> ressource or add it to the vm template? >> How can i invoke this? >> >> Regards, >> Tobias >> >> 2013/3/25 Carlos Martín Sánchez <[email protected]> >>> >>> Hi, >>> >>> What kind of results will you present to the user? >>> >>> >>> The commands that trigger a driver action are asynchronous, so you can't >>> get any result immediately to the user. >>> >>> If you are modifying the c++ core, you could add your custom information >>> in the VM template, this way the users will see it in the 'onevm show' >>> output. >>> >>> Or you could use the generic document resources [1]. These are just like >>> the templates managed with onetemplate; they will store anything and will >>> also have owner, group, acl, permissions... >>> >>> Regards >>> >>> [1] >>> http://opennebula.org/documentation:rel3.8:api#actions_for_document_management >>> -- >>> Carlos Martín, MSc >>> Project Engineer >>> OpenNebula - The Open-source Solution for Data Center Virtualization >>> www.OpenNebula.org | [email protected] | >>> @OpenNebula<http://twitter.com/opennebula> >>> >>> On Fri, Mar 22, 2013 at 2:37 PM, Tobias Zillner >>> <[email protected]>wrote: >>> >>>> Hello, >>>> >>>> my Name is Tobias Zillner and I am working on a master thesis about >>>> forensics in the cloud. >>>> >>>> For my practical part I decided to add forensic services to open >>>> nebula. >>>> >>>> So i extended the onevm commad and added a function forensics. I >>>> extended the whole way through the code. >>>> At the moment i am able to execute my own vmmd script, in which a >>>> forensic tool (volatility) is called and analyse the virtual machines. >>>> >>>> But now I don't know how to get the result back to the user. Basically >>>> i want to create a scenario like this: >>>> >>>> A user just puts in onevm forensics id and the analysis result is shown >>>> to him in the screen. >>>> >>>> Anybody can help me here and tell me what funtions to call or how the >>>> way back to the user shell works? >>>> Does anyone have experience with adding funtions and the flow of >>>> commands through the code? >>>> >>>> THX! >>>> >>>> Best Regards, >>>> >>>> Tobias >>>> _______________________________________________ >>>> Users mailing list >>>> [email protected] >>>> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org >>>> >>> >>>
_______________________________________________ Users mailing list [email protected] http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
