locked
PNRP Name Resolution RRS feed

  • Question

  • Hi,

    I've been reading this article and trying to understand PNRP name resolution, but it seems like there is one important thing that is not explained. Initially, when a program wants to resolve a peer using only its peer name, how does it do it? Clearly, the peer on which the program is running has no cache yet (well only its own PNRP ID), so how does it resolve the peer? The actual process of resolving a peer name once the cache has at least one other ID in it is fine, but I'm wondering how the process gets started. Perhaps if the cache is empty, a multicast message is sent out over the network? But PNRP is meant to work over the internet, so I'm not sure how that would be possible outside a LAN... Basically what I'm asking is how does a computer actually become part of a mesh and known by other peers? From what I read it seems like even when a peer becomes part of a mesh, it does not necessarily know about all the other peers in it.

    Any explanation of the details of PNRP or a pointer to a more informative article would be very much appreciated.

    Thanks,

    Alex

    Sunday, May 18, 2008 12:07 PM

Answers

  • As far as I know, what your understanding would be correct in case of Local Subnet called LAN. In case of Internet outside LAN, I believe it bootstraps by sync with the SeedServer and make intial cache. Using intial cache, the PNRP node can proceed the naming resolution task. You can see the SeedServer info by typing "netsh p2p pnrp cloud show initialization" for more info. Hope it helps. 

    Thursday, May 29, 2008 11:10 PM

All replies

  • As far as I know, what your understanding would be correct in case of Local Subnet called LAN. In case of Internet outside LAN, I believe it bootstraps by sync with the SeedServer and make intial cache. Using intial cache, the PNRP node can proceed the naming resolution task. You can see the SeedServer info by typing "netsh p2p pnrp cloud show initialization" for more info. Hope it helps. 

    Thursday, May 29, 2008 11:10 PM
  • Thanks for the explanation - I also found some text that seems to confirm what you're saying. All is clear now.
    Friday, May 30, 2008 11:13 AM
  • Thanks for the explanation - I also found some text that seems to confirm what you're saying. All is clear to me now.
    Friday, May 30, 2008 11:13 AM