On Mar 5, 2014, at 11:07 PM, Tero Kivinen <[email protected]> wrote:

> In section 2 it says:
> 
>   Note that IKEv2 and IPsec session do not need to be on the same node
>   as IKEv2 and IPsec context are different.
> 
> This is not so easy. The RFC5996 says:
> 
> ----------------------------------------------------------------------
> 2.4.  State Synchronization and Connection Timeouts
> ...
>       An implementation needs to stop sending over any SA if
>   some failure prevents it from receiving on all of the associated SAs.
>   If a system creates Child SAs that can fail independently from one
>   another without the associated IKE SA being able to send a delete
>   message, then the system MUST negotiate such Child SAs using separate
>   IKE SAs.
> ----------------------------------------------------------------------
> 
> I.e. if any of the IPsec SAs fail, then all of IPsec SAs created using
> same IKE SA, and the IKE SA must also fail. If IPsec SAs and IKE SA
> are on separate nodes, that do set up new kind of requirements for
> those nodes. I.e. if one node having IPsec SAs fails, the node having
> IKE SA needs to detect this, and send delete notification for each
> IPsec SA that were in that node. Also if the node having the IKE SA
> will fail, then all the IPsec SAs associated with that IKE SA, must
> stop sending, i.e. they needs to be destroyed.                  

Tero's comment nails the concern I had when reading the document. IKEv2 ties 
Child SAs to their parent SA, and using the context of one without the other 
seems dangerous.

--Paul Hoffman
_______________________________________________
IPsec mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/ipsec

Reply via email to