Thanks, Jon! Thanks, Robyn! I'll fix this for the next release.
Lindsay -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Robyn Orosz Sent: October 8, 2007 7:24 AM To: Jon Cc: [email protected] Subject: Re: [Vyatta-users] OSPF Passive Hi Jon, You are correct that the XORP user manual does correctly describe the current behavior. This behavior has been reported as an issue in Vyatta Bugzilla: https://bugzilla.vyatta.com/show_bug.cgi?id=1793 And in XORP Bugzilla: http://www.xorp.org/bugzilla/show_bug.cgi?id=566 I'll forward this message to our tech writer so she can either make a note in our documentation or add a release note. The behavior described in the Vyatta Command Reference is the expected behavior of an OSPF passive interface but, due to issue 1793, the network attached to the passive interface is always advertised as a /32. To work around this issue, you can remove the passive interfaces from the OSPF configuration and instead advertise their networks in an OSPF export policy. Thank you again for noticing this discrepancy in our documentation. I'll make sure something is noted until issue 1793 is fixed. -Robyn Jon wrote: > Hi, > > In the Vyatta command reference, the use of ospf passive mode is described > as: (With small typo) > > Optional. Determines whether OSPF sends hello messages out >ON< this > interface. > > If hello messages are not sent, neighbor relationships will not be > established on that interface. However, because the interface is still > part of the OSPF configuration, the subnet attached to this interface will > still be included in the internal OSPF routes. This can be a useful > security mechanism at the edge of a network. > Supported values are as follows: > true: Hello messages will not be sent over this interface. > false: Hello messages will be sent over this interface. > The default is false. > > In the xorp user manual this mode is described as: > > passive: This takes the value true or false. The default setting is false > it can be set to true to set the interface in loopback. The protocol is no > longer run on this address and the Router-LSA contains a host route for > this address. > > Tests we have done with this mode suggests that the xorp description is > the correct one, since the subnet on the interface set to passive is no > longer distributed. > > Can someone confirm this mode of operation? > > -- Jon > _______________________________________________ > Vyatta-users mailing list > [email protected] > http://mailman.vyatta.com/mailman/listinfo/vyatta-users > _______________________________________________ Vyatta-users mailing list [email protected] http://mailman.vyatta.com/mailman/listinfo/vyatta-users _______________________________________________ Vyatta-users mailing list [email protected] http://mailman.vyatta.com/mailman/listinfo/vyatta-users
