09-16-2013 09:53 PM - edited 03-07-2019 03:30 PM
Im studying for CCNP switch right and was trying to figure out what happens to communication if the links from the Active switch to the core go down and their the connection between the active and standby, would all the traffic go to the active switch and just stop? Becuase all the links try and find the best way to the root and the root would remain the same.
maybe im just over thinking it...
see attatched.
thanks
09-16-2013 10:42 PM
Hi Jeffrey,
as long as the core-switches have layer-2 connectivity to each other in the affected VLANs, the HSRP status will not change.
If the link between the core switches fails and also the link between one access-switch and the primary core, the spanning-tree will calculate a new topology through another access-switch (if another one exists). This will result in an undesired forwarding path, especially for clients connected to the access-switch which lost the primary link, since forwarding has to follow the spanning-tree topology (because learning MAC-addresses and building the MAC-address tables also do).
If there is no other layer-2 path between the core-switches, the secondary will change from HSRP standby to active, the MAC-address of it's virtual IP-address will be the same as for the other core (which also is in active state) and clients attached to the access-switch will still be able to reach their default gateway after a short outage. What other destinations (e.g. internet) they can reach depends on how routing is implemented.
Hope that helps
Rolf
09-17-2013 12:19 AM
Good day, Jeffrey.
would all the traffic go to the active switch and just stop? Becuase all the links try and find the best way to the root and the root would remain the same.
Yes it will but not because active switch as well STP root, but cos it will still have active role in HSRP and all default gateways will point to it. Indeed some of previously blocked link to one of access switches will be unblocked after link between switch1 (RSTP root and HSRP active) and switch 2 will fail. So traffic from any access switch will have possibile way to get as to Switch1 as well to Switch2, but over suboptimal path - for eg if bottom middle switch will unblock it port traffic flow to Switch2 from bottom right switch will be: Switch1-> middle boottom switch ->Switch2. So connectivity in segment still present and all we need to do to prevent traffic from stopping on Switch1 is use one of two solutions:
2) Add additional L3 link between Switch1 and Switch2 and run some of IGP (OSPF for eg.)
Regards,
09-17-2013 01:25 AM
So as long as HSRP interface tracking was used against the uplink to the core, the standby would become active, and traffic would still find a way out but passing through the original switch(upper left) because it is the root.
Makes since thank You
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide