When Path Switch Request Message is sent from eNodeB to MME?

Question

As one failure message is defined in specification for the path switch procedure then what would be the behavior of target eNodeB after getting failure response from the MME ? Action would be vendor specific or standard action has been defined for this scenario ?

Answer 1

During X2 Handover procedure, the target eNB sends the path switch request message to the MME to inform that the UE has changed cell & the SGW needs to forward the incoming packets to a different destination.
So the MME sends a modify bearer request message to the SGW to update the downlink TEIDs & ip address.
The SGW is supposed to use this updated TEID (tunnel end point identifier) & address to forward any incoming packet to the target enodeB.

The SGW changes the downlink data path to the target enB. But before that the SGW sends the end marker packets on the old path to the source eNB informing that these are the last few packets you are receiving after which the incoming packets will be sent to the target enodeB.
SGW responds back with a modify bearer response message to the MME.

The MME responds back with path switch ack message.
Now the target enodeB informs the source enodeB about the successful handover by sending a UE context release msg so that the source enodeB releases all the radio and control plane related resources associated with that particular UE.

In short you can say that the path switch request procedure is used to inform the SGW to switch the path of downlink packets & re-route the incoming packets to a new destination.

Hope this helps!

Answer 2

When an x2 based handover takes place

Comments

Prior to answering your question I would put some light on why handover takes place in a simple way.
A UE keeps measuring its signal strength in intervals and reports to the enodeB. If the UE  measurement report says the measured signal strength is below a particular threshold, enodeB has to handover to a different cell that can provide a better signal strength. So enodeB is the deciding entity whether to handover or not.
Now coming to your question as to when X2 handover takes place. It depends on the availability of an X2 interface between 2 enodeBs. Generally an enodeB tends to go for an X2 based handover. But in case where an X2 interface is not present between 2 enodeBs, we go for an S1 based handover, i.e. via MME.

---

Hi,
i am facing issue, when MME receive path switch request message from target enodeB , it sends create session request towards SGW instead of modify bearer request.Due to which failure occuring. And MME sending path switch failure in response to target enodeb with cause code 4(unspecified).

---

On receiving the Path Switch Req msg, MME has to figure out if there is a need for SGW Relocation or not. If an SGW relocation is required, MME sends a Create Session Req msg to the NEW SGW explicitly indicating that the SGW should send Modify Bearer Req msg to PGW. PGW responds back with Modify Bearer Response to SGW. The SGW responds back to MME with Create Session Response msg. If an SGW relocation is not required, MME sends a Modify Bearer Req to SGW.
Now if your MME is sending CSR to SGW, its clear that the SGW relocation is required. Now why are you receiving a failure? Honestly it can be anything. Maybe the handling of Path Switch Req at MME has some issue or maybe your new SGW is rejecting CSR coz of load balancing. It is difficult to comment on the issue unless you share the wireshark capture between eNB & MME, MME & SGW, SGW & PGW.