top button
Flag Notify
Site Registration

Handle error scenarios in the PCRF (over Gx interface)

+3 votes

I am trying to handle all the error scenarios seen while testing the Gx with the third party. As of now I had coded for just the successful scenarios.

I am looking for some of inputs on these error scenarios and how to proceed and exit gracefully (rather than crashing the code)

  • Dynamic rule config with wrong parameters., PGW sends CCR with the CCR type=TERMINATION_REQUEST. How can PCRF respond (just log error and comeout ?)

  • Static rule config with the "wrong rule name". When the PCRF sends CCA-I with the wrong rule name, PGW sends CCR-U with the Charging Rule Report with the "Rule-Failure-Code" as "Unknown Rule Name". In this case remove those rules and continue OR just comeout with just logging error)

posted Sep 18, 2015 by Pdk

Share this question
Facebook Share Button Twitter Share Button LinkedIn Share Button

3 Answers

+2 votes
Best answer

Just a small correction
Dynamic rule config with wrong parameters., PGW sends CCR with the CCR type=TERMINATION_REQUEST. How can PCRF respond (just log error and comeout ?)

In this case the CCR-Type=UPDATE_REQUEST(2) should go.

Also, it depends on how policies are configured on PCRF side. Maybe PCRF can send CCA and then PGW will continue with session creation or CCA can be sent with a new rule name.(Need to check spec once)

Peeyush Sharma

answer Sep 18, 2015 by Peeyush Sharma
Thanks a lot Peeyush , yes that looks correct approach, but how can PCRF know which parameters are having issues..
The CCR U is not very clear in responding with the error causes. (to pinpoint which is causing this issue). So this may keep happening , as PCRF may keep sending wrong value and PGW keeps rejecting..(This may trigger some timer expiry at PGW and SGW for the attach). Not sure whether this is right to wait till timer expiry..Or try some reasonable times and exit ...
+1 vote

On PCRF user can create a policy that if CCR-U comes with Rule-Failure-Code then some action should be taken. Else anyways PCRF has to handle the CCR message received and you can put some condition
E.g. If CCR-U has some rule-failure-code or not., if yes then what action should be taken. I guess that would be a costly affair though.

Peeyush Sharma

answer Sep 18, 2015 by Peeyush Sharma
Thanks, yes..But as of in my basic design, not much error handling..
In pcrf I create a map of records (like for this IMSI, these are apns, and these are rules and tfts in that apn). So I update the map on CCR, RAR etc .
So as of now I can handle this case as part of CCR message handling..But may be as you suggested there are better ways of handling this
+1 vote

Why are you storing info for a subscriber in PCRF(like APN, quota etc and rules)

You can use SPR for half of the things and then design a policy rule engine where a operator/customer can define policy by itself.

Let's say I want a IMS APN and I want to block the http traffic over default bearer then maybe operator can define a static rule where if Bearer-Usage is IMS_SGINALLING send static rule in CCA

Make it more flexible from the beginning is what I would suggest. It would be a tough task though. There are a lot of parameters which you need to provide in policy design.

All the best

Peeyush Sharm

answer Sep 18, 2015 by Peeyush Sharma
Thanks ..I thought SPR (Or in our case HLR) has info regarding the user subscription. (like user allowed Apns, bit rates etc). Could we add info like currently active apns , rules and their parameters (like TFT) in the SPR ? I thought it is stored in PCRF..May be my initial assumptions are wrong :(...Quite difficult to change design now..But I will try ..
No no, in-call related stuff should not be copied to SPR. Just the subscription info should be in SPR. In-call attributes should remain with PCRF only.

Peeyush Sharma
Similar Questions
+3 votes

Looking for the information related to PCRF.

Another specific question. When the PCRF decides on a QOS and sends across to P-GW. P-GW might not have got any info related to Dedicated bearer to be established earlier. So how does it map this newly sent QOS for the new dedicated bearer to the one it requested by P-CSCF.

I am trying to understand, how to map the QOS info sent by PCRF to the "dedicated bearer request by the P-CSCF. Can PCRF on sending the QOS, also create the new dedicated bearer id, which is sent back to P-CSCF ?

Any info regarding with is very helpful.

+1 vote

Looking at interoperability issues.
The CCRequest message sent from PGW to PCRF during attach. Does the CCA sent back from PCRF to PGW needs to have TFT info?

+2 votes

Could you help me figure out CCRequest basic message contents sent from the PGW to PCRF. I am in a process of replacing our Gx interface from the static calls to the Diameter interface.

I am interested in any discussion about this interface.

I am also trying to understand , how the APN info is passed from the PGW(PCEF) to PGW. I think the Called-Party-AVP is used. It is bit contradicting info in the net than the one given in spec, (which says it is used to indicate Emergency APN)

+1 vote

Hi all,

We have complexity linked to Gy and Gx interfaces.

Let's say that we have two rules configured as static on PGW with :

Rule1= Reports RatingGroup1
Rule2= Reports RatingGroup2

When PCRF install Rule 1 on Gx, the ratingGroup reported by PGW to OCS is RatingGroup1
When PCRF uninstal Rule1 then install Rule2, the ratingGroup reported by PGW remain RatingGroup1 till a Gy trigger is reached (revalidation time, GSU consumption...).
Then the problem on our system is that when PCRF change the rules on Gx interface the PGW keep reporting the old rating group on Gy interface ie: Gx change do not lead to a Gy change similtanously. I m wondering if there is a standard behaviour for such a use case?

Thanks a lot