PCCE 11.6 - difference in Termination call details tables

[touma.kayal]

Dears,

I have a PCCE solution version 11.6, after restarting side A (Rogger, AW, CVP, PG), calls when to side B normally.

now after side A when up again, i saw a difference in the number of records in Termination_Call_Detail tables between side A and Side B. 

i tried to reset both logger A and Logger B services to force sync between these 2 Sides, then i restarted the distributor service on both side A and Side b AWHDS. but this didn't help. to note that the difference is the number of calls that were routed to side B when side A was down. 

To note that sync is working fine between loggers as well as between logger A/AWHDS A and logger B/AWHDS B. but only these records are not synchronized.

Appreciate your prompt support on this matter

Regards,

[bill.king1]

This is certainly possible, as you might have a gap somewhere in the database, but now the two are in sync so that missing data that is in one will never be pulled over to the other without manual intervention.

If you really want to have both of them match and you're sure that one side has everything you need, you could try and follow a process like this:

https://www.cisco.com/c/en/us/support/docs/customer-collaboration/unified-contact-center-enterprise/200747-Manually-Resync-HDS-B-to-HDS-A.html

[touma.kayal]

Hi Bill,

Thank you for your reply, but manual intervention is somehow inefficient for me as this restart happens once a month on each side( for windows updates purposes).

is there anyway or any restart sequence that can prevent this gap ?

thank you

Regards,

[bill.king1]

Yes, do one side at a time, with a gap between sides so that after each reboot, the systems can get back in sync. This way you allow the systems to be synched and then resynch when they come back up.

[touma.kayal]

But there is already 1 week timeframe between each restart

[bill.king1]

So in the most basic sense, say side A shows 100 rows of data and B side shows 100 rows of data.

You stop A, do your work, and bring A back up but don't start services.

So A still has 100 rows, but B has 110 rows for calls that came in while A was down.

You start services on A, does your A side catch up those missing rows so that at some point, both A and B then show 115 rows (so A catches up the 10 it missed, plus 5 more new calls).

If it doesn't, then you have some type of replication problem.

[touma.kayal]

Hi Bill,

The 10 rows will not replicate to side A , but each and every call after A is up, gets synchronized and replicates. Then after side A is up, everything goes back to normal behavior.

So it is only the 10 rows that did not replicate.

Regards,

[bill.king1]

You should open a trouble ticket then, as that means some of the most basic redundancy is not working. The design guides go over this in detail, but unless it is a case where the one system is down for hours, when the one comes back up, it should "catch up" automatically.

It might be worth doing an exit router during a maintenance window as part of the attempt to make sure everything is in synch as well.

https://www.cisco.com/c/en/us/td/docs/voice_ip_comm/cust_contact/contact_center/icm_enterprise/icm_enterprise_11_6_1/Design/Guide/ucce_b_soldg-for-unified-cce-116/ucce_b_soldg-for-unified-cce-116_chapter_0110.html#concept_116F80365E2B04FD61DCF4C87FBE...

[Ramamoorthy Shanmugam]

Hi,

Please verify below registry setting for Logger & AW-HDS on both the side.

HKEY_LOCAL_MACHINE<SOFTWARE<Cisco Systems, Inc.\ICM\{Instance Name}\LoggerA\NodeManager\CurrentVersion\Processes\rpl.

side - A
/db pcce_sideA /server /name PCCE-RGRA /replicationport 40026 /recoveryport 40028

Side -B
/db pcce_sideB /server /name PCCE-RGRB /replicationport 41026 /recoveryport 41028

HKEY_LOCAL_MACHINE<SOFTWARE<Cisco Systems, Inc.\ICM\{Instance Name}\Distributor\NodeManager\CurrentVersion\Processes\rpl.

Side -A

/db pcce_hds /client /name pcce-rgra /replicationport 40026 /recoveryport 40028 /all

Side-B
/db pcce_hds /client /name pcce-rgrb /replicationport 41026 /recoveryport 41028

Between Logger A & B if huge recovery key mismatch or central controller partner is offline more than 6 hours there is no automated way to pull the old records.

We need to sync data between A to B or B to A .( check the rttest for router sync) and check the Logger RCD table RecoveryKey counts for A/B.

example :-

Select max(RecoveryKey) from Termination_Call_Detail where DateTime between '2019-07-18 00:00:00' and '2019-07-18 23:59:00'
as you can see the Logger has a recovery key value of “827019xxxxxxxxx”, which is lower than the AW based on your issue .

Ram.S