Activation process cannot activate xxxx. due to error: id xxxx is locked by MulticastServicePool-1#

[yatinb]

Hello

We are getting the message during activation on CC after creating mass orders.

Activation process cannot activate [23262077] due to error: id 22610979 is locked by MulticastServicePool-1#

Also, the activation process takes a lot of time per partition to scan and complete the activation.

This is a single host setup for development system.

Any pointers would be helpful.

Thanks

Yatin B

[yatinb]

Additional information from rater trc file:

#2.#2020 10 20 17:23:34:815#0-700#Path#com.highdeal.pnr.hci.server.RaterBulkOperation#

###C0000ACF40580032000002C000007CCC###com.highdeal.pnr.hci.server.RaterBulkOperation.run()#<system>#######Thread[RaterActivationProcess-20,5,main]#Plain##

Caught com.highdeal.dal.CannotLockException: id 22610979 is locked by MulticastServicePool-1

at com.highdeal.dal.sql.ExclusiveLock.lock(ExclusiveLock.java:67)

at com.highdeal.dal.sql.ExclusiveLockManager.lock(ExclusiveLockManager.java:73)

at com.highdeal.dal.sql.SQLDALTransaction.lockExclusive(SQLDALTransaction.java:680)

at com.highdeal.dal.sql.SQLDALTransaction.lockExclusive(SQLDALTransaction.java:654)

at com.highdeal.dal.sql.SQLSubscriberAccountDAL.getSubscriberAccountById(SQLSubscriberAccountDAL.java:231)

at com.highdeal.dal.sql.SQLSubscriberAccountDAL.get(SQLSubscriberAccountDAL.java:194)

at com.highdeal.dal.sql.SQLSubscriberAccountDAL.get(SQLSubscriberAccountDAL.java:127)

at com.highdeal.dal.data.DALChargingContractEntity.getSubscriberAccount(DALChargingContractEntity.java:44)

at com.highdeal.pnr.hci.server.CleanupRateOpImpl.getSubscriberAccount(CleanupRateOpImpl.java:446)

at com.highdeal.pnr.hci.server.CleanupRateOpImpl.execute(CleanupRateOpImpl.java:142)

at com.highdeal.pnr.hci.server.RaterChargingContractActivationHandlerImpl.doOnSubscriberAccountSubElement(RaterChargingContractActivationHandlerImpl.java:274)

at com.highdeal.pnr.hci.server.RaterBulkOperation$SubscriptionBulkOpHandler.run(RaterBulkOperation.java:730)

at com.highdeal.pnr.hci.server.SharedThreadPoolFactory$SharedThreadPoolRunnable.run(SharedThreadPoolFactory.java:175)

at com.highdeal.hci.util.ThreadPool$ThreadPoolRunnable.run(ThreadPool.java:261)

at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)

^Cat java.util.concurrent.FutureTask.run(FutureTask.java:266)

at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)

at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)

at java.lang.Thread.run(Thread.java:836)

#

[former_member428395]

Dear Yatin,

please find below my responses:

1 - please check the value of the parameter ACTIVATION_SCHEDULER_MAX_THROUGHPUT and increase the parameter value if requested.

2 - see also the parameter ACTIVATION_SCHEDULER_MAX_THROUGHPUT.
Please take care to not create provider contract with an effective date far in the past. Indeed, in this case, SAP CC has to compute all events between the effective date and the current date.

3 - As this is on S/4 side, please open a new question on S/4 Tag

Best regards,

Eric F.

[yatinb]

Hi Eric,

We are doing Mock cutover on QA systems with the mass order creations as well as Integration testings. So if we run the activation scheduler only nightly basis, for Integration testing, the team needs to activate test orders manually in hundreds, hence our scheduler runs every couple of hours and getting into locking issues.

Also, once the 300-400k new orders are replicated in CC, the rater runs very show even though we increased activation threads, queue sizes, cache etc.

Below are few points where if you could provide some details, it would be helpful.

1. When rater takes long time inactivation, not heavy CPU or memory usage experienced . Also the HANA DB server also seen at 5-10 % usage. How can the activation be improved ?

2. While the activation seem be stuck on a partition for a long time, there is really not much load seen on the system as well the top processes show limited Rating thread using 50-80% CPU even though more than 64 activation threads are assigned.

3. On S/4 side, table DFKK_CC_DO has about 1.8GB size in memory but on the disk it is 160GB and growing very rapidly. I do not find any details on how to control it.

Thanks

Yatin

[former_member428395]

Hello Yatin,

the Cannot lock resource that you noticed is most probably due to another process which is running and is locking the contract which has to be activated.

Take care to not have mass provisioning job (or any other mass job) during the activation. Usually, the activation is done during the night when there is no other activity (or when there is less activity) .

Eric F.