JDBC Sender not processing message

[Former Member]

Hello

I have a scenario JDBC SENDER -> RFC -> JDBC RECEIVER. When I have an exception on the mapping from JDBC Sender to RFC, the following messages from the JDBC Sender are not processed.

In communication channel in Rwb, there´s only :

12/4/08 5:16:42 PM Processing started

12/4/08 5:14:42 PM Processing started

12/4/08 5:12:42 PM Processing started .....

But no messages are processed. I already checked the SQL in the DB and there are some messages that should be processed.

I already checked with the Dba and there´s no user blocking the table.

I already tried to delete the Lock entries (in LOCKING ADAPTER) in Visual Administration.

I already tried to restart the JDBC service in Visual Administration

I already tried to restart the Communication Channel.

Any other ideias ? I don´t want to restart the Java from XI, because this error may happen in PRD system.

Thanks

[Former Member]

This issue was urgent so I changed the scenario to BPM. By BPM the JDBC Sender doesn´t lock when an error occurs.

Later I´ll try to do this interface again without Bpm.

Santosh, on my next try I´ll check the logs on visual admin. I already checked that note, my SP is 13 and I already tried to make this configuration, but nothing happened.

Prateek, there aren´t two communication channels pooling the same table. This error happens when there´s an unprocessed message in the communiation channel (due a mapping error). So it "locks" the following pooled messages.

Thanks for everyone´s help

[Former Member]

vitor,

this is the extract of sap note : 1083488 Kindly try this...

Symptom

Sometimes, some of the XI FTP/JDBC sender channels stopped polling permanently. However this is resolved by re-starting these channels.

Other terms

FTP sender channels, JDBC sender channels, polling stopped, stopped polling, polling adapters.

Reason and Prerequisites

Program Error

Solution

For FTP adapter, please set the appropriate FTP timeout parameter in sender channel configurations. Existing timeout implementation has a bug while reading from Input. This patch solves this problem.

For JDBC adapter, we introduce a new advanced mode table parameter 'taskTimeout' parameter This is case-sensitive. This is to be configured in seconds and always greater than 0. Behavior is not known if you keep any value other than above.

XI 3.0 SP20 ADAPTER FRAMEWORK CORE 3.0

Support Package 20, Patch Level 8 (SAPXIAFC20P_8.SCA)

XI 3.0 SP19 ADAPTER FRAMEWORK CORE 3.0

Support Package 19, Patch Level 19 (SAPXIAFC19P_19.SCA)

XI 3.0 SP18 ADAPTER FRAMEWORK CORE 3.0

Support Package 18, Patch Level 23 (SAPXIAFC18P_23.SCA)

XI 3.0 SP17 ADAPTER FRAMEWORK CORE 3.0

Support Package 17, Patch Level 22 (SAPXIAFC17P_22.SCA)

PI 7.0 SP10 ADAPTER FRAMEWORK CORE 7.0

Support Package 10, Patch Level 11 (SAPXIAFC10P_11.SCA)

PI 7.0 SP11 ADAPTER FRAMEWORK CORE 7.0

Support Package 11, Patch Level 8 (SAPXIAFC11P_8.SCA)

PI 7.0 SP12 ADAPTER FRAMEWORK CORE 7.0

Support Package 12, Patch Level 3 (SAPXIAFC12P_3.SCA)

PI 7.0 SP13 ADAPTER FRAMEWORK CORE 7.0

Support Package 13, Patch Level 3 (SAPXIAFC13P_3.SCA)

The archives and the support package stack guide can be found on the SAP service Marketplace as described in SAP Note 952402.

Regards,

"

[Former Member]

Vitor,

Is it the latest poll time or something in past...

if in past than that is the catch..even if polling is correct message is not forming up..can you check the jdbc logs at visual admin level..

Regards,

[Former Member]

Vitor,

Is it polling on a a continuous basis...?

if yes at what time interval..?

and are you getting continuous blank message links as per the polling interval at RWB ?

i mean to say if polling interval is 5 sec are you getting blank message links after every 5 secs...?

Regards,

[Former Member]

Santosh, the pooling interval is 60 seconds. There isn´t any blank message, only "Process Started" in RWB.

Thanks

[prateek]

By any chance, is it possible that two similar communication channels are trying to poll same database table?

Regards,

Prateek

[Former Member]

Is ur update query working fine at sql level..

[Former Member]

Yes, it works.

[Former Member]

don't restart java stack but you can restart jdbc adapter only @ visual admin..!!!!!

[dharamveer_gaur2]

Hi

Now Check your generated SQL query format is correct? follow the process mentioned below.

At runtime you can find the genereated sql statements by doing configuraitons in Receiver JDBC adapter.

In the JDBC Receiver adapter you have the Advanced Properties .

Over there enter the following

left column logSQLStatement

right column true

To see the query created ..

Login to adapter monitoring ..select the relevant jdbc adapter.

Now when any message is processed by the jdbc adapter in adapter monitoring at that time you will see a message link. When you click on that link a new window will open. In that window if you click on page down you will get to see the sql statement generated by the jdbc

[Former Member]

Dharamveer, the generated SQL query format is right because this interface was working fine. It stoped working when an exception happened in mapping and now even after fixing the mapping, no records are being picked from database. The only way to restore this is by restarting the Java, but I can´t do this because this can happen in Prd system.

Thanks

[Former Member]

Hi Vitor,

Have you used the taskTimeout parameter in Advance mode parameters

Note 1083488 - XI FTP/JDBC sender channel stop polling\

regards

Ganga

[Former Member]

Santosh and gangaprasad, I´ve already tried both solutions...but the lock persists.

[Former Member]

Vitor,

If select query is working fine at sql level than your select query is perfect at xi but since your update query is having some issue so transaction is rolled back as select and update will work together. an dwillbe consideed as one transaction .

so kindly check update query also at sql prompt .

I guess you have proper rights on table ..means update rights .kindly check with dba .

if it also works fine :

than kindly perform following steps :

1. Change the isolation level to serializable .

2 Start stop the Communication channel .

Regards,

[Former Member]

Unni, exactly.

VJ, when I release the lock, another lock is created and so on. It never releases and the message never processes.

Thanks

[Former Member]

HI Victor

This is a problem with adapter locking. If you are facing continues locking problem then make modification to your JDBC connector service and change it to enqueue locking instead of database locking.

Change the Isolation level to READ_UNCOMMITTED.

Check for implementing enqueue lock implementation guide on SDN.

It will help as enqueue locks can be removed with the expiration of lifetime.

Thanks

Gaurav

[Former Member]

Hi Gaurav.

I´ve already tried to change the isolation level to NONE, but the error persists. About the enqueue locking, how can I do that ? Can you provide any link ?

Thanks

Vitor Zaninotto

[Former Member]

Hi Victor,

Did you find any entries in the Locking Entries in VA.

JDBC Adapter might got locked.Check the Visual Admin

Check in the Visual Admin >Cluster> "LOCKING ADAPTER"

we get a option of Display Locks .Check for an entry with

NAME : $XIDBAD.JDBC2XI

reset the locks and restart the CC ,now it works fine.

Note 1083488 - XI FTP/JDBC sender channel stop polling indefinitely(04/04S)

regards

Ganga

[Former Member]

gangaprasad, I already did that and when I release the lock, another lock is created. It never releases and the message never processes.

Santosh, the user have the correct rights. I already tried all isolation levels, already tried to stop and start communication channel many times, but no success.

Thanks

[Former Member]

Hi,

Can you post both the queries, select & update from the communication channel

regards

Unni

[Former Member]

SELECT NAME FROM ASD_O012_01 WHERE CD_STATUS_SSA IS NOT null AND CD_STATUS_SSA <> 'S' AND CD_STATUS_SSA <> 'E' AND TEST IS NOT null

UPDATE ASD_O012_01 SET CD_STATUS_SSA = 'R' WHERE CD_STATUS_SSA IS NOT null AND CD_STATUS_SSA <> 'S' AND CD_STATUS_SSA <> 'E' AND TEST IS NOT null

[Former Member]

So are you saying if you run the query,

SELECT NAME FROM ASD_O012_01

WHERE

CD_STATUS_SSA IS NOT null

AND CD_STATUS_SSA 'S'

AND CD_STATUS_SSA 'E'

AND TEST IS NOT null

in the DB you get some records, but XI is not picking? Please confirm.

[VijayKonam]

With my prev experience.. this is absolutely to do with the locking of the adapter in Visual Admin only. Once you cleat that and inactivate and activate the channel.. it should work normally.

VJ