Skip to main content

Hi All,

One of our customers having an issue like that they facing a deadlock in the PROD environment and the trace log files on Database Drive E experienced unexpected growth due to deadlock trace details, leading to the application becoming unavailable.

During further analysis, we have identified that the below query is causing the deadlock in the environment 

----- SQL Statement for this session -----
UPDATE MOBILE_SYNC_DATA_TAB SET SYNC_STATE=DECODE(SYNC_STATE, :B10 , :B10 , :B12 ), REFERENCED_OJBVERSION = :B11 , MODIFIED_COLUMNS = DECODE(SYNC_STATE, :B10 , MODIFIED_COLUMNS, MODIFIED_COLUMNS || '^' || :B9 ) WHERE APP_NAME = :B8 AND ENTITY = :B7 AND REFERENCED_OBJKEY = :B6 AND NOT (APP_NAME = NVL(:B5 , :B3 ) AND DEVICE_ID = NVL(:B4 , :B3 ) AND USER_ID = :B2 ) AND SYNC_STATE <> :B1

Can someone help on this to understand how this SQL query can be a reason for a deadlock? What are the possible ways to happen and how we can avoid it tp happening again?

 

IFS Applications 10

Hi,

We recently done a improvement (App10 UPD22) to minimize the chance to get the deadlock. But deadlock can still visible if the application does lot if frequent changes to Push type entities. So please review the Push Queue to identify such frequent changes.    What’s the 2nd SQL which blocks this DL condition? 


Reply