Troubleshooting
Problem
Database host server jobs, QZDASOINIT or QZDASSINIT, may continue to run even after the client application that they are servicing had ended.
Cause
When a client application using the database host server is ended, the client TCP/IP stack will send a reset (RST) packet to the server to indicate that the connection has been forcefully ended. The server receives this and it is queued up in the server's TCP/IP queue. If the server job is in a normal waiting state, the message is processed immediately and the server job will end and log an appropriate message in the job log indicating that the connection was ended (typically a CPE3424). If the server job is already processing a previous request then it will not read the RST message until it has completed the processing of the current request. So if a client job has initiated a query or any other SQL process that takes a very long time to run, ending the client job will not cause the server job to end.
If the application is written in such a way that it can send an SQLCancel request to the server, that will cause the job to end sooner in most cases. Even with the SQLCancel function though, there are some cases where the job will not end immediately. For example, if the server job is running functions that are below the MI, the cancel will not be processed until those function are complete and control is returned to the operating system.
Related Information
Document Location
Worldwide
[{"Type":"MASTER","Line of Business":{"code":"LOB68","label":"Power HW"},"Business Unit":{"code":"BU070","label":"IBM Infrastructure"},"Product":{"code":"SWG60","label":"IBM i"},"ARM Category":[{"code":"a8m0z0000000CRyAAM","label":"Host Servers-\u003EDatabase"}],"ARM Case Number":"","Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"All Versions"}]
Was this topic helpful?
Document Information
Modified date:
09 September 2024
UID
ibm17167926