Fixed Issues - 11.3.6 SP3

This article details the issues fixed in CA Workload Automation AE Release 11.3.6 SP3.
wcc114sp5p
This article details the issues fixed in CA Workload Automation AE Release 11.3.6 SP3.
Contents
FREQUENT DEADLOCKS IN SCHEDULER
Problem Number: 4272
The scheduler displayed frequent deadlock messages while running on MSSQL; thus leading to EVENT_HDLR_ERROR alarms and causing scheduling problems. This issue was fixed in Release 11.3.6 SP2.
The application server displayed frequent deadlock messages while running on MSSQL; thus leading to EVENT_HDLR_ERROR alarms. This issue is fixed in Release 11.3.6 SP3.
ASM DISKGROUP VERIFICATION CHECK FAILS DURING INSTALL
Problem Number: 4282
During installation, if you select the ASM (Automatic Storage Management) option, the Data Disk Group not found error occurred and prevented from continuing with the installation.
SCHEDULER TRUNCATES ISDBGACTIV TRACES
Problem Number: 4294
When the UNIX scheduler runs under ISDBGACTIV=JOB traces and it evaluates the start of jobs across different time zones, the scheduler begins to output just the file and line number but no other debug information.
DECOUPLE CONFIGURATION OF AGENT LISTENER FROM RECV_TIMEOUT
Problem Number: 4315
The RECV_TIMEOUT environment variable is used primarily to configure the CA Workload Automation AE clients to wait longer for the application server to process client commands and respond. Over time, the RECV_TIMEOUT environment has been associated with other time outs that are not related to request/response interactions. One example is the scheduler agent listener, which also uses RECV_TIMEOUT to reduce the time to wait for incoming data to arrive from an agent that has established a connection over the network. To avoid conflicting uses of RECV_TIMEOUT, a new environment variable named MSGSVC_LISTEN_TIMEOUT is introduced.
 If CA Technical Support previously instructed you to set the RECV_TIMEOUT variable to a low value to time out poor socket connections sooner due to frequent network delays and you upgraded from Release 11.3.6 SP2 or lower, unset the RECV_TIMEOUT variable and set the MSGSVC_LISTEN_TIMEOUT variable to the same low value instead. Restart the scheduler or application server for the changes to take effect. For more information about the related network communication environment variables, see Configuring CA Workload Automation AE.
There is no relationship between RECV_TIMEOUT and MSGSVC_LISTEN_TIMEOUT. After upgrading to Release 11.3.6 SP3 or higher, you may set both without consequence.
CHASE/AGENT INTEGRATION FOR DELAYED STATUS
Problem Number: 4323
Part of the chase utilities task is to contact each agent where a job is running and confirm with the agent that the job is still actually in the RUNNING state. However, in times of intense load (lots of events to process, from many agents), the scheduler can sometimes briefly fall behind processing incoming events from the agents (that is, SU/FA) and subsequently the agent cannot report the jobs completion status in a timely manner. At that moment, if chase is run, it raised an alarm indicating that the job should be running on the agent but it is in fact not running.
As part of this fix, the CA Workload Automation AE application server now requests the patched agent for the list of jobs that have completed execution but whose statuses have not yet been communicated to the scheduler. This fix requires CA WA Agent for UNIX, Linux, Windows or i5/OS 11.3 SP4 Incremental 2. For all agents earlier than the specified version, the chase utility contin