On a frequent basis, our +arch disk space holding Oracle archive logs is getting full due to which it is impacting our DB server.
Release : 12.0
Component : CA Workload Automation AE (AutoSys)
During the database failover process, the primary scheduler will mark the contents of ujo_alamode.int_val to value '-2' for the 'DB' row. While the value is -2 in the table, the application server will run the below query in an infinite loop and could cause filling up transaction logs or archive logs of the databases.
update ujo_alamode set str_val = (select str_val from ujo_alamode where type='AUTOSERV') where type = 'AUTOSERV'
Database transaction logs or archive logs will fill up exponentially, causing database downtime.
This has been accepted as a bug and will be fixed in 12.1 release.
The current workaround solution is to restart all the AutoSys services on all server machines.
https://support.broadcom.com/nl/web/ecx/problemdetails?problemno=17929&componentcode=ATSYS