15.8. IMS diagnostic data collection for WAIT/HANG conditions
The difficulty in capturing specific data related to a wait or hang in a multi-address space environment is knowing which address spaces to dump since it is often difficult to determine who is actually causing the wait/hang. Often people misinterpret a loop for a hang, and while meaningful processing in both cases is not possible, the key indicator is the lack of CPU usage. A hang will usually be confirmed by no CPU activity, or no I/O activity; whereas a loop will often show high, and even excessive CPU activity.
While it is necessary to capture a dump of primary IMS address spaces (preferably using the previously discussed procedures), the wait/hang could be related, for example, ...
Get z/OS Diagnostic Data Collection and Analysis now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
