During events like memory leaks, enough information is not available from a core and or gstack to provide information for the issue. Currently we have to install debug libs and set special conditions to retrieve enough information for the vendor to find the issue. I large and heavy use environments, this can be bad to clients and disrupt the environment even more. And sometimes it hard to reproduce the event after it has happened or you have to wait a week for the situation to occur again.

Please provide a easier way to set a debug mode to get heaps and or all the information the vendor needs to diagnose the issue without a restart. A possible "heap" mode setting to turn on in real time?

by: Donald G. | over a year ago | Other