Currently, all directories in the /var/crash directory are checked for their times, comparing them to that of the time recorded by the kdump test when it ran the first portion, prior to rebooting/panic. The latest directory matching the naming convention to be created (taken from name that's generated when it's saved) is selected. If that is outside of the maximum allowed time, it will error. We are changing this in the following ways: - When saving the crash core file, a UUID is generated and used to create the directory to look in when the second portion of the kdump test runs (upon restarting the panicked machine). This will be unique per test run. - We know that there should only be one directory one level down from the new UUID directory. We will raise an error if more than one is detected. - We no longer raise an error if the directory's created time (taken from directory name) is outside of the range previously permitted. Instead, now, we will raise a warning and select the closest one (the only one in the directory, else it would error).
verified in redhat-certification-hardware-1.7.1-20150828.el7 Changed vmcore & crash directory's timestamp. kdump test completed successfully.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://rhn.redhat.com/errata/RHBA-2015-1709.html