Collecting details of issues by using logs, dumps, and traces
The problem determination tools that are provided with IBM Spectrum Scale™ are intended to be used by experienced system administrators who know how to collect data and run debugging routines.
You can collect various types of logs such as GPFS™ logs, protocol service logs, operating system logs, and transparent cloud tiering logs. The GPFS™ log is a repository of error conditions that are detected on each node, as well as operational events such as file system mounts. The operating system error log is also useful because it contains information about hardware failures and operating system or other software failures that can affect the IBM Spectrum Scale system.
The IBM Spectrum Scale system also provides a system snapshot dump, trace, and other utilities that can be used to obtain detailed information about specific problems.
- Time stamp in GPFS log entries
The time stamp in a GPFS log entry indicates the time of an event. - Logs
This topic describes various logs that are generated in the IBM Spectrum Scale. - Setting up core dumps on a client RHEL or SLES system
No core dump configuration is set up by IBM Spectrum Scale by default. Core dumps can be configured in a few ways.
For information about SLES core dumps, see SUSE Knowledgebase
article: How to obtain application core dumps: https://www.suse.com/support/kb/doc/?id=3054866
- Configuration changes required on protocol nodes to collect core dump data
To collect core dumps for debugging programs in provided packages, these system configuration changes need to be made on all protocol nodes in the cluster. - Setting up an Ubuntu system to capture crash files
This is the procedure for setting up an Ubuntu system for capturing crash files and debugging CES NFS core dump. - Trace facility
The IBM Spectrum Scale system includes many different trace points to facilitate rapid problem determination of failures.