Free Trial

Safari Books Online is a digital library providing on-demand subscription access to thousands of learning resources.


Share this Page URL
Help

Chapter 19: Logs for problem determinati... > 19.1: Job logs and system log - Pg. 214

19.1 Job logs and system log WebSphere is a collection of server instances working together. Each server instance is made up of a controller region and some number of servant regions. Each of these regions is an address space that, as with any other address space, has a job log that you can view through SDSF. If you run JES3, then you might be using EJES, the JES2 spool. The system log (SYSLOG) is a record of the console messages. WebSphere messages issued to SYSLOG also show up in the job log. In most problem cases, these logs show information about exceptions, information about abnormal situations, or simply warning messages in the system. These are normally the first logs that you should examine when software problems occur. 19.1.1 When to use system log and job logs The JES2 spool writes the information to the job logs of the address spaces for the different regions. There are several types of regions that are of interest: For each server instance, there is one controller region and 0 or more servant regions (except for the daemon, for which there is only a controller region). ­ Controller regions, to put it simply, handle communication, receiving requests from clients and sending back responses. ­ Servant regions are given the requests to process, so they do the actual work. You might also have address spaces from local clients and your HTTP server. Now, given all the job logs available, how do you know which ones have useful information? The answer depends on what is taking place at the time of the error.