Log streams and expected output

Investigating the logging and tracing output for the application server is an excellent way to observe performance, diagnose problems, and gain a general understanding of how the application server is working within your environment. The expected output locations for logging and trace information can be different depending on the operating system on which the application server is running.

Note: 這個主題參照一或多個應用程式伺服器日誌檔。 此外,在分散式和 IBM® i 系統上,另外也建議您可以配置伺服器來使用「高效能可延伸記載 (HPEL)」日誌和追蹤基礎架構,而不使用 SystemOut.logSystemErr.log, trace.logactivity.log 檔案。HPEL 與原生 z/OS® 記載機能也可以一起使用。如果您使用 HPEL,則可以從伺服器設定檔 bin 目錄,利用 LogViewer 指令行工具來存取您所有的日誌和追蹤資訊。請參閱有關利用 HPEL 疑難排解應用程式的資訊,以取得更多使用 HPEL 的相關資訊。
[AIX Solaris HP-UX Linux Windows][IBM i]
Table 1. Log and trace output for distributed and IBM i. This table lists the expected output for logging and tracing output streams for distributed and IBM i operating systems
Log or trace stream Expected output - basic log and trace mode Expected output - HPEL log and trace mode
Java™ trace Trace.log. You must have tracing enabled.
  • HPEL trace repository.
  • HPEL TextLog*.log, when you enable this log.
Java logs
  • Trace.log, when you enable tracing.
  • Activity.log. The Activity.log file is also directed to the Showlog output.
  • SystemOut.log
  • HPEL log repository
  • HPEL TextLog*.log, when you enable this log.
System.out stream
  • Trace.log
  • SystemOut.log
  • HPEL log repository
  • HPEL TextLog*.log, when you enable this log.
System.err stream SystemErr.log
  • HPEL log repository
  • HPEL TextLog*.log, when you enable this log
cout (the C or C++ output stream) Native stdout Native stdout
cerr (the C or C++ error stream) Native stderr Native stderr
Figure 1. Distributed and IBM i - Basic log and trace mode. Distributed and IBM i - Basic log and trace modeGraphical representation of logging on distributed and IBM i
Figure 2. Distributed and IBM i - HPEL log and trace mode. Distributed and IBM i - HPEL log and trace modeGraphical representation of HPEL log and trace on distributed and IBM i
[z/OS]
Table 2. Log and trace output for z/OS. This table lists the expected output for logging and tracing output streams when the application server is installed on z/OS.
Log or trace stream Expected output - traditional log and trace mode Expected output - HPEL log and trace mode
Java trace SYSPRINT
避免困難 避免困難: Trace information, which includes events at the Fine, Finer and Finest levels, is written only to this output. If you do not enable diagnostic trace, setting the log detail level to Fine, Finer, or Finest does not affect the logged data.gotcha
  • HPEL trace repository
  • HPEL TextLog*.log, when you enable this log
Java logs (Audit level) Transformed into write-to-operator (WTO) output to the hard copy logging stream Same outputs as traditional log and trace mode, plus the following:
  • HPEL log repository
  • HPEL TextLog*.log, when you enable this log
Java logs (other levels) SYSOUT or your configured error stream Same outputs as traditional log and trace mod, plus the following:
  • HPEL log repository
  • HPEL TextLog*.log, when you enable this log
System.out SYSPRINT
  • HPEL log repository
  • HPEL TextLog*.log, when you enable this log
System.err SYSOUT
  • HPEL log repository
  • HPEL TextLog*.log, when you enable this log
cout (the C or C++ output stream) SYSPRINT SYSPRINT
cerr (the C or C++ error stream) SYSOUT SYSOUT
Native trace SYSPRINT
避免困難 避免困難: Trace information, which includes events at the Fine, Finer and Finest levels, is written only to this output. If you do not enable diagnostic trace, setting the log detail level to Fine, Finer, or Finest does not affect the logged data.gotcha
SYSPRINT
避免困難 避免困難: Trace information, which includes events at the Fine, Finer and Finest levels, is written only to this output. If you do not enable diagnostic trace, setting the log detail level to Fine, Finer, or Finest does not affect the logged data.gotcha
Native Message logs SYSOUT or your configured error stream SYSOUT or your configured error stream
Note: 這個主題參照一或多個應用程式伺服器日誌檔。 此外,在分散式和 IBM i 系統上,另外也建議您可以配置伺服器來使用「高效能可延伸記載 (HPEL)」日誌和追蹤基礎架構,而不使用 SystemOut.logSystemErr.log, trace.logactivity.log 檔案。HPEL 與原生 z/OS 記載機能也可以一起使用。如果您使用 HPEL,則可以從伺服器設定檔 bin 目錄,利用 LogViewer 指令行工具來存取您所有的日誌和追蹤資訊。請參閱有關利用 HPEL 疑難排解應用程式的資訊,以取得更多使用 HPEL 的相關資訊。
When considering the direction of log and trace streams, keep in mind the following acronyms and abbreviations:
cerr
C or C++ error steam
cout
C or C++ output steam
DD
Data Description statements
HFS
Hierarchical File System
JCL
Job Control Language
JES
Job Entry Subsystem
MVS™
Multiple Virtual Storage
WTO
Write-to-operator
WTOR
Write-to-operator with reply
Figure 3. z/OS - Traditional log and trace mode. z/OS - Traditional log and trace modeGraphical representation of traditional logging on z/OS
Figure 4. z/OS - HPEL log and trace mode. z/OS - HPEL log and trace modeGraphical representation of HPEL logging on z/OS

指出主題類型的圖示 概念主題



時間戳記圖示 前次更新: July 9, 2016 11:11
http://www14.software.ibm.com/webapp/wsbroker/redirect?version=cord&product=was-nd-mp&topic=cprf_logging
檔名:cprf_logging.html