Where to view detailed log and trace information in a TSAMP cluster

Technote (FAQ)


Question

Diagnostic logs and traces specific to an IBM Tivoli System Automation for Multiplatforms (TSAMP) environment

Answer

There are 4 principle files that are very relevant to TSAMP related activity:

1. Syslogs

2. Error Report (AIX only)

3. IBM.RecoveryRM trace_summary

4. IBM.GblResRM trace_summary

1. Syslogs:

Syslogs can be located anywhere but the syslog.conf is by default located in /etc/ directory. From this file you can tell where your syslog daemon is configured to write messages to. For example, checking the /etc/syslog.conf file, ignoring all lines that begin with "#" there is very little of real use here, for a sample:
daemon.debug;user.debug /var/log/syslog.out rotate time 1d files 14 

The above line writes to the file /var/log/syslog.out which should then be checked for current or recent events. Also, this is setup to rotate every day and keep the last 14 days of logs, those logs will be named syslog.1.out, syslog.2.out and so on with the lowest number being the most recently archived log files. In the syslog output you can view all of the logged messages from the control scripts for applications like DB2 and SAP.


2. Error Report (AIX only):
The error report is shown by typing "errpt" from the command line. This will give you a simple list of all the errors listed in the error report with a simple description for most of them. Using "errpt -a" will give a detailed list with quite a bit more useful information, for example:

errpt:
IDENTIFIER TIMESTAMP  T C RESOURCE_NAME  DESCRIPTION
3B16518D   0128102513 I S ConfigRM       The node is online in the domain indicat

errpt -a:
---------------------------------------------------------------------------
LABEL:          CONFIGRM_ONLINE_ST
IDENTIFIER:     3B16518D

Date/Time:       Mon Jan 28 10:25:04 2013
Sequence Number: 7540
Machine Id:      00CF67EE4C00
Node Id:         samp2
Class:           S
Type:            INFO
Resource Name:   ConfigRM

Description
The node is online in the domain indicated in the detail data.

Probable Causes
A user ran the 'startrpdomain' or 'startrpnode' commands.
The node rebooted while the node was online.
The configuration manager recycled the node through an offline/online
transition to resynchronize the domain configuration or to recover
from some other failure.

User Causes
A user ran the 'startrpdomain' or 'startrpnode' commands.
The node rebooted while the node was online.
The configuration manager recycled the node through an offline/online
transition to resynchronize the domain configuration or to recover
from some other failure.

        Recommended Actions
        None

Detail Data
DETECTING MODULE
...skipping...
---------------------------------------------------------------------------

3. IBM.RecoveryRM trace_summary:
IBM.RecoveryRM is the "brains" of TSAMP. It decides what, where, when, why a resource is started or stopped. To format and show the trace_summary file use the following to find the path and then use the rpttr command to format the file into text.

lssrc -ls IBM.RecoveryRM | grep trace_summary

For example the output can look like this:
/var/ct/v97_hadr/log/mc/IBM.RecoveryRM/trace_summary -> spooling not enabled

Now use the rpttr command to format the trace_summary file and write it to a file in your temp directory.
As root:
rpttr -odtic /var/ct/v97_hadr/log/mc/IBM.RecoveryRM/trace_summary > /tmp/trace.out

You can read the file in any text editor.

4. IBM.GblResRM trace_summary:
IBM.GblResRM is the global resource manager that actually runs the start/stop and monitor scripts as well as assigns IP Aliases on the nodes at the direction of IBM.RecoveryRM. To format and show the trace_summary file use the following to find the path and then use the rpttr command to format the file into text.

lssrc -ls IBM.GblResRM | grep trace_summary

For example the output can look like this:
/var/ct/v97_hadr/log/mc/IBM.GblResRM/trace_summary -> spooling not enabled

Now use the rpttr command to format the trace_summary file and write it to a file in your temp directory.
As root:
rpttr -odtic /var/ct/v97_hadr/log/mc/IBM.GblResRM/trace_summary > /tmp/trace.out

You can read the file in any text editor.

Additional Information: Samlog
A new command added in TSAMP 3.2.1 allows for the collection, format, merge and display of related log and trace files into one contiguous file that is ordered chronologically.

For collection of all trace_summary files for IBM.GblResRM and IBM.RecoveryRM on all nodes in the domain (the domain and nodes must be online) for the last day enter the following as root (with redirect to a file):
samlog 1> /tmp/samlog.out
For more information on this command see the man page or the reference guide for TSAMP.


Rate this page:

(0 users)Average rating

Document information


More support for:

Tivoli System Automation for Multiplatforms

Software version:

3.1, 3.2, 3.2.1, 3.2.2

Operating system(s):

AIX, Linux, Solaris

Reference #:

1625077

Modified date:

2013-02-11

Translate my page

Machine Translation

Content navigation