17.5.2 Data to be collected in the event of trouble (in UNIX)
- Organization of this subsection
(1) Log information about the OS
The following table lists the log information about the OS to be collected.
Type of information |
Outline |
Default file name |
Collected by the jpcras command |
Collected by the jpcwras command |
---|---|---|---|---|
System log |
syslog |
|
Y#1 |
Y#1 |
Process information |
List of processes |
N/A |
Y |
Y |
System file |
hosts file |
/etc/hosts |
Y |
Y |
/etc/inet/ipnodes#2 |
Y#3 |
N |
||
services file |
/etc/services |
Y |
Y |
|
OS information |
System information |
-- |
Y |
Y |
Patch information |
-- |
Y |
Y |
|
Kernel information |
-- |
Y |
Y |
|
Version information |
-- |
Y |
Y |
|
Network status |
-- |
Y |
Y |
|
Environment variables |
-- |
Y |
Y |
|
Host name |
-- |
Y |
Y |
|
Firewall information(Linux) |
-- |
Y |
N |
|
Dump information |
core file#4 |
-- |
Y |
N |
- Legend:
-
Y: Can be collected
N: Cannot be collected
--: There is no default file.
- #1
-
This cannot be collected in a system where output is not set to the default path or file name. In this case, use a different method to collect this information.
- #2
-
The /etc/inet/ipnodes file is available only under Solaris. Collect it together with the /etc/hosts file.
- #3
-
This file can only be collected with the jpcras command included in PFM - Manager 09-00 or PFM - Base 09-00 or later.
- #4
-
Under HP-UX 11i V3 (IPF), you can use the coreadm command to rename the core file. If the core file is renamed to a file name that does not start with core, the jpcras command cannot collect it. In this case, collect the file manually.
(2) Performance Management information
You need to collect the information about Performance Management listed below. In the case of a network error, you also need to collect applicable files from the connection-destination machine. The Performance Management information required to be collected is shown below.
Type of information |
Outline |
Default file name |
Collected by the jpcras command |
---|---|---|---|
Common message log |
Message log output from Performance Management (sequential file method) |
/opt/jp1pc/log/jpclog{01|02}#1 |
Y |
Message log output from Performance Management (wraparound file method) |
/opt/jp1pc/log/jpclogw{01|02}#1 |
Y |
|
Common message log (logical host use) |
Message log output from Performance Management (sequential file method) |
environment-directory/jp1pc/log/jpclog{01|02}#1 |
Y |
Message log output from Performance Management (wrap-around file method) |
environment-directory/jp1pc/log/jpclogw{01|02}#1 |
Y |
|
Configuration information |
Each configuration information file |
|
Y |
Database information |
Name Server |
|
Y |
Master Manager |
|
Y |
|
Master Store |
|
Y |
|
View Server |
|
Y#7 |
|
Agent Store and Remote Monitor Store |
|
Y#7 |
|
Agent Store (health check agent) |
|
Y#7 |
|
Database information (logical host use) |
Name Server |
|
Y |
Master Manager |
|
Y |
|
Master Store |
|
Y |
|
View Server |
|
Y#7 |
|
Agent Store and Remote Monitor Store |
|
Y#7 |
|
Agent Store (health check agent) |
|
Y#7 |
|
Trace log |
Trace information of each service in the Performance Management program |
N/A#4 |
Y |
Install log#5 |
Standard log of Hitachi Program Product Installer |
|
N |
- Legend:
-
Y: Can be collected
N: Cannot be collected
N/A: There is no default file.
- #1
-
For details on the output format of the log files, see 16.5 Detecting problems by linking with the integrated system monitoring product.
- #2
-
xxxx indicates the service key of each PFM - Agent or PFM - RM. For details on the service keys of each PFM - Agent or PFM - RM, see the naming rules in an appendix of the JP1/Performance Management Planning and Configuration Guide.
- #3
-
For a PFM - Agent or PFM - RM monitoring an application program which can start multiple service sets on a single host, there is a directory for each instance.
- #4
-
For details on the trace log destination directory, see 17.4.2 Details on log information.
- #5
-
Collect this information if the installation fails.
- #6
-
Collect this file when installation has failed but the file exists.
- #7
-
Collected only when a directory name, a service key, or the all or data option is specified. If the option is omitted, the file is not collected.
(3) PFM - Web Console information
You need to collect the information about PFM - Web Console listed below. In the case of a network error, you also need to collect applicable files from the connection-destination machine. The PFM - Web Console information that needs to be collected is shown below.
Type of information |
Outline |
Default file name |
Collected by the jpcwras command |
---|---|---|---|
Web server information |
Server log of Web server or Web container |
N/A |
Y |
Operation status log#1 |
Message log output from PFM - Web Console |
/opt/jp1pcwebcon/log/jpcwtracelog-file-number#2.log |
Y |
Message logs output from PFM - Web Console commands |
|
Y |
|
File information |
List of PFM - Web Console installation files |
N/A |
Y |
Configuration Info |
PFM - Web Console configuration information |
/opt/jp1pcwebcon/conf/*.* /opt/jp1pcwebcon/sample/conf/*.* |
Y |
- Legend:
-
Y: Can be collected
--: There is no default file.
- #1
-
For details on the output format of the operation status log, see the chapter describing the log information in the manual JP1/Performance Management Reference.
- #2
-
The log file number is the number of output log files, starting from 1.
- Operation status log output
-
The output size of the operation status log (excluding the log for the jpcrpt command) can be set by using logFileSize × logFileNumber in the initialization file (config.xml).
- #3
-
For details about the log output format of the jpcrpt command, see 17.4.2(1)(a) Log output format of the jpcrpt command.
(4) Operation information
You need to collect the following information about the operation being performed when an error occurs:
-
Details of the operation
-
Time the error occurred
-
Machine configuration (for example, the OS version, host name, and PFM - Manager and PFM - Agent or PFM - RM configuration)
-
Whether the error is replicable
-
When logged on from PFM - Web Console, the Performance Management user name at logon
-
Arguments specified in a command if a problem occurred while the command was executed
(5) Error information
You need to obtain the following error information:
-
Messages output to the console, if the error occurred during command execution
(6) Docker environment information (Linux)
If you are using Performance Management in a Linux Docker environment, you have to collect the following Docker environment information:
Type of information |
Outline |
Collected by the jpcras command |
---|---|---|
Docker environment information |
Docker version |
Y |
Docker container list |
Y |
|
Process within the Docker container |
Y |
|
Docker container information |
Y |
- Legend:
-
Y: Can be collected