A P P E N D I X  C

Utilities

This appendix provides information about SunVTS utilities. Topics include:


VTSreportgenerate Utility

The vtsreportgenerate script generates a standard report to stdout in a verbose and summary format. Output can also be generated in a report text file.

This script generates a standard SunVTS Test Report that provides actionable details on the findings of SunVTS after a stress test. Diagnosed faults and actionable system messages are provided to the user.

Two types of reports can be generated.

Syntax for VTSreportgenerate

vtsreportgenerate [ -s | --summary] [ -v |--verbose ]
[ -o |--output
file ]


-s, --summary

Summary Report

-v, --verbose

Verbose Report

-o, --output

Output filename

-h, --help

This usage message


SunVTS Summary Test Report

This is the default report generated with no options passed to vtsreportgenerate. The SunVTS Summary test report will include the following categories of information.:


# vtsreportgenerate [ -s | --summary ]

SunVTS Test Report Header

The header information will provide data on the test run itself:


SunVTS Summary Test Report
SunVTS Version 7.0 PS5 Build4
Start Time:   Fri Jan 16 10:00:54 IST 2009
End Time:     Fri Jan 16 11:00:54 IST 2009
Hostname:     diag061.goa7.sun.com

The Test Status is "PASS" if the error count is not greater than 0. Else, "FAIL".

Test Status

Each Test status is displayed with a Pass or Fail depending on the outcome of the test that was run.


Test Status

 

Disk

PASS [ | FAIL ]

Environment:

PASS [ | FAIL ]

Interconnect:

PASS [ | FAIL ]

Media:

PASS [ | FAIL ]

Memory:

PASS [ | FAIL ]

Network:

PASS [ | FAIL ]

Processor:

PASS [ | FAIL ]


Faults and Suspects

Any faults triggered due to a SunVTS Test Run are displayed in the standard Fault Management Architecture format (FMA, or Solaris Predictive Self Healing).

The existing Solaris fault messages are displayed in the output of FMA and the fields and actions are reported from FMA. An example fault report is shown below.


Solaris Fault Example Using fmadm (1M)

Faults Detected
--------------- ------------------------------------  -------------- ---------
TIME            CACHE-ID                              MSG-ID         SEVERITY
--------------- ------------------------------------  -------------- ---------
Oct 21 07:00:18 cdfe927d-f0d5-c7a6-c4b2-b71eb7b9ccea  PCIEX-8000-8R  Critical 
Fault class : fault.io.pciex.device-invreq
Affects     : dev:////pci@3c,0/pci10de,376@f
                  degraded but still in service
FRU         : "MB" (hc://:product-id=Sun-Fire-X4540:chassis-id=0000000000:server-id=localhost/motherboard=0)
                  faulty
 
Description : The transmitting device sent an invalid request.
              Refer to http://sun.com/msg/PCIEX-8000-8R for more information.
Response    : One or more device instances may be disabled
Impact      : Loss of services provided by the device instances associated with
              this fault
Action      : Ensure that the latest drivers and patches are installed.
              Otherwise schedule a repair procedure to replace the affected
              device(s).  Use fmdump -v -u <EVENT_ID> to identify the devices
              or contact Sun for support.

High Priority SunVTS Messages

High priority error messages generated by SunVTS are in their own section.


SunVTS Messages
04/28/09 16:10:50 diag062 SunVTS7.0ps6build5: VTSID 1368 Media.cddvdtest.ERROR rdsk/c4t0d0s0: "DKIOCGMEDIAINFO failed

syslog(3C) Messages

SunVTS stress tests generate high priority error messages in syslog that would be useful to a user to gain insight into causes of failures. Such messages can also be used for empirical analysis and correlations among service actions and human intuitive diagnosis.

Any syslog(3C) messages of a severity like this:

LOG_ERR, LOG_CRIT, LOG_ALERT and LOG_EMERG

will be reported as part of the SunVTS Summary Test Report. See syslog(3C) for more information on the severity of messages.


Syslog Messages
Oct 21 07:00:12  rpcbind: [ID 362760 daemon.alert] no value for config/enable_tcpwrappers (entity not found). Using default "false"
Oct 21 07:00:12  rpcbind: [ID 362760 daemon.alert] no value for config/enable_tcpwrappers (entity not found). Using default "false"
Oct 21 07:00:12  rpcbind: [ID 362760 daemon.alert] no value for config/enable_tcpwrappers (entity not found). Using default "false"
Oct 21 07:00:12  rpcbind: [ID 362760 daemon.alert] no value for config/verbose_logging (entity not found). Using default "false"

SunVTS Verbose Test Report

The verbose SunVTS Test Report provides verbose forms of data from the Summary.:


# vtsreportgenerate  -v | --verbose 

The header information will provide data on the test run itself.


SunVTS Verbose Test Report
SunVTS Version 7.0 PS5 Build4
Start Time:   Fri Jan 16 10:00:54 IST 2009
End Time:     Fri Jan 16 11:00:54 IST 2009
Hostname:     diag067.goa7.sun.com

Test Status

Each Test status will be displayed with a PASS or FAIL depending on the outcome of the test that was run. Additionally, the verbose report will include details on individual tests within the test and the number of "passes" and "errors".


Test Status
 
	 Disk                       passes: 22 errors: 0
	 c1t1d0p0(disktest)       passes: 44 errors: 0
	 c1t0d0(disktest)         passes: 31 errors: 0
	 c1t1d0(disktest)         passes: 35 errors: 0
	 
	 Environment                passes: 0  errors: 2118
	 BMC(bmcenvironment)      passes: 0  errors: 2118
	 
	 Interconnect               passes: 28 errors: 0
	 bus(bustest)             passes: 169 errors: 0
	 c1t1d0p0(disktest)       passes: 56 errors: 0
	 c1t0d0(disktest)         passes: 33 errors: 0
	 c1t1d0(disktest)         passes: 51 errors: 0
	 bge0(nettest)            passes: 977 errors: 0
	 
	 Media                      passes: 0  errors: 779
	 c0t1d0(cddvdtest)        passes: 0  errors: 779
	 
	 Memory                     passes: 12 errors: 0
	  Pass Level 1
	  kmem(vmemtest)           passes: 1  errors: 0
	  kmem(vmemtest).1         passes: 1  errors: 0
	  mem(pmemtest)            passes: 18 errors: 0
	  mem(pmemtest).1          passes: 18 errors: 0
	  kmem(vmemtest)           passes: 1  errors: 0
	  kmem(vmemtest).1         passes: 1  errors: 0
	  mem(pmemtest)            passes: 19 errors: 0
	  mem(pmemtest).1          passes: 19 errors: 0
	  kmem(vmemtest)           passes: 1  errors: 0
	 ....
	 ....
	 Network                    passes: 884 errors: 0
	 bge0(nettest)            passes: 884 errors: 0
	 bge1(netlbtest)          passes: 3951 errors: 0
	 
	 Processor                  passes: 2  errors: 0
	 
	  Pass Level 1
	  CPU(cputest)             passes: 610 errors: 0
	  FPU(fputest)             passes: 16 errors: 0
	  l2sram(l2sramtest)       passes: 1  errors: 0
	  l1cache(l1dcachetest)    passes: 1  errors: 0
	  CPU(cputest)             passes: 3  errors: 0
	  l1cache(l1dcachetest)    passes: 1  errors: 0
	  FPU(fputest)             passes: 2  errors: 0
	 
	  Pass Level 2
	  system(systest)          passes: 1  errors: 0
	  system(systest)          passes: 1  errors: 0
	  l1cache(l1dcachetest)    passes: 2  errors: 0
	  l2sram(l2sramtest)       passes: 1  errors: 0
	  FPU(fputest)             passes: 77 errors: 0
	  CPU(cputest)             passes: 151 errors: 0
	  FPU(fputest)             passes: 91 errors: 0
	  l2sram(l2sramtest)       passes: 1  errors: 0

Faults and Suspects

Any faults triggered due to SunVTS Test Run are displayed in the standard FMA format.

The existing Solaris Fault messages are predicated on the current output of FMA and the fields and actions reported from FMA. An example of an FMA fault is shown below.


Fault Example using fmadm (1M)

Faults Detected
--------------- ------------------------------------  -------------- ---------
TIME            CACHE-ID                              MSG-ID         SEVERITY
--------------- ------------------------------------  -------------- ---------
Oct 21 07:00:18 cdfe927d-f0d5-c7a6-c4b2-b71eb7b9ccea  PCIEX-8000-8R  Critical 
Fault class : fault.io.pciex.device-invreq
Affects     : dev:////pci@3c,0/pci10de,376@f
                  degraded but still in service
FRU         : "MB" (hc://:product-id=Sun-Fire-X4540:chassis-id=0000000000:server-id=localhost/motherboard=0)
                  faulty
 
Description : The transmitting device sent an invalid request.
              Refer to http://sun.com/msg/PCIEX-8000-8R for more information.
Response    : One or more device instances may be disabled
Impact      : Loss of services provided by the device instances associated with
              this fault
Action      : Ensure that the latest drivers and patches are installed.
              Otherwise schedule a repair procedure to replace the affected
              device(s).  Use fmdump -v -u <EVENT_ID> to identify the devices
              or contact Sun for support.

SunVTS Log Messages

SunVTS generates error messages after a stress test. These also aid root cause analysis of faulty HW. These messages are reported as part of the SunVTS Verbose Test Report:


01/06/09 10:19:14 diags197.goa78.sun.com SunVTS6.4: VTSID 8011 disktest.FATAL sda: "Failed lock mtab semaphore. "semop" system call failure, errmsg: Invalid argument." Probable_Cause(s): <disktest instances exceeds system semaphore operation limitation (default system limit for seminfo_semmnu = 30)><System software error>  Recommended_Action(s): <Add the line "set semsys:seminfo_semmnu=0x100" to your /etc/system file and reboot the machine> <If the problem persists, call your authorized Sun service provider.>
 
01/06/09 10:44:24 diags197.goa78.sun.com SunVTS6.4: VTSID 8011 disktest.FATAL sdc: "Failed lock mtab semaphore. "semop" system call failure, errmsg: Invalid argument." Probable_Cause(s): <disktest instances exceeds system semaphore operation limitation (default system limit for seminfo_semmnu = 30)><System software error>  Recommended_Action(s): <Add the line "set semsys:seminfo_semmnu=0x100" to your /etc/system file and reboot the machine> <If the problem persists, call your authorized Sun service provider.>

syslog(3C) Messages

SunVTS stress tests generate high priority error messages in syslog that help you gain insight into failure causes. The messages can also be used for empirical analysis and correlations among service actions and human intuitive diagnosis. Any syslog(3C) messages of a severity like this:

LOG_NOTICE, LOG_WARNING, LOG_ERR, LOG_CRIT, LOG_ALERT and LOG_EMERG

will be reported as part of the SunVTS Verbose Test Report. See syslog(3C) for more information on severity of messages. An example is shown below:


Apr 24 18:29:00 diag062 automountd[398]: [ID 476599 daemon.warning] sr1-cnwd03-02: no NFS service
 
Apr 27 18:12:17 diag062 gnome-session[733]: [ID 702911 daemon.warning] WARNING: Could not connect to ConsoleKit: Could not get owner of name ’org.freedesktop.ConsoleKit’: no such name
 
Apr 27 18:21:25 diag062 gnome-session[2182]: [ID 702911 daemon.warning] EggSMClient-WARNING: Desktop file ’/usr/share/gnome/autostart/desktop-print-management-applet.desktop’ has malformed Icon key ’print-manager.png’(should not include extension)
 
Apr 27 18:30:25 diag062 gnome-session[2182]: [ID 702911 daemon.warning] WARNING: Could not connect to ConsoleKit: Could not get owner of name ’org.freedesktop.ConsoleKit’: no such name
 
Apr 27 18:30:36 diag062 scsi: [ID 107833 kern.warning] WARNING: /pci@0,0/pci108e,4843@1d,7/storage@3/disk@0,0 (sd2):
 
Apr 27 18:30:36 diag062 scsi: [ID 107833 kern.warning] WARNING: /pci@0,0/pci108e,4843@1d,7/storage@3/disk@0,0 (sd2):