| Solaris |
|
|
This document provides information about how to run Smoke Tests for Open HA Cluster. Smoke Tests include tests for HA Communication Infrastructure and Services Software in Open HA Cluster. For the most up-to-date build instructions, see the Building and Installing the Open HA Cluster Automated Test Suite page. This document contains the following sections:
Before you run the Smoke test suite, ensure that you have met the following prerequisites.
For failover and scalable tests:
For zone tests:
NOTE: You can use the same hostname and global raw device in all three types of tests (failover, scalable, and zone). Top
To determine whether the SCATE server process is already running, use one of the following commands:
$ **/usr/scate/svc/bin/scate status**
or
$ **pgrep scateserver**
# **/usr/scate/svc/bin/scate start**
Top === How to Create the Test Request File
$ **/usr/scate/tset/bin/smokerun**
The following is a sample screenshot of a smokerun invocation.
Name of a cluster node ? [phys-foo-1] **pono1**
Obtaining cluster configuration information from node pono1...
Specify Test Results Log Directory that is NFS exported and world writable.
Ex : /home/username/log (or) /net/machine/export/home/log
Test Results Log Directory? **/home/mytest/scate_logs**
Test Request Filename ? [/home/mytest/scate_logs/TestRequest/testrequest1.out]
~-------------
Test Category
~-------------
1. Smoke Tests for HA Communication Infrastructure
2. Smoke Tests for HA Services
Enter the number that corresponds to the tests you want to run.
To select more than one option, Enter a comma seperated list.
Enter your selection : **2**
1. Failover Services
2. Scalable Services
3. Private Interconnect
4. Non-global Zone
Enter the number that corresponds to the test you want to run.
To select more than one test, you can either enter a comma
seperated list (1,2,3...) or specify a range (1-3) or
enter "all" to run all the tests.
Enter your selection : **1**
***** Generating Entry for Failover Services *****
Some tests you selected require information about Network Hostname.
Please enter the hostname to be used for configuring Logical Host and Shared Address resource :**ono-1**
Getting node list from the cluster...
Please enter comma seperated node list on which "ono-1" will be hosted [pono1,pono2] : **//Return//**
Validating the Network hostname...
Getting the "NetIfList"(IPMP group info). Please wait...
********************************************************************************
Tests require one global raw device. Device Name should be of the form
/dev/md/foo-1/rdsk/dx for SVM or /dev/global/rdsk/dxsy for a H/W RAID device.
The device should NOT be mounted either globally or locally on any cluster node.
Also, the device should NOT be mounted at boot thru /etc/vfstab. If the device
is present in /etc/vfstab, set 'mount at boot' field to 'no' on all the nodes.
********************************************************************************
Enter the raw disk slice(s) and end the list with a blank line.
Disk name? **/dev/md/ono-1/rdsk/d0**
Disk name? **//Return//**
You selected 1 disks
Raw Disk 1 = /dev/md/ono-1/rdsk/d0
Is this correct? [y/n] ? **y**
You selected 1 disks
Raw Disk 1 = /dev/md/ono-1/rdsk/d0
Test Request File is in /home/mytest/scate_logs/TestRequest/testrequest1.out
Run the tests using smokerun -f /home/mytest/scate_logs/TestRequest/testrequest1.out
The following is an example test request file for Smoke Failover Services testing:
# SCATE Test Request File
# Generated: n1c33 Thu May 22 07:34:20 PDT 2008
LOGDIR=/home/mytest/scate_logs
CLUSTER
NODENAME=pono1
END
# Smoke - Failover Services
REQUEST cti
SEQUENCE=sanity
SCENARIO=failover
#SCENARIO=tc_mytest
#TEST_NUMBER="{1}"
#TEST_NUMBER="{1-3,5}"
#TEST_NUMBER=all
RUN_NEWCLI=TRUE
NETWORK_HOSTNAME_1=ono-1
NETWORK_HOSTNAME_ADAPTERS_1="sc_ipmp0@1,sc_ipmp0@2"
NETWORK_HOSTNAME_HOSTS_1=pono1,pono2
NETWORK_HOSTNAME_TYPE_1=0
RAW_DISK_1=/dev/md/ono-1/rdsk/d0
FS_TYPE=ufs
END
Top === How to Run the Smoke Test
The test request file is created under the log directory that is passed to the smokerun utility. In the previous example, the test request file name is /home/mytest/scate_logs/TestRequest/testrequest1.out. To run the test suite, invoke the following command on the test server as user scate.
$ **/usr/scate/svc/bin/smokerun -f ///path/////testrequestfile//**
Example:
$ **/usr/scate/svc/bin/smokerun -f /home/mytest/scate_logs/TestRequest/testrequest1.out** Executing the tests in /home/mytest/scate_logs/TestRequest/testrequest1.out... Info: (From Server) Host is rsh'able. Operation Success: (From Server) Job ID is 1 Operation Success: (From Server) Log directory is /home/mytest/scate_logs/05-22-08_09.57.48_pono1_1
Top === How to Run Individual Test Cases==== For CTI Requests
SCENARIO=tc_//testcase//
TEST_NUMBER="{//N//}"
For "{//N//}", specify the numbers of the tests to run by using the following format, including the quotation marks (" ") and the braces ({ }):
$ **smokerun -f //requestfile//**
SCENARIO=tc_//testcase// TEST_NUMBER="//N//"
For "//N//", specify the numbers of the tests to run by using the following format, including the quotation marks (" ") but without the braces you use in CTI request files:
$ **smokerun -f //requestfile//**
Top === How to Stop the SCATE Server Process
# **/usr/scate/svc/bin/scate stop**
Top === Logs and Reports for the Test Suites The test suites generate the following logs and reports.
///Logdirectory/////timestamp//_//clusternode//_//N///
520|0 1 00123575 1 43|May 22 08:10:35 : Sysid 1, file McExecCmd.c, line 129 : ex ecute_cmd: gosu ksh -c '/usr/cluster/bin/clrt register SUNW.tsvc2'
The following are the sample contents of the summary file:
TC /tests/tc_failover TP 1 tc_failover PASS TP 2 tc_failover PASS TP 3 tc_failover PASS
Top === Monitoring Live Logs
You can use the scatestatus -l command to monitor the progress of the smokerun test invocations. The live logs for individual test suites for 'cti' REQUEST entries in the test request file are available in the following location:
///Logdirectory/////timestamp//_//clusternode//_//N///1_//xxx///results/0001e/*.AAA
Use the tail -f command on the live log to show the log messages from the tests that are currently in progress.
If the tests are aborted for some reason, the logs from the tests that were in progress are stored under the live_logs/ subdirectory.
Top
Terms of Use
|
Privacy
|
Trademarks
|
Copyright Policy
|
Site Guidelines
|
Site Map
|
Help
Your use of this web site or any of its content or software indicates your agreement to be bound by these Terms of Use.
© 2012, Oracle Corporation and/or its affiliates.