1. Tool introduction

LTP is a system test suite developed by the Linux Test Project. It develops a test combination based on the utilization statistics of system resources to provide enough pressure for the system. Judge the stability and reliability of the system through stress testing. Stress test is a destructive test, that is, the operation of the system under abnormal and overloaded conditions. It is used to evaluate how the system will operate when the maximum load is exceeded. It is a test of the system's ability to withstand a certain load intensity under normal conditions.

2. Parameter introduction

-a email_to Send all reports via email to this email address
-c NUM_PROCS Run LTP under additional background CPU load
-C FAILCMDFILE Command file containing all failed test cases
-d TMPDIR specifies the folder where temporary files are created
-D Run LTP under the additional background load of auxiliary storage (separated by commas)
-h Help Help, print all available options
-g HTMLFILE creates an additional HTML output document
-N Run all network tests
-n Run all network LTP in the background
-p Log file in readable format
-l Record the test results in the log file.
-o output file Redirect the test output to a file.
-s PATTERN only run test cases that match the pattern
-t DURATION Execute the test suite for the given duration. Example:
-t 60s =60 seconds
-t 45m=45 minutes
-t 24h = 24 hours
-t 2d =2 days
-v Print more detailed output to the screen.
-x INSTANCES instance to run multiple instances of this test suite

3. Test execution

Networking environment
LTP tool download address: https://github.com/linux-test-project/ltp
Compile and install all tests: make make install
Run all tests: Cd /opt/ltp ./runltp
For specific usage, check the document in the download address
[https://blog.csdn.net/cui841923894/article/details/81463048]

4 Command run

/opt/ltp/runltp -p -l /tmp/resultlog.20051221 -d /tmp -o /tmp/ltpscreen.20051221 -t 24h

After running, use tail -f /tmp/ltpscreen.20051221 to monitor the running status. If the operation continues, do not shut down and wait for the result within 24 hours (modify the target time as needed).

testscripts/ltpstress.sh -d sar -l log -t 72 -i 60

It means to run ltpstress continuously for 72 hours and use sar to record system activities every 60 seconds. The log file of sar is /tmp/ltpstress.xxx.sar, and the test log file of ltpstess is /tmp/ltpstress.xxx.log (it is necessary to use ltpstress.sh Turn on the rsh service of the machine and allow root to log in to the machine)

5 Result analysis

1. Analyze the /tmp/ resultlog.20051221 log

2. Analyze the /tmp/ltpscreen.20051221 log

3. Analyze /tmp/resultlog.20051221: filter out all FAILs to get a complete testcase of all FAILs.
The method is as follows: use sort to sort the FAIL items, and then use uniq to exclude duplicates and output to a file:
grep FAIL /tmp/ resultlog.20051221 | sort | uniq >failcase.txt

　 So far, the failcase.txt obtained is the name of all FAIL testcases. Pay attention to the analysis of the reasons for the failure of the case, and come to the conclusion: it is a configuration problem (all the case failed in the 72h test, and there was no success record), or a stability problem (both failure and success). And add the conclusion in failcase.txt for easy viewing.
　The stress test part can refer to:
　https://blog.csdn.net/a1317480843/article/details/80006028?utm_medium=distribute.pc_relevant.none-task-blog-OPENSEARCH-1&depth_1-utm_source=distribute.pc_relevant.none-task-blog-OPENSEARCH-1