Fuego Object Details
Here is a list of objects:
- Docker image
- Docker container
- Fuego board
- Fuego test
- test source
- base script
- test specs
- test criteria
- test reference
- results elements:
- test results (aggregate data)
- test charts
- Fuego testplan
- Fuego run
- test results (individual)
- test logs
- test artifacts
- Jenkins node - references a Fuego board
- Jenkins job - references a Fuego board, test and spec
- Jenkins build - references a Fuego run
Docker objects [edit section]
docker image [edit section]
A docker image is built when Fuego is installed. It contains a Debian distribution, and tools, libraries, and packages needed to perform test execution and support the Fuego system.
docker container [edit section]
This is runtime instance of the image, that has been instantiated and has persistent state. The container is the object that is actually executed by docker, and which Jenkins and Fuego run inside of, in order to perform tests.
Fuego objects [edit section]
Fuego board [edit section]
A board is defined by:- fuego board file: /fuego-ro/boards/<name>.board
- defines: IPADDR, SSH_PORT, LOGIN, BOARD_TESTDIR, and many more
- defines: TOOLCHAIN
- defines: DISTRIB
- defines: TRANSPORT
- toolchain file: /fuego-ro/toolchains/<TOOLCHAIN>-tools.sh
- defines: AR, ARCH, CC, and many other variables for the toolchain
Also called 'Device' by Jenkins, and 'NODE_NAME' by Jenkins
- dynamic data:
- fuego board dynamic variables: /fuego-rw/boards/<name>.vars
Fuego test [edit section]
A test is defined by:
- fuego test definition files:
- in /fuego-core/engine/tests/<test_name>
- fuego_test.sh
- tarball
- spec.json
- in /fuego-core/engine/tests/<test_name>
- see Test Specs and Plans
- reference.json
- parser.py
- test p/n logs - see Log files
- chart_config.json
- test.yaml file
- Test documentation
Fuego test results [edit section]
Fuego stores results for tests in a number of files.
- fuego individual results files (from one run):
- in the "run" directory (/fuego-rw/logs/<test_name>/<run_id>
- structured results ("run.json")
- log files - see Log files
- in the "run" directory (/fuego-rw/logs/<test_name>/<run_id>
- fuego aggregate results files:
- in /fuego-rw/logs/<test_name>
- results.json
- flat_plot_data.txt
- flot_chart_data.json
- in /fuego-rw/logs/<test_name>
test elements table [edit section]
Here is the same information in table formThe "test directory" is /fuego-core/engine/tests/{test_name}
A test consists of the following files or items:
File or item ^ | format ^ | location ^ | description ^ | test type ^ |
p/n logs | text | test directory | Are logs with the results (positive or negative) parsed out, for determination of test pass/fail | functional only |
spec.json | JSON | test directory | Has groups of variables (and their values) that can be used with this test | all |
test.yaml | YAML | test directory | Has the "manifest" for packaging a test | all |
tarfile | tar format | test directory | Has the source code for the test program | all |
test plan(s) | JSON | inside a batch test, or in fuego-core/overlays/testplans-default.json | Has the testplan(s) for this test | all |
reference.json | Fuego-specific | test directory | Has the threshold values and comparison operators for benchmark metrics measured by the test | benchmarks only |
base script | shell script | test directory | Is the shell script that implements the different test phases in Fuego | all |
criteria.json | JSON | test directory, or user-specified | Provides the criteria for evaluating the results of the test. For example, it can indicate that some testcase results may be ignored. For benchmarks, this indicates threshold values that are used to determine pass or fail for the test. | |
parser.py | python | test directory | Python program to parse benchmark metrics out of the test's log, and provide a dictionary of results to Fuego for processing | benchmarks only |
patches | patch format | test directory | Zero or more patches to customize the test program (applied during the unpack phase | all |
chart_config.json | JSON | test directory | provides the names of metrics to be plotted for this benchmark | benchmarks only |
(See also Adding a new test, Benchmark parser notes, parser.py, reference.log
Fuego run [edit section]
A test run is defined by:
- fuego log files:
- /fuego-rw/logs/<test_name>/<board>.<spec>.<build_id>.<build_number>/
- testlog.txt
- syslog.before.txt
- syslog.after.txt
- consolelog.txt
- testlog.p.txt, testlog.n.txt
- (coming soon: fres.json)
- /fuego-rw/logs/<test_name>/<board>.<spec>.<build_id>.<build_number>/
- benchmark results:
- /fuego-rw/logs/<test_name>
- plot.data
- plot.png
- <testname>.info.json
- <testname>.<metric1>.json
- with a file for each metric
- /fuego-rw/logs/<test_name>
Fuego testplan [edit section]
- This file indicates a set of tests to run, and associated test variants (test specs) and other parameters (build variables from the spec)
- /fuego-core/engine/overlays/testplans/testplan_<name>.json
See Test Specs and Plans for more information.
Jenkins objects [edit section]
Jenkins node [edit section]
- This is the Jenkins object to represent a Fuego board
- /var/lib/jenkins/nodes/<node_name>/config.xml
Jenkins job [edit section]
- This is the Jenkins object to represent a unit of test execution
- It refers to a node, spec, and test
- The job name is composed of: <node_name>.<spec>.<test_type>.<test_name>
- jenkins job file:
- /var/lib/jenkins/jobs/<job_name>/config.xml - expected to be static
- provides form for: 'Reboot', 'Rebuild', 'Target_Cleanup'
- specifies shell command sequence to cause build
- this references the script for the test, at:
- $FUEGO_CORE/engine/tests/$TESTDIR/$TEST_NAME.sh
- this references the script for the test, at:
- /var/lib/jenkins/jobs/<job_name>/nextBuildNumber - changes for each run
- /var/lib/jenkins/jobs/<job_name>/config.xml - expected to be static
Jenkins build [edit section]
- jenkins files:
- /var/lib/jenkins/jobs/<job_name>/builds/<build_id>/
- build.xml
- changelog.xml
- log - this is the console log
- /var/lib/jenkins/jobs/<job_name>/builds/<build_id>/