Re: [Fuego] [PATCH v3] initscripts: Add test cases for commands of initscripts.

See comments inline below.

> -----Original Message-----
> From: [email protected] [mailto:fuego-
> [email protected]] On Behalf Of Wang Mingyu
> Sent: Wednesday, September 18, 2019 9:21 PM
> To: [email protected]
> Subject: [Fuego] [PATCH v3] initscripts: Add test cases for commands of
> initscripts.
> 
> This test set is used to check che inittab command halt, ifdown-ifup and
che-> the

> sendsigs.
> 
> Signed-off-by: Wang Mingyu <[email protected]>
> ---
>  tests/Functional.initscripts/fuego_test.sh    | 18 ++++++++++++
>  .../initscripts_test.sh                       |  4 +++
>  tests/Functional.initscripts/spec.json        |  6 ++++
>  .../tests/initscripts_halt.sh                 | 13 +++++++++
>  .../tests/initscripts_ifdown-ifup.sh          | 29 +++++++++++++++++++
>  .../tests/initscripts_sendsigs.sh             | 13 +++++++++
>  6 files changed, 83 insertions(+)
>  create mode 100644 tests/Functional.initscripts/fuego_test.sh
>  create mode 100755 tests/Functional.initscripts/initscripts_test.sh
>  create mode 100644 tests/Functional.initscripts/spec.json
>  create mode 100644 tests/Functional.initscripts/tests/initscripts_halt.sh
>  create mode 100644 tests/Functional.initscripts/tests/initscripts_ifdown-
> ifup.sh
>  create mode 100644 tests/Functional.initscripts/tests/initscripts_sendsigs.sh
> 
> diff --git a/tests/Functional.initscripts/fuego_test.sh
> b/tests/Functional.initscripts/fuego_test.sh
> new file mode 100644
> index 0000000..b3fb94a
> --- /dev/null
> +++ b/tests/Functional.initscripts/fuego_test.sh
> @@ -0,0 +1,18 @@
> +function test_pre_check {
> +    assert_has_program ifup
> +    assert_has_program ifdown
> +}
> +
> +function test_deploy {
> +    put $TEST_HOME/initscripts_test.sh $BOARD_TESTDIR/fuego.$TESTDIR/
> +    put -r $TEST_HOME/tests $BOARD_TESTDIR/fuego.$TESTDIR/
> +}
> +
> +function test_run {
> +    report "cd $BOARD_TESTDIR/fuego.$TESTDIR;\
> +    ./initscripts_test.sh"
> +}
> +
> +function test_processing {
> +    log_compare "$TESTDIR" "0" "TEST-FAIL" "n"
> +}
> diff --git a/tests/Functional.initscripts/initscripts_test.sh
> b/tests/Functional.initscripts/initscripts_test.sh
> new file mode 100755
> index 0000000..dd5ce37
> --- /dev/null
> +++ b/tests/Functional.initscripts/initscripts_test.sh
> @@ -0,0 +1,4 @@
> +#!/bin/sh
> +for i in tests/*.sh; do
> +    sh $i
> +done
> diff --git a/tests/Functional.initscripts/spec.json
> b/tests/Functional.initscripts/spec.json
> new file mode 100644
> index 0000000..6560900
> --- /dev/null
> +++ b/tests/Functional.initscripts/spec.json
> @@ -0,0 +1,6 @@
> +{
> +    "testName": "Functional.initscripts",
> +    "specs": {
> +        "default": {}
> +    }
> +}
> diff --git a/tests/Functional.initscripts/tests/initscripts_halt.sh
> b/tests/Functional.initscripts/tests/initscripts_halt.sh
> new file mode 100644
> index 0000000..5fadd19
> --- /dev/null
> +++ b/tests/Functional.initscripts/tests/initscripts_halt.sh
> @@ -0,0 +1,13 @@
> +#!/bin/sh
> +
> +# In target, verify that the file is executable.
> +# option: none
> +
> +test="halt"
> +
> +if [ -x /etc/init.d/halt ]
> +then
> +    echo " -> $test: TEST-PASS"
> +else
> +    echo " -> $test: TEST-FAIL"
> +fi

Would be nice to actually test if the program can halt the machine here.
But that could be an enhancement for the future.

> diff --git a/tests/Functional.initscripts/tests/initscripts_ifdown-ifup.sh
> b/tests/Functional.initscripts/tests/initscripts_ifdown-ifup.sh
> new file mode 100644
> index 0000000..ed1d2eb
> --- /dev/null
> +++ b/tests/Functional.initscripts/tests/initscripts_ifdown-ifup.sh
> @@ -0,0 +1,29 @@
> +#!/bin/sh
> +
> +# In target, run command ifdown/ifup.
> +# option: none
> +
> +test="ifdown/ifup"
> +
> +if [ ! -f /etc/sysconfig/network-scripts/ifcfg-lo ]
> +then
> +    echo " -> $test: ifcfg file is not exist."
is not exist -> was not found

> +    echo " -> $test: TEST-SKIP"
> +    exit 0
> +fi
> +
> +if ifdown lo
> +then
> +    echo " -> $test: ifdown lo succeeded."
> +else
> +    echo " -> $test: ifdown lo  failed."
> +    echo " -> $test: TEST-FAIL"
> +    exit 0
> +fi

Should check here that the 'lo' interface (loopback)
is no longer present in the output of 'ifconfig'

I added some lines to do this.

> +
> +if ifup lo
> +then
> +    echo " -> $test: TEST-PASS"
> +else
> +    echo " -> $test: TEST-FAIL"
> +fi
> diff --git a/tests/Functional.initscripts/tests/initscripts_sendsigs.sh
> b/tests/Functional.initscripts/tests/initscripts_sendsigs.sh
> new file mode 100644
> index 0000000..633acad
> --- /dev/null
> +++ b/tests/Functional.initscripts/tests/initscripts_sendsigs.sh
> @@ -0,0 +1,13 @@
> +#!/bin/sh
> +
> +# In target, verify that the file is executable.
> +# option: none
> +
> +test="sendsigs"
> +
> +if [ -x /etc/init.d/sendsigs ]
> +then
> +    echo " -> $test: TEST-PASS"
> +else
> +    echo " -> $test: TEST-FAIL"
> +fi
> --
> 2.17.1

OK - just checking that the program is present and executable
is not very deep testing.

However, I'll accept this as a starting point for this test.

I'm applying this, but will re-word a few items, and I have added some
extra checking for the correct status of the loopback device,
 as described above.
 -- Tim