[PATCH] oeqa/runtime: Added one runtime testcase in connman.

[PATCH] oeqa/runtime: Added one runtime testcase in connman.

[Daniel Istrate]

(testcase 223) Check that only one connmand runs in background.

Signed-off-by: Daniel Istrate <daniel.alexandrux.istrate@intel.com>
---
 meta/lib/oeqa/runtime/connman.py | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/meta/lib/oeqa/runtime/connman.py b/meta/lib/oeqa/runtime/connman.py
index cc537f7..b040400 100644
--- a/meta/lib/oeqa/runtime/connman.py
+++ b/meta/lib/oeqa/runtime/connman.py
@@ -28,3 +28,26 @@ class ConnmanTest(oeRuntimeTest):
         if status != 0:
             print self.service_status("connman")
             self.fail("No connmand process running")
+
+    @testcase(223)
+    def test_only_one_connmand_in_background(self):
+        """
+        Summary:     Only one connmand in background
+        Expected:    There will be only one connmand instance in background.
+        Product:     BSPs
+        Author:      Alexandru Georgescu <alexandru.c.georgescu@intel.com>
+        AutomatedBy: Daniel Istrate <daniel.alexandrux.istrate@intel.com>
+        """
+
+        # Make sure that 'connmand' is running in background
+        (status, output) = self.target.run(oeRuntimeTest.pscmd + ' | grep [c]onnmand')
+        self.assertEqual(0, status, 'Failed to find "connmand" process running in background.')
+
+        # Start a new instance of 'connmand'
+        (status, output) = self.target.run('connmand')
+        self.assertEqual(0, status, 'Failed to start a new "connmand" process.')
+
+        # Make sure that only one 'connmand' is running in background
+        (status, output) = self.target.run(oeRuntimeTest.pscmd + ' | grep [c]onnmand | wc -l')
+        self.assertEqual(0, status, 'Failed to find "connmand" process running in background.')
+        self.assertEqual(1, int(output), 'Found {} connmand processes running, expected 1.'.format(output))
-- 
2.1.4

Re: [PATCH] oeqa/runtime: Added one runtime testcase in connman.

[Randy Witt]

Hi Daniel and Alex,


On 07/15/2015 07:35 AM, Daniel Istrate wrote:
> (testcase 223) Check that only one connmand runs in background.
>
> Signed-off-by: Daniel Istrate <daniel.alexandrux.istrate@intel.com>
> ---
>   meta/lib/oeqa/runtime/connman.py | 23 +++++++++++++++++++++++
>   1 file changed, 23 insertions(+)
>
> diff --git a/meta/lib/oeqa/runtime/connman.py b/meta/lib/oeqa/runtime/connman.py
> index cc537f7..b040400 100644
> --- a/meta/lib/oeqa/runtime/connman.py
> +++ b/meta/lib/oeqa/runtime/connman.py
> @@ -28,3 +28,26 @@ class ConnmanTest(oeRuntimeTest):
>           if status != 0:
>               print self.service_status("connman")
>               self.fail("No connmand process running")
> +
> +    @testcase(223)
> +    def test_only_one_connmand_in_background(self):
> +        """
> +        Summary:     Only one connmand in background
> +        Expected:    There will be only one connmand instance in background.
> +        Product:     BSPs
> +        Author:      Alexandru Georgescu <alexandru.c.georgescu@intel.com>
> +        AutomatedBy: Daniel Istrate <daniel.alexandrux.istrate@intel.com>
> +        """
> +
> +        # Make sure that 'connmand' is running in background
> +        (status, output) = self.target.run(oeRuntimeTest.pscmd + ' | grep [c]onnmand')
> +        self.assertEqual(0, status, 'Failed to find "connmand" process running in background.')
> +
> +        # Start a new instance of 'connmand'
> +        (status, output) = self.target.run('connmand')
> +        self.assertEqual(0, status, 'Failed to start a new "connmand" process.')

If multiple instances of connmand can run, is that not a bug it connmand rather 
than in our configuration? I'm asking because it seems like this would be more 
suited as a connman test that would be part of ptest.

Otherwise, why not test all daemons for multiple instances?

Re: [PATCH] oeqa/runtime: Added one runtime testcase in connman.

[Burton, Ross]

[-- Attachment #1: Type: text/plain, Size: 512 bytes --]

On 16 July 2015 at 18:46, Randy Witt <randy.e.witt@linux.intel.com> wrote:

> If multiple instances of connmand can run, is that not a bug it connmand
> rather than in our configuration? I'm asking because it seems like this
> would be more suited as a connman test that would be part of ptest.
>

The test doubles as "did connman start" (for images in which connman is
expected to be present).  With sufficiently broken kernels, connman will
exit on startup (this test exists for a reason...)

Ross

[-- Attachment #2: Type: text/html, Size: 934 bytes --]