[Libvirt-cim] [PATCH] [TEST]Update RPCS/04 to validate that the Network child pool can be created through the providers

[Kaitlin Rupert]

yunguol at cn.ibm.com wrote:
> # HG changeset patch
> # User Guolian Yun <yunguol at cn.ibm.com>
> # Date 1239868524 25200
> # Node ID 860c994006a12104618e29bf051730993568bcc1
> # Parent  4ec367c94c356de7fac5a19ffe215c316d0cdcd1
> [TEST]Update RPCS/04 to validate that the Network child pool can be created through the providers
>
>
> Follow up patch will valide Disk child pool creation and verification in the same tc
>   
libvirt supports several different disk pool types. Not all of these are 
supported in libvirt-cim yet, but they will be in the future. I would 
put the disk pool verification in a different test, otherwise, this test 
might become to cluttered / confusing.
> Tested for KVM with current sources
> Signed-off-by: Guolian Yun<yunguol at cn.ibm.com>
>
> diff -r 4ec367c94c35 -r 860c994006a1 suites/libvirt-cim/cimtest/ResourcePoolConfigurationService/04_CreateChildResourcePool.py
> --- a/suites/libvirt-cim/cimtest/ResourcePoolConfigurationService/04_CreateChildResourcePool.py	Wed Apr 08 02:22:53 2009 -0700
> +++ b/suites/libvirt-cim/cimtest/ResourcePoolConfigurationService/04_CreateChildResourcePool.py	Thu Apr 16 00:55:24 2009 -0700
> @@ -52,32 +52,84 @@
>   
Can you update the comment block at the top of the test?

> +
> +def verify_pool(pool_list, poolname):
> +    status = PASS
> +    if len(pool_list) < 1:
> +        logger.error("Returen %i instances, expected at least one instance",
>   
Typo here.. this should be "Return"
> +                     len(pool_list))
> +        return FAIL
> +    
> +    for i in range(0, len(pool_list)):
> +        ret_pool = pool_list[i].InstanceID
> +        if ret_pool == poolname:
> +            break
> +        elif ret_pool != poolname and i == len(pool_list)-1:
> +            logger.error("Can not find expected pool")
> +            status = FAIL
>   
This for loop is a little odd. Why not set status to FAIL at the top of 
the function. If you find the pool you are looking for, set status to 
PASS and break from the loop. If you don't find the pool, then status 
will be FAIL at the end of the loop.
> +    elif curr_cim_rev >= libvirt_cim_child_pool_rev: 
> +        nprasd = get_typed_class(options.virt, 
> +                                 'NetPoolResourceAllocationSettingData')
> +        np_id = 'NetworkPool/%s' % testpool
> +        iname = CIMInstanceName(nprasd,
> +                                namespace = 'root/virt',
> +                                keybindings = {'InstanceID':np_id})
> +        logger.info('iname is %s', iname)
> +        nrasd = CIMInstance(nprasd, path = iname,
> +                            properties ={
> +                                         "Address" : "192.168.0.30",
> +                                         "Netmask" : "255.255.255.0",
> +                                         "IPRangeStart" : "192.168.0.31",
> +                                         "IPRangeEnd" : "192.168.0.57",
> +                                         "ForwardMode":"route eth1"})
>   
Since this test will be network pool specific, can you also test the 
other types of network pools 
(http://libvirt.org/formatnetwork.html#examples)
> +        try:
> +            rpcs_conn.CreateChildResourcePool(ElementName=testpool, 
> +                                              Settings=[nrasd.tomof()])
> +        except pywbem.CIMError, details:
> +            logger.error("Invoke CreateChildResourcePool() error")
> +            logger.error(details)
>              return FAIL
> -     
> -    logger.error("The execution should not have reached here!!")
> -    return FAIL
> +         
> +        np = get_typed_class(options.virt, 'NetworkPool')
> +        netpool = EnumInstances(options.ip, np)
> +        status = verify_pool(netpool, np_id)
> +
> +        destroy_netpool(options.ip, options.virt, testpool)
>   
Also be sure to undefine the pool as well.