zhausong / pysphere Goto Github PK

Connect to pysphere using Kerberos / AD

From: https://groups.google.com/forum/#!topic/pysphere/jcFn6ZtwUgk

This should be done by implementing this function: 
http://www.vmware.com/support/developer/vc-sdk/visdk41pubs/ApiReference/vim.Sess
ionManager.html#loginBySSPI

What steps will reproduce the problem?
1. have a default proxy configured that does not allow access to the ESXi hosts.
2. configure urllib2 to not use a proxy:

proxy_handler = urllib2.ProxyHandler({})
opener = urllib2.build_opener(proxy_handler)
urllib2.install_opener(opener)

3. try uploading a file to a guest with vm.send_file()

What is the expected output? What do you see instead?

This causes send_file() to abort with an exception because the connection
to the ESX host can't be established.

What version of the product are you using? On what operating system?

pysphere 0.1.7, python 2.6.8, Windows 7

urllib2.urlopen() has accepted a request object since at least python 2.5.

Apply the following patch:

--- vi_virtual_machine.py,0     2012-10-09 18:04:01.576577700 +0200             
+++ vi_virtual_machine.py       2012-10-09 19:12:04.782569200 +0200             
@@ -1276,10 +1276,9 @@                                                          

             url = url.replace("*", urlparse(self._server._proxy.binding.url    
                                                                      ).hostname
)                                                                               
-            opener = urllib2.build_opener(urllib2.HTTPHandler)                 
             request = urllib2.Request(url, data=content)                       
             request.get_method = lambda: 'PUT'                                 
-            resp = opener.open(request)                                        
+            resp = urllib2.urlopen(request)                                    
             if not resp.code == 200:                                           
                 raise VIException("File could not be send",                    
                                   FaultTypes.TASK_ERROR)

Being able to delete a VM would be extremely useful. Is this a limitation on 
the vSphere SDK or on pysphere. If its pysphere I would be more than happy to 
write the module for it. 

What steps will reproduce the problem?
for ds, name in server.get_datastores().items():

What is the expected output? What do you see instead?
Traceback (most recent call last):
  File "./check_vsphere.py", line 195, in <module>
    sys.exit(main(sys.argv[1:]))
  File "./check_vsphere.py", line 56, in main
    return datastore(server, warning, critical)
  File "./check_vsphere.py", line 170, in datastore
    for ds, name in server.get_datastores().items():
  File "/usr/lib/python2.6/site-packages/pysphere/vi_server.py", line 189, in get_datastores
    return self._get_managed_objects(MORTypes.Datastore, from_mor)
  File "/usr/lib/python2.6/site-packages/pysphere/vi_server.py", line 759, in _get_managed_objects
    obj_type=mo_type)
  File "/usr/lib/python2.6/site-packages/pysphere/vi_server.py", line 687, in _retrieve_properties_traversal
    raise VIApiException(e)
pysphere.resources.vi_exception.VIApiException: [InvalidPropertyFault]:


What version of the product are you using? On what operating system?
Machine being queried by script is ESXi 3.5.0
Machine running script is CentOS 6

Please provide any additional information below.
Code works against newer ESXi machine.

see https://groups.google.com/forum/#!topic/pysphere/BXUJs5CTRIo

What steps will reproduce the problem?
1.Calling CloneVM with sync_run true results in Attribute error
2.
3.

What is the expected output? What do you see instead?
What I try to do: use a template to clone a VM from a template residing on 
another datastore.  Since the command executes flawless (I monitor it in 
vsphere), my argument are correct and so is the core section of the 
CloneVM_task in pysphere.  It goes wrong in the wrap-up of the call. I run the 
task against vCenter instead of the hypervisor (since only vcenter supports 
cloning)

IMHO there are two issues:
1. Minor: CloneVM returns different object types based on the value of one of 
its arguments, the sync_run parameter:  sync_run=true --> return 
VIVirtualMachine; sync_run=false --> returns VITask. While technically OK, it's 
a bit confusing and you need to be careful to test the value or sync_run to 
deal with the correct return type.  Other pysphere tasks do not have that 
behavior.

2. Major: IMHO the call to "return VIVirtualMachine(self._server, 
vi_task.get_result())" is not correct. vi_task.get_result() does not return a 
MOR but a VIProperty, hence you get the attribute non existing error for the 
call to instantiate the VIVirtualMachine class which expects a MOR.  The 
correct call would be:  "return VIVirtualMachine(self._server, 
vi_task.get_result()._obj)"

When I implemented this change my whole code ran OK.

What version of the product are you using? On what operating system?
Pysphere: R78.
Host: Windows 2008 R2, 64-bit
Python: 2.7 in 32 bit mode.
Vcenter and Vsphere: 5.0 U1

Please provide any additional information below.
My code:
vm.clone(name, sync_run=True, folder=None, resourcepool=rp_mor,
                                             datastore=ds_mor, host=host_mor, power_on=False)

Where I believe the problem is: vi_virtual_machine.py around line 530

            if sync_run:
                status = vi_task.wait_for_state([vi_task.STATE_SUCCESS,
                                                 vi_task.STATE_ERROR])
                if status == vi_task.STATE_ERROR:
                    raise VIException(vi_task.get_error_message(),
                                      FaultTypes.TASK_ERROR)
                return VIVirtualMachine(self._server, vi_task.get_result()) 

            return vi_task

Fist make the python 2.X version as compatible as possible with python 3 (e.g. 
by modifying ZSI's object types comparisons with the isinstance built-in)

Then run 2to3 and add the necessary fixes

What steps will reproduce the problem?
1. Instantiate a VIServer object and connect to vcenter server;
2. Try to get registered VM's using a resource pool as a filter;

What is the expected output? What do you see instead?
The list of vm's inside the resource pool.

What version of the product are you using? On what operating system?
Vmware Vcenter 5.0.

Client libs:

Ubuntu 12.10
Python 2.7
pysphere==0.1.7

What steps will reproduce the problem?
1. server.connect("vmhost","DOMAIN\user","passsword")

What is the expected output? What do you see instead?
No expected output.
pysphere.resources.vi_exception.VIApiException: [InvalidLoginFault]: Cannot 
complete login due to an incorrect user name or password.


What version of the product are you using? On what operating system?
0.1.8 on CentOS 6 64 and Cygwin64, currently the VMWare version is 5.1


Please provide any additional information below.
pysphere works well with non DOMAIN accounts, tested on 5.1 and 4.1 versions of 
VMWare

This is an enhancement proposal. It would be fantastic if we could add a 
__getitem__ method to VIVirtualMachine classes so that one might be able to 
index a VIVirtualMachine instance—like so: vm['memory_mb']—and get the 
pertinent information. This would essentially be a shorthand for the following:
``
vm.get_properties()['memory_mb']
``

What steps will reproduce the problem?
1. Set up your server connection
2. Get a VM entity
3. Call `vm.send_file(bla, bla)`

What is the expected output? What do you see instead?
I expect the local file to be uploaded to the guest.  Instead, I've been seeing 
stuff like this:

  File "/usr/lib/python2.7/site-packages/pysphere/vi_virtual_machine.py", line 1193, in send_file
    resp = opener.open(request)
  File "/usr/lib/python2.7/urllib2.py", line 400, in open
    response = self._open(req, data)
  File "/usr/lib/python2.7/urllib2.py", line 418, in _open
    '_open', req)
  File "/usr/lib/python2.7/urllib2.py", line 378, in _call_chain
    result = func(*args)
  File "/usr/lib/python2.7/urllib2.py", line 1215, in https_open
    return self.do_open(httplib.HTTPSConnection, req)
  File "/usr/lib/python2.7/urllib2.py", line 1174, in do_open
    h.request(req.get_method(), req.get_selector(), req.data, headers)
  File "/usr/lib/python2.7/httplib.py", line 958, in request
    self._send_request(method, url, body, headers)
  File "/usr/lib/python2.7/httplib.py", line 992, in _send_request
    self.endheaders(body)
  File "/usr/lib/python2.7/httplib.py", line 954, in endheaders
    self._send_output(message_body)
  File "/usr/lib/python2.7/httplib.py", line 812, in _send_output
    msg += message_body
UnicodeDecodeError: 'ascii' codec can't decode byte 0x89 in position 0: ordinal 
not in range(128)

What version of the product are you using? On what operating system?
I'm the latest svn revision, or r75 of `vi_virtual_machine.py`.  Only testing 
with Python 2.7 on Arch Linux.

Please provide any additional information below.
This problem only started appearing recently--I've been able to send files to 
guest systems without a problem since I started using pysphere a month or two 
ago.  I'm not sure what changed, exactly, but somewhere along the line it 
appears that the "msg" you see in the last line of the stack trace is a unicode 
object while the "message_body" from the same line is a str object.  The "msg" 
variable holds a bunch of header information, while the "message_body" contains 
the actual binary content of the file I'm trying to send.

I've watched/debugged this off and on for a few days, and it seems that the 
problem goes away if I simply force the URL to be a str object sometime before 
this line in vi_virtual_machine.py:

    request = urllib2.Request(url, data=content)

Perceive real timeout values as requested. IE: timeout=5

In [19]: time.clock() - start_time
Out[19]: 0.010006999999999877

In [20]: time.clock() - start_time
Out[20]: 0.013527999999999984

In [21]: start_time = time.time()

In [22]: time.time() - start_time
Out[22]: 10.152903079986572

In [23]: time.time() - start_time
Out[23]: 10.824795961380005

Add an optional parameter to the clone() method allowing specific settings to 
be passed (e.g. network settings, os settings, etc)

See https://groups.google.com/forum/#!topic/pysphere/QoiUxFQcH1M

See 
http://groups.google.com/group/pysphere/browse_thread/thread/2bba503390fbf2a5/97
7cbcfdde6268c7#977cbcfdde6268c7

e.g.

class VIAlarm:

  def list_alarms(...)

  def silence_alarm(...)

we might add some callback functionality where a user can provide a function to 
be called when a particular alarm are sets off

I tried to find information about overall architecture/design of pysphere in 
vain. My question is using pysphere, can I access every property, method, type 
available at vsphere web services level 
(http://www.vmware.com/support/developer/vc-sdk/visdk41pubs/ApiReference/index.h
tml)? Or am I restricted to what's exposed by pysphere client libraries and any 
new access would require enhacing the client lib?

For instance vijava gives complete (100%) access to underlying web services 
functionality

One important part of maintaining snapshots on modern vSphere is consolidating 
disks. pysphere should support this. Related VMWare docs 
http://pubs.vmware.com/vsphere-51/index.jsp?topic=%2Fcom.vmware.wssdk.apiref.doc
%2Fvim.VirtualMachine.html

The method returns a dictionary where keys are the resource pool paths and the 
values their MORs.

This path key is formed by the resource pool name and all its ancestors names 
separated by '/'. For example:
/Resources/AnotherRP/This Resource Pool

However there might be more than one root resource pool. The root rp (not shown 
from the VI Client) generally is named "Resources"
So that is not a unique key for the dictionary, in your case you have two rp 
named "/Resources" so adding the second to the dictionary overwrites the first.

I should probably invert the dictionary, i.e. make their MOR (which have to be 
unique) be the keys and the paths, their values).

This will impact on other functionality of pysphere that uses 
_get_resource_pools, so I'll have to check all the code.

Implement upload/download files from datastores.

This functionality must be implemented out of the vsphere sdk, they are just 
simple PUT/GET requests to the server.

See: http://communities.vmware.com/message/1540767

Dear pysphere-project-team!

First of all thank you for this really good library.

We are using a nagios environment to monitor our vmware host and their virtual 
machine. How about to get some current useful information like cpu utilization, 
cpu usage, mem usage, net usage, io statistics or storage information over this 
api. In my opinion the execution of a script to monitor a host is much than 
over the perl api. 

A well know perl-script is developed from op5 (check_vmware_api.pl). It is good 
but perl isn't so powerful than python xD

best regards

realburns 

We are looking at a way to boot the system into our own ISO image. So a way to 
add an ISO image from a datastore, as well as changing the boot-order of 
devices.

This second feature is strictly required as we could expect the order to be to 
boot from CD before disk, and simply connecting an image would boot it on 
reboot.

Is this currently already possible ?

I am adding a search feature into my app, and I'm looking for the best way to 
search "everything". Below is my current method. If there a more efficient 
method using the current library?

# search term
search = "esx"
pattern = re.compile(search, flags=re.IGNORECASE)

print "searching..."
items = server.get_hosts()
for item in items:
     if pattern.search(items[item]):
            print "host: %s" % items[item]

items = server.get_datastores()
for item in items:
    if pattern.search(items[item]):
        print "datastore: %s" % items[item]

items = server.get_clusters()
for item in items:
    if pattern.search(items[item]):
        print "cluster: %s" % items[item]

items = server.get_resource_pools()
for item in items:
    if pattern.search(items[item]):
        print "resource_pools: %s" % items[item]

vms = server.get_registered_vms()
for vm in vms:
    if pattern.search(vm):
        print "vm: %s" % vm

What steps will reproduce the problem?
class AEsxiServer(object):
    def __init__(self, servername, password,  username='root'):
        self.servername = servername
        self.username = username
        self.password = password
        self.vmlist = []
        self.vmstatusdict = {}
        self.vmmaclist = {}
        self.vmdict = {}
        self.inUseDict = {}

    def connect(self):
        print "connecting to %s" % self.servername
        self.server = VIServer()
        self.server.connect(self.servername, self.username, self.password)
        print "connected to %s" % self.servername
        print "Currently server version is: %s %s" \
            % (self.server.get_server_type(), self.server.get_api_version())

    def get_vmmaclist(self):
        for vmPath in self.get_vmlist():
            vm = self.server.get_vm_by_path(vmPath)
            vminfo = vm.get_properties()
            vmmacinfo = vminfo['mac_address']
            print vmmacinfo

def main():
    print "This is Esxi Server Object"
    S=AEsxiServer('xxxx','xxxx')
    S.connect()
    S.get_vmmaclist()

if __name__ == '__main__':
    main()

results:
print vmmacinfo # it will return "None"

S.get_vmmaclist() will return the following errors
Traceback (most recent call last):
  File "./AsterEsxiServer.py", line 133, in <module>
    main()
  File "./AsterEsxiServer.py", line 125, in main
    asterS.get_vmmaclist()
  File "./AsterEsxiServer.py", line 60, in get_vmmaclist
    vmmacinfo = vminfo['mac_address']
KeyError: 'mac_address'

What is the expected output? What do you see instead?
It should output mac address, but it prompts that don't have the key 
'mac_address'
BTW, I can get 

What version of the product are you using? On what operating system?
RHEL6.5

Please provide any additional information below.

After deleting a disk from a VM, I could not run server.get_vm_by_name

This alteration seems to fix this issue.

Here's a small patch that implements vSphere API Method "RelocateVM_Task" as 
member function of the VIVirtualMachine class:

 * pysphere.vi_virtual_machine.VIVirtualMachine.relocate()


> What version of the product are you using? On what operating system?
Against latest SVN checkout; tested on both Ubuntu 12.04 x64 and Windows XP x64.

-- Chris

PS: Thanks for the quality module.

What steps will reproduce the problem?
1. Used a Windows VM
2. Issued a remote command that takes a long time
   (you can simulate this by a batch program that just waits 
    around for 20 or more minutes)
3. Issue the VM's list_processes waiting for the exit_code to be set

What is the expected output? What do you see instead?
Expect the exit_code to eventually get set.  Instead, get the following error:
  File "test.py", line 154, in _getProcess
    processes = vm.list_processes() 
File 
"/root/Perforce/ToolsAutomation/PEToolsDepot/ToolsAutomation/VSAInstaller/python
/pysphere/vi_virtual_machine.py", line 1236, in list_processes
    raise VIApiException(e)
pysphere.resources.vi_exception.VIApiException: 
[GuestOperationsUnavailableFault]: The guest operations agent could not be 
contacted.

What version of the product are you using? On what operating system?
0.1.6 
Linux (CentOS 5.3)
(also failed same way under Windows, so it has more to do with the SOAP 
interface)

Please provide any additional information below.

Some vSphere managed objects besides Datastore should be wrapped under this 
VIDatastore class.

For e.g. HostDatastoreBrowser (see 
https://groups.google.com/forum/?fromgroups=#!topic/pysphere/v1QLEPRbAyg)


This class should also implement what is requested on Issue #12 
(https://code.google.com/p/pysphere/issues/detail?id=12)

Original mail from John Calcote. (see 
http://groups.google.com/group/pysphere/browse_thread/thread/81d97515a223e5e9#)

Hi Seba, 
I think I ran into a bug in pysphere - it may be in ZSI itself. I got 
the following TraceBack when attempting to call 
AcquireCIMServicesTicket on the VIServer instance with a managed 
object reference to a host that was currently disconnected: 
Traceback (most recent call last): 
  File "c:\fusion\fio-esx-agent-mgr-script.py", line 15, in <module> 
    fioesxagentmgr.main() 
  File "c:\users\jcalcote\dev\hg\management\src\util\vmware\fio-agent- 
cim\src\fioesxagentmgr\__init__.py", line 80, in main 
    esx_manager.poll_for_config_changes() 
  File "c:\users\jcalcote\dev\hg\management\src\util\vmware\fio-agent- 
cim\src\fioesxagentmgr\esxmgr.py", line 251, in 
poll_for_config_changes 
    ticket = self._get_host_services_ticket(mor) 
  File "c:\users\jcalcote\dev\hg\management\src\util\vmware\fio-agent- 
cim\src\fioesxagentmgr\esxmgr.py", line 150, in 
_get_host_services_ticket 
    response = self._server._proxy.AcquireCimServicesTicket(request) 
  File "c:\users\jcalcote\dev\hg\management\src\util\vmware\fio-agent- 
cim\eggs\pysphere-0.1.6-py2.7-win32.egg\pysphere\resources 
\VimService_services.py", line 1416, in AcquireCimServicesTicket 
    response = 
self.binding.Receive(AcquireCimServicesTicketResponseMsg.typecode) 
  File "c:\users\jcalcote\dev\hg\management\src\util\vmware\fio-agent- 
cim\eggs\pysphere-0.1.6-py2.7-win32.egg\pysphere\ZSI\client.py", line 
549, in Receive 
    return _Binding.Receive(self, replytype, **kw) 
  File "c:\users\jcalcote\dev\hg\management\src\util\vmware\fio-agent- 
cim\eggs\pysphere-0.1.6-py2.7-win32.egg\pysphere\ZSI\client.py", line 
467, in Receive 
    msg = FaultFromFaultMessage(self.local.ps) 
  File "c:\users\jcalcote\dev\hg\management\src\util\vmware\fio-agent- 
cim\eggs\pysphere-0.1.6-py2.7-win32.egg\pysphere\ZSI\fault.py", line 
253, in FaultFromFaultMessage 
    pyobj = ps.Parse(FaultType.typecode) 
  File "c:\users\jcalcote\dev\hg\management\src\util\vmware\fio-agent- 
cim\eggs\pysphere-0.1.6-py2.7-win32.egg\pysphere\ZSI\parse.py", line 
326, in Parse 
    return how.parse(self.body_root, self) 
  File "c:\users\jcalcote\dev\hg\management\src\util\vmware\fio-agent- 
cim\eggs\pysphere-0.1.6-py2.7-win32.egg\pysphere\ZSI\TCcompound.py", 
line 203, in parse 
    value = what.parse(c_elt, ps) 
  File "c:\users\jcalcote\dev\hg\management\src\util\vmware\fio-agent- 
cim\eggs\pysphere-0.1.6-py2.7-win32.egg\pysphere\ZSI\TCcompound.py", 
line 206, in parse 
    value = what.parse(c_elt, ps) 
  File "c:\users\jcalcote\dev\hg\management\src\util\vmware\fio-agent- 
cim\eggs\pysphere-0.1.6-py2.7-win32.egg\pysphere\ZSI\TC.py", line 
1455, in parse 
    pyobj = what.parse(elt, ps) 
  File "c:\users\jcalcote\dev\hg\management\src\util\vmware\fio-agent- 
cim\eggs\pysphere-0.1.6-py2.7-win32.egg\pysphere\ZSI\TCcompound.py", 
line 147, in parse 
    what = TypeDefinition.getSubstituteType(self, elt, ps) 
  File "c:\users\jcalcote\dev\hg\management\src\util\vmware\fio-agent- 
cim\eggs\pysphere-0.1.6-py2.7-win32.egg\pysphere\ZSI\schema.py", line 
296, in getSubstituteType 
    (self.type[0], self.type[1], pyclass), ps.Backtrace(elt)) 
TypeError: ("Substitute Type (urn:vim25, NoHost) is not derived from 
<class 'pysphere.resources.VimService_services_types.NoHost_Def'>", u'/ 
soapenv:Envelope/soapenv:Body/soapenv:Fault/detail/NoHostFault') 

It looks like it should have thrown a NoHost exception, but instead it 
throws a TypeError. 
Thanks in advance, 
John

Methods triggering tasks returns different object types depending on the value 
of the sync_run parameter which might be confusing. These should probably be 
implemented on separate methods.

If this is implemented, the sync_run parameter should still be supported for a 
few versions (showing a deprecation warning).

Pros:
 - clarity, a method will only return a specified type of result

Cons:
 - lots of new methods will be added. E.g. clone_async (or clone_task), revert_to_snapshot_async (revert_to_snapshot_task), etc. This might be even more confusing
 - handle backwards compatibility

Related to: https://code.google.com/p/pysphere/issues/detail?id=20

What steps will reproduce the problem?
1. vm = server.get_vm_by_name('some-existing-vm')
2. len(vm.get_snapshots()) # f.i 3
3. vm.refresh_snapshot_list()
4. len(vm.get_snapshots()) # 6 - duplicated

worse case (my case :) ):
1. vm = server.get_vm_by_name('some-existing-vm') # vm has 4 snapshots
2. vm.create_snapshot('12/12/2011') # snapshot will be 'refreshed' transparently
3. # now i'm removing snapshots over KEEP_OLD_SNAPSHOTS=4
4. i lost all my snapshots in that keys instead of oldest one

What is the expected output? What do you see instead?
 * snapshots list should be cleared and next filled with current state


What version of the product are you using? On what operating system?
 * pysphere 0.1.5
 * python 2.6
 * debian squeeze


Please provide any additional information below.

in file 
http://code.google.com/p/pysphere/source/browse/trunk/pysphere/vi_virtual_machin
e.py at line 1430:
 * snapshots are appended into self._root_snapshots
 * but self._root_snapshots is only cleared while creating new VIVirtualMachine in __init__
 * i think it should by cleared before "for root_snap in self.properties.snapshot.rootSnapshotList:
"

see https://groups.google.com/forum/?fromgroups=#!topic/pysphere/4XPkfP9z0TA

See attachement

Thanks

What steps will reproduce the problem?

    1. Do `vm.get_status()`
    2. Log into ESXi server using vSphere Client, look in Recent Tasks

What is the expected output? What do you see instead?

    I don't expect to have an error in the log, but instead I see the attached file.

    ("Create task collector" --- "The operation is not supported on the object.")

What version of the product are you using? On what operating system?

    pysphere==0.1.6
    VMware ESXi 4.1.0

Please provide any additional information below.

    Despite the error, functionally it appears to work.

What steps will reproduce the problem?

While running a long running command, the API call will return before
completing.   This causes race conditions and there is no option
to run synchronously or get_state.


What version of the product are you using? On what operating system?
1.7, RedHat 6

The current API does not include the GuestDiskInfo data as described in 
http://www.vmware.com/support/developer/vc-sdk/visdk41pubs/ApiReference/

The attached patch adds them. I'm happy to rename the property or a different 
approach if you prefer. Feedback welcome.

Cheers,
 Michael

No really a technical issue but setup.py and the PKG-INFO files
are pointing to the older 0.1.5 release.

Please add SDK 5.1 Support

Is there any way to get the allocated disk space for a specific host?
I need to run a comprehensive Python script against a few thousand machines and 
one of the key things i need to fetch is allocated disk space.


I can not find any documentation on it, nor can i find anything in the source 
codes, is this feature completely left out?

1. pid = self.vm.start_process(r'c:\windows\system32\calc.exe', arg)
2. vm.list_processes() >>> pid found as expected.
3. kill the process.
4. vm.list_processes() >>> pid still being listed.

That's an error i guess.

pysphere 0.1.7, Python 2.6, ESXi 5.0, Windows XP SP3 guest, Windows 7 64bits 
host.

I'm writing a function to start process in guest and wait until process be dead.

Reported by Harish


Code:
{{{
from pysphere import VIServer
server = VIServer()
server.connect("192.168.3.30", "root", "root123")
vmlist = server.get_registered_vms()
print server.get_registered_vms()
vm1 = server.get_vm_by_name("Win64-7")
vm2 = server.get_vm_by_path("[datastore2] Win64-7/Win64-7.vmx")´
print vm1.get_status()
print vm2.get_status()
server.disconnect() 
}}}

Error:

{{{
Python 2.7.1 (r271:86832, Nov 27 2010, 18:30:46) [MSC v.1500 32 bit (Intel)] on 
win32 Type "copyright", "credits" or "license()" for more information. 

'[datastore2] RHEL64-5.3 ES/RHEL64-5.3 ES.vmx', 'datastore2? 
Win64-7/Win64-7.vmx', '[datastore2] New Virtual Machine/New Virtual 
Machine.vmx'] 

Traceback (most recent call last): 

File "C:\Documents and Settings\hmunagapati\Desktop\vm_py\test.py", line 
10, in <module> 
vm1 = server.get_vm_by_name("Win64-7") 
File "C:\Python27\lib\site-packages\pysphere\vi_server.py", line 218, in 
get_vm_by_name 
return VIVirtualMachine(self, mor) 
File "C:\Python27\lib\site-packages\pysphere\vi_virtual_machine.py", line 
47, in init 
self.update_properties() 
File "C:\Python27\lib\site-packages\pysphere\vi_virtual_machine.py", line 
741, in update_properties 
self.mor, get_all=True) 
File "C:\Python27\lib\site-packages\pysphere\vi_server.py", line 396, in 
get_object_properties 
return request_call(request) 
File "C:\Python27\lib\site-packages\pysphere\vi_server.py", line 696, in 
call_retrieve_properties 
return self.proxy.RetrieveProperties?(request).returnval 
File 
"C:\Python27\lib\site-packages\pysphere\resources\VimService?_services.py", 
line 61, in RetrieveProperties? 
response = self.binding.Receive(RetrievePropertiesResponseMsg?.typecode) 
File "C:\Python27\lib\site-packages\pysphere\ZSI\client.py", line 544, in 
Receive 
return Binding.Receive(self, replytype, kw) 
File "C:\Python27\lib\site-packages\pysphere\ZSI\client.py", line 469, in 
Receive 
reply = self.ps.Parse(tc) 
File "C:\Python27\lib\site-packages\pysphere\ZSI\parse.py", line 326, in 
Parse 
return how.parse(self.body_root, self) 
File "C:\Python27\lib\site-packages\pysphere\ZSI\TCcompound.py", line 
203, in parse 
value = what.parse(c_elt, ps) 
File "C:\Python27\lib\site-packages\pysphere\ZSI\TCcompound.py", line 
203, in parse 
value = what.parse(c_elt, ps) 
File "C:\Python27\lib\site-packages\pysphere\ZSI\TCcompound.py", line 
203, in parse 
value = what.parse(c_elt, ps) 
File "C:\Python27\lib\site-packages\pysphere\ZSI\TC.py", line 1392, in 
parse 
pyobj = what.parse(elt, ps) 
File "C:\Python27\lib\site-packages\pysphere\ZSI\TCcompound.py", line 
203, in parse 
value = what.parse(c_elt, ps) 
File "C:\Python27\lib\site-packages\pysphere\ZSI\TC.py", line 446, in 
parse 
pyobj = self.text_to_data(v, elt, ps) 
File "C:\Python27\lib\site-packages\pysphere\ZSI\TCtimes.py", line 211, 
in text_to_data 
retval = fix_timezone(retval, tz_from = m.groupdict().get('tz'), tz_to 
= None) 
File "C:\Python27\lib\site-packages\pysphere\ZSI\TCtimes.py", line 98, in 
fix_timezone 
ltv:6? = datetime((ltv:6? + [0, 
tz_from])).astimezone(tz_to).timetuple():6? 
File "C:\Python27\lib\site-packages\pysphere\ZSI\TCtimes.py", line 48, in 
utcoffset 
dt.hour, dt.minute, dt.second, dt.weekday(), 0, -1))) 
OverflowError?: mktime argument out of range 
}}}

I could see an attribute check in method "get_entity_statistic" of 
vi_performance_manager.py

if composite:
            if hasattr(query, "Entity"):
                stats.extend(query.Entity.Value)
            if hasattr(query, "ChildEntity"):
                for item in query.ChildEntity:
                    stats.extend(item.Value)
        else:
            if hasattr(query[0], "Value"):
                stats = query[0].Value


When will happen this situation ???
AttributeError: PerfCompositeMetric instance has no attribute 'entity'
AttributeError: PerfCompositeMetric instance has no attribute 'ChildEntity'

Is this the only way to do it. 

Why can't we check value of attibutes (currentSupported or summarySupported) of 
PerfProviderSummary and proceed.

What steps will reproduce the problem?
1.
properties = ["summary.targetBalance"]

results = s._retrieve_properties_traversal(property_names=properties,
                                           obj_type=MORTypes.ClusterComputeResource)


for item in results:
    print "=" * 80
    print "MOR:", item.Obj
    for p in item.PropSet:
        print p.Name, "=>", p.Val


What is the expected output? 
the same as effectiveCpu

What do you see instead?
Invalid Property Type

What version of the product are you using? On what operating system?
pysphere-0.1.7

Please provide any additional information below.
They keys are there
['AdmissionControlInfo', 'CurrentBalance', 'CurrentEVCModeKey', 
'CurrentFailoverLevel', 'DasData', 'DynamicProperty', '
DynamicType', 'EffectiveCpu',....


Output from

mors = s.get_datacenters().keys() + s.get_clusters().keys() + 
s.get_hosts().keys() + s._get_managed_objects(MORTypes.VirtualMachine).keys()
# for mor in mors: print mor
props = {MORTypes.ClusterComputeResource:[]}
results = s._get_object_properties_bulk(mors, props)


for host in results:
    for prop in host.PropSet:
       print "prop Name"
       print "========="
       print dir(prop.Name)
       print "prop Val"
       print "========="
       print dir(prop.Val)

Was able to log in to ESXi and find the VM by name, then any attempts to log in 
to the VM fail. I am able to connect and login to a Windows guest just fine. I 
created additional users in my unix VM and tested, all of which failed the same 
with the following error:
Traceback (most recent call last):
  File "./test.py", line 20, in <module>
    vm.login_in_guest("user1", "password")
  File "/usr/local/lib/python2.7/dist-packages/pysphere/vi_virtual_machine.py", line 1044, in login_in_guest
    self.__validate_authentication(auth)
  File "/usr/local/lib/python2.7/dist-packages/pysphere/vi_virtual_machine.py", line 1722, in __validate_authentication
    raise VIApiException(e)
pysphere.resources.vi_exception.VIApiException: [InvalidGuestLoginFault]: 
Failed to authenticate with the guest operating system using the supplied 
credentials.

System info of the VM running the script:
root@test:~# uname -a
Linux archive 3.2.0-51-generic-pae #77-Ubuntu SMP Wed Jul 24 20:40:32 UTC 2013 
i686 athlon i386 GNU/Linux
root@test:~# cat /proc/version
Linux version 3.2.0-51-generic-pae (buildd@komainu) (gcc version 4.6.3 
(Ubuntu/Linaro 4.6.3-1ubuntu5) ) #77-Ubuntu SMP Wed Jul 24 20:40:32 UTC 2013
root@test:~# lsb_release -a
No LSB modules are available.
Distributor ID: Ubuntu
Description:    Ubuntu 12.04.2 LTS
Release:        12.04
Codename:       precise

See ubuntu.png for the version information of the VM (only able to get console 
access and no copy/paste for text mode)

The VM is running an updated version on VMware tools.
-See ubuntu2.png picture

Finally, I tested this on an Ubuntu Desktop VM and a different Ubuntu Server 
VM, all of which returned the same error as above.

Let me know if I need to provide any further information.

What steps will reproduce the problem?
1. Start python console
2. Load pysphere module
3. Create pysphere instance and connect to ESXi server (successful) 
4. create vm instance using get_vm_by_name()
5. Set guest credentials using vm.login_in_guest()
6. Attempt to push local file to guest file system using vm.send_file() 
function.  This produces the error below.  

What is the expected output? What do you see instead?
Successful push of file from local system to guest system

What version of the product are you using? On what operating system?
Current - 0.1.8

Please provide any additional information below.

From initial trouble shooting it appears the IPv6 address somewhere in the 
process to use the urllib2.py library.  Looks like the ":" colons are being 
used somewhere to split the address up expecting a port and is ignoring the 
wrapping square brackets "[]".
initial ipv6 address = "[fd30:f8fe:a3af:4466::100]"
resulting url var looking like this....
url = "[fd30:"

Currently I am seeing this by looking in some print statements in the 
vm_virtual_machine.py around line 1292. "url = url.replace("*", 
urlparse(self._server._proxy.binding.url).hostname)"


Python session below
###############################################

Python 2.6.6 (r266:84292, May 27 2013, 05:35:12) 
[GCC 4.4.7 20120313 (Red Hat 4.4.7-3)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> from pysphere import *
>>> s = VIServer()
>>> s.connect("[fd30:f8fe:a3af:4466::100]", "admin", "adminpass")
>>> vm = s.get_vm_by_name("IPV6guest")
>>> vm.login_in_guest("guestuser","guestpass")
>>> stat = 
vm.send_file("/etc/puppet/manifests/site.pp","/etc/puppet/manifests/site.pp", 
True)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/pyve/mdms/lib/python2.6/site-packages/pysphere/vi_virtual_machine.py", line 1301, in send_file
    resp = urllib2.urlopen(request)
  File "/usr/lib64/python2.6/urllib2.py", line 126, in urlopen
    return _opener.open(url, data, timeout)
  File "/usr/lib64/python2.6/urllib2.py", line 391, in open
    response = self._open(req, data)
  File "/usr/lib64/python2.6/urllib2.py", line 409, in _open
    '_open', req)
  File "/usr/lib64/python2.6/urllib2.py", line 369, in _call_chain
    result = func(*args)
  File "/usr/lib64/python2.6/urllib2.py", line 1198, in https_open
    return self.do_open(httplib.HTTPSConnection, req)
  File "/usr/lib64/python2.6/urllib2.py", line 1165, in do_open
    raise URLError(err)
urllib2.URLError: <urlopen error [Errno -2] Name or service not known>

This is two related requets. Since most other systems rely on the VM's uuid, I 
propose:

1. Have a function or flag on get_registered_vms that returns the uuid of a vm 
(..., returnUUID=False). I do see that this information is available under 
VirtualMachineConfigSummary_Def.

2. Implement a get_vm_by_uuid function.

Additionally, as other object (such as vcenter) have a uuid, start looking at 
how to expose this information and select object by uuid.

Are there any plans to implement the 'VixVM_WaitForToolsInGuest' function.  
Right now, if you revert from a snapshot and attempt to access guest functions, 
vmware tools needs time to initialize or you can crash the guest.

What steps will reproduce the problem?
   Using the code in the SVN trunk
   Doing 'python ./setup.py install'
   Then at an interactive python (v2.6) prompt, typing: 
        import pysphere

   This results in the following output:
   [ ... ]
     File "pysphere/vi_performance_manager.py", line 31, in <module>
       from pysphere import VIProperty
   ImportError: cannot import name VIProperty
   >>>>

This happens on Linux (Ubuntu 10.04.03), as well as on MacOSX (Python 2.7 in 
that environment).
Applying the attached patch (then re-installing) resolved this problem for me.

It would be great if there was a "public" get_mor() method (I'm currently using 
the _mor property).

Use case: This is an easy way to get snapshots and display them in a web page 
(a lot faster that either of the clients), see option 2 here:
http://blogs.vmware.com/vsphere/2013/01/capturing-virtual-machine-screenshots-in
-vsphere.html

ie: use the URL: https://10.24.145.65/screen?id=vm-162

What steps will reproduce the problem?
1. vm.start_process(r"c:\my.py"),  

you can read more about the problem in: 
https://groups.google.com/group/pysphere/browse_thread/thread/9b705990d68b968d

What is the expected output? What do you see instead?
running python module. or a python script easily 

What version of the product are you using? On what operating system?
latest.

Please provide any additional information below.

What steps will reproduce the problem?
Unknown.  Problem happens randomly on 10% of the VM guests.

What is the expected output? What do you see instead?


What version of the product are you using? On what operating system?
PySphere 0.1.8.  OS is Windows XP through 8.1 (not version specific).

Please provide any additional information below.

Traceback (most recent call last):
  File "C:\p4root\depot\Systems_Software\Products\Client_IDS\temporary_dev_area\tnili\cidsauto\vsphere\vsphere.py", line 46, in run
    vm1 = server.get_vm_by_path(server_path)
  File "C:\Python27\lib\site-packages\pysphere\vi_server.py", line 286, in get_vm_by_path
    return VIVirtualMachine(self, vm)
  File "C:\Python27\lib\site-packages\pysphere\vi_virtual_machine.py", line 55, in __init__
    self.__update_properties()
  File "C:\Python27\lib\site-packages\pysphere\vi_virtual_machine.py", line 1801, in __update_properties
    p['name'] = self.properties.name
  File "C:\Python27\lib\site-packages\pysphere\vi_property.py", line 71, in __getattr__
    self._get_all()
  File "C:\Python27\lib\site-packages\pysphere\vi_property.py", line 53, in _get_all
    oc = self._server._get_object_properties(self._obj, get_all=True)
  File "C:\Python27\lib\site-packages\pysphere\vi_server.py", line 484, in _get_object_properties
    ret = request_call(request)
  File "C:\Python27\lib\site-packages\pysphere\vi_server.py", line 733, in call_retrieve_properties_ex
    request)._returnval
  File "C:\Python27\lib\site-packages\pysphere\resources\VimService_services.py", line 114, in RetrievePropertiesEx
    response = self.binding.Receive(RetrievePropertiesExResponseMsg.typecode)
  File "C:\Python27\lib\site-packages\pysphere\ZSI\client.py", line 545, in Receive
    return _Binding.Receive(self, replytype, **kw)
  File "C:\Python27\lib\site-packages\pysphere\ZSI\client.py", line 470, in Receive
    reply = self.local.ps.Parse(tc)
  File "C:\Python27\lib\site-packages\pysphere\ZSI\parse.py", line 325, in Parse
    return how.parse(self.body_root, self)
  File "C:\Python27\lib\site-packages\pysphere\ZSI\TCcompound.py", line 199, in parse
    value = what.parse(c_elt, ps)
  File "C:\Python27\lib\site-packages\pysphere\ZSI\TCcompound.py", line 199, in parse
    value = what.parse(c_elt, ps)
  File "C:\Python27\lib\site-packages\pysphere\ZSI\TCcompound.py", line 199, in parse
    value = what.parse(c_elt, ps)
  File "C:\Python27\lib\site-packages\pysphere\ZSI\TCcompound.py", line 199, in parse
    value = what.parse(c_elt, ps)
  File "C:\Python27\lib\site-packages\pysphere\ZSI\TC.py", line 1392, in parse
    pyobj = what.parse(elt, ps)
  File "C:\Python27\lib\site-packages\pysphere\ZSI\TCcompound.py", line 187, in parse
    if callable(what): what = what()
  File "C:\Python27\lib\site-packages\pysphere\ZSI\schema.py", line 350, in _reveal_type
    **self.__kw)
  File "C:\Python27\lib\site-packages\pysphere\resources\VimService_services_types.py", line 43114, in __init__
    ns0.GuestScreenInfo_Def.__bases__ = tuple(bases)
TypeError: duplicate base class DynamicData_Def

We are using vCenter 5.0 and our team has created many templates from which 
VM's are deployed.  I would like to know if pysphere can be used to automate 
this process.

The docs for VIVirtualMachine.clone seem to indicate that this is possible 
although I could not get it to work.

>>> vms = server.get_registered_vms()
>>> templates = [vm for vm in vms if '[iSCSI-Templates]' in vm]
>>> templates[1]
'[iSCSI-Templates] 
WindowsVista-SP1-IE8-x86-Template/WindowsVista-SP1-IE8-x86-Template.vmtx'
>>> vm = server.get_vm_by_path(templates[1])
>>> new_vm = vm.clone("Clone")
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "C:\Python27\lib\site-packages\pysphere\vi_virtual_machine.py", line 537,
 in clone
    FaultTypes.TASK_ERROR)
pysphere.resources.vi_exception.VIException: [Task Error]: The operation is not
allowed in the current state of the host.