ZenPacks : WMI Windows Performance : Comments

RE: WMI Windows Performance

community@zenoss.org — Mon, 05 Oct 2009 20:55:44 GMT

Egor,

What controls the polling Interval for your WMI datasources?

RE: WMI Windows Performance

community@zenoss.org — Thu, 17 Dec 2009 19:39:09 GMT

SNMP Performance Cycle Interval (perfsnmpCycleInterval) - 300 sec

RE: WMI Windows Performance

community@zenoss.org — Thu, 28 Jan 2010 16:45:59 GMT

Hi, Whay you take a look in the the Disc monitor in the MS SQL Server ZenPack.

http://community.zenoss.org/docs/DOC-4641

Thank you....

RE: WMI Windows Performance

community@zenoss.org — Fri, 26 Mar 2010 19:09:32 GMT

Hi all. I am new to Zenoss (been using it since December 2009) so I am still feeling my way around the community site. I have been testing the RC 2.0 of WMIPerf and I have some issues that occur during that adding of a device. Most of these deal with WMI syntax.

Is this the appropriate place to post them?

Also Egor I have a question about the OSProcess that you are implementing. Is the goal to be able to monitor processes on windows machines (up/down status) without SNMP or do you still need SNMP to monitor status?

Thanks

RE: WMI Windows Performance

community@zenoss.org — Fri, 26 Mar 2010 19:59:56 GMT

Hi ocongil,

SNMP not necessary anymore, you can monitor your server with WMI only.

Egor

RE: WMI Windows Performance

community@zenoss.org — Fri, 26 Mar 2010 23:08:10 GMT

Thanks for the reply Egor. That will be awesome. I am testing the packs now.

I have run into a bit of a problem with the data source. Whenever I try to test an existing or add a new WMI data source for testing the result always comes back

Get WMI Instance from //10.0.14.35/root/cimv2 Traceback (most recent call last): File "/usr/local/zenoss/zenoss/ZenPacks/ZenPacks.community.WMIDataSource-2.1-py2.4.egg/ZenPacks/community/WMIDataSource/WMIClient.py", line 336, in ? results = WmiGet(url, query, properties) File "/usr/local/zenoss/zenoss/ZenPacks/ZenPacks.community.WMIDataSource-2.1-py2.4.egg/ZenPacks/community/WMIDataSource/WMIClient.py", line 278, in WmiGet url[2], device.zWmiProxy = url[2].split('@') ValueError: need more than 1 value to unpack DONE in 6 seconds

I am pretty sure my zWinUser and zPassword are correct cause I can use the older version of the WMI zenpack.

By the way, if this isn't the perfered way to discuss dev issues please let me know. I am still trying to navigate the community.

RE: WMI Windows Performance

community@zenoss.org — Sat, 27 Mar 2010 14:08:41 GMT

after install the 2.0 RC1, can't add device.

RE: WMI Windows Performance

community@zenoss.org — Sun, 28 Mar 2010 15:35:42 GMT

check if your password contains '@:. ' symbols.

RE: WMI Windows Performance

community@zenoss.org — Mon, 29 Mar 2010 00:46:47 GMT

excuse me, do you means the linux root password of zenoss? or Zenoss administrator password? or other?

Please let me knows. Thanks,

RE: WMI Windows Performance

community@zenoss.org — Mon, 29 Mar 2010 00:59:23 GMT

He's referring to the username and password for the windows account which you are using to perform your WMI monitoring. These would be specified in zWinUser and zWinPassword zProperties.

RE: WMI Windows Performance

community@zenoss.org — Mon, 29 Mar 2010 01:50:15 GMT

Thanks Ryan.

My case is after installed the RC version, the Zenoss can't add device and show error, and Egor said check the password for special char. such as "@:.", but I doesn't know which password that Egor means.

I've changed the Zenoss password and trying to install this RC zenpack again, if the error exist, I'll copy to here.

Thanks~

RE: WMI Windows Performance

community@zenoss.org — Mon, 29 Mar 2010 02:03:12 GMT

Egor,

Atatched the error for your reference.

Thanks.

Using Zenoss 2.5.2 on VM appliance. Installed WMI data source 2.1 and WMI perf 1.97

Type: ValueError

Value: too many values to unpack

Traceback (innermost last):

    * Module ZPublisher.Publish, line 119, in publish
    * Module ZPublisher.mapply, line 88, in mapply
    * Module ZPublisher.Publish, line 42, in call_object
    * Module Products.Five.browser.metaconfigure, line 417, in __call__
    * Module Shared.DC.Scripts.Bindings, line 313, in __call__
    * Module Shared.DC.Scripts.Bindings, line 350, in _bindAndExec
    * Module Products.PageTemplates.PageTemplateFile, line 126, in _exec
    * Module Products.PageTemplates.PageTemplate, line 98, in pt_render
    * Module zope.pagetemplate.pagetemplate, line 115, in pt_render
    * Module zope.tal.talinterpreter, line 271, in __call__
    * Module zope.tal.talinterpreter, line 346, in interpret
    * Module zope.tal.talinterpreter, line 891, in do_useMacro
    * Module zope.tal.talinterpreter, line 346, in interpret
    * Module zope.tal.talinterpreter, line 536, in do_optTag_tal
    * Module zope.tal.talinterpreter, line 521, in do_optTag
    * Module zope.tal.talinterpreter, line 516, in no_tag
    * Module zope.tal.talinterpreter, line 346, in interpret
    * Module zope.tal.talinterpreter, line 957, in do_defineSlot
    * Module zope.tal.talinterpreter, line 346, in interpret
    * Module zope.tal.talinterpreter, line 536, in do_optTag_tal
    * Module zope.tal.talinterpreter, line 521, in do_optTag
    * Module zope.tal.talinterpreter, line 516, in no_tag
    * Module zope.tal.talinterpreter, line 346, in interpret
    * Module zope.tal.talinterpreter, line 949, in do_defineSlot
    * Module zope.tal.talinterpreter, line 346, in interpret
    * Module zope.tal.talinterpreter, line 536, in do_optTag_tal
    * Module zope.tal.talinterpreter, line 521, in do_optTag
    * Module zope.tal.talinterpreter, line 516, in no_tag
    * Module zope.tal.talinterpreter, line 346, in interpret
    * Module zope.tal.talinterpreter, line 536, in do_optTag_tal
    * Module zope.tal.talinterpreter, line 521, in do_optTag
    * Module zope.tal.talinterpreter, line 516, in no_tag
    * Module zope.tal.talinterpreter, line 346, in interpret
    * Module zope.tal.talinterpreter, line 891, in do_useMacro
    * Module zope.tal.talinterpreter, line 346, in interpret
    * Module zope.tal.talinterpreter, line 536, in do_optTag_tal
    * Module zope.tal.talinterpreter, line 521, in do_optTag
    * Module zope.tal.talinterpreter, line 516, in no_tag
    * Module zope.tal.talinterpreter, line 346, in interpret
    * Module zope.tal.talinterpreter, line 949, in do_defineSlot
    * Module zope.tal.talinterpreter, line 346, in interpret
    * Module zope.tal.talinterpreter, line 536, in do_optTag_tal
    * Module zope.tal.talinterpreter, line 521, in do_optTag
    * Module zope.tal.talinterpreter, line 516, in no_tag
    * Module zope.tal.talinterpreter, line 346, in interpret
    * Module zope.tal.talinterpreter, line 891, in do_useMacro
    * Module zope.tal.talinterpreter, line 346, in interpret
    * Module zope.tal.talinterpreter, line 536, in do_optTag_tal
    * Module zope.tal.talinterpreter, line 521, in do_optTag
    * Module zope.tal.talinterpreter, line 516, in no_tag
    * Module zope.tal.talinterpreter, line 346, in interpret
    * Module zope.tal.talinterpreter, line 623, in do_insertText_tal
    * Module Products.PageTemplates.Expressions, line 206, in evaluateText
    * Module zope.tales.tales, line 696, in evaluate
      URL: quickstart_macros
      Line 80, Column 0
      Expression: <StringExpr '\n var DEFAULT_SNMPCOMMUNITIES = ${view/default_communities};\n var DEVICE_TYPES = ${view/device_types};\n'>
      Names:

      {'container': <DataRoot at /zport/dmd>,
       'context': <DataRoot at /zport/dmd>,
       'default': <object object at 0xb7c9d528>,
       'here': <DataRoot at /zport/dmd>,
       'loop': {},
       'nothing': None,
       'options': {'args': (<Products.Five.metaclass.SimpleViewClass from /home/zenoss/Products/ZenModel/easyAddDevice.pt object at 0xb32ba8ac>,)},
       'repeat': <Products.PageTemplates.Expressions.SafeMapping object at 0xb338458c>,
       'request': <HTTPRequest, URL=http://zenoss:8080/zport/dmd/easyAddDevice>,
       'root': <Application at >,
       'template': <ImplicitAcquirerWrapper object at 0xb3362e4c>,
       'traverse_subpath': [],
       'user': <PropertiedUser 'srvadmin01'>,
       'view': <Products.Five.metaclass.SimpleViewClass from /home/zenoss/Products/ZenModel/easyAddDevice.pt object at 0xb32ba8ac>,
       'views': <zope.app.pagetemplate.viewpagetemplatefile.ViewMapper object at 0xb32b250c>}

    * Module zope.tales.expressions, line 261, in __call__
    * Module zope.tales.expressions, line 217, in __call__
    * Module Products.PageTemplates.Expressions, line 139, in _eval
    * Module Products.PageTemplates.Expressions, line 103, in render
    * Module Products.ZenUtils.json, line 41, in inner
    * Module Products.ZenWidgets.browser.quickstart.views, line 96, in device_types
    * Module Products.ZenWidgets.browser.quickstart.views, line 76, in _assemble_types_list

ValueError: too many values to unpack

RE: WMI Windows Performance

community@zenoss.org — Tue, 30 Mar 2010 18:15:12 GMT

Hi Egor

I will try it with a new password as it does have one of those symbols. Thanks

RE: WMI Windows Performance

community@zenoss.org — Tue, 30 Mar 2010 18:48:06 GMT

I explained to you which password he meant. zWinPassword in zProperties.

RE: WMI Windows Performance

community@zenoss.org — Wed, 31 Mar 2010 01:39:50 GMT

Thanks, Ryan.

I've checked the zWinpassword with no include the symbols "@.:"

I"ve tried to deploy the 2.5.2 VM appliance and install this zenpack with no machine, has not configure any zProperties, the problem exists.

RE: WMI Windows Performance

community@zenoss.org — Mon, 12 Apr 2010 23:20:59 GMT

Great plugins Egor. I've been looking for a simpler way to get WMI information. I've run into a situation where a host that has both IPv4 and an IPv6 address is tracebacking during modeling. I'm running Zenoss 2.5.2 and the following packs:

ZenPacks.community.WMIDataSource	community	Egor Puzanov	2.2	Yes
ZenPacks.community.WMIPerf_Windows	community	Egor Puzanov	1.97	Yes

During modeling, the following happens after wmi.ProductMap:

Remote exception: Products.ZenUtils.IpUtil.IpAddressError: fe80::994:xxxx:70f6:3984 is an invalid addressTraceback (most recent call last): File "/usr/local/zenoss/zenoss/Products/ZenHub/PBDaemon.py", line 75, in inner return callable(*args, **kw) File "/usr/local/zenoss/zenoss/Products/ZenHub/services/ModelerService.py", line 108, in remote_applyDataMaps if adm._applyDataMap(device, map): File "/usr/local/zenoss/zenoss/Products/DataCollector/ApplyDataMap.py", line 167, in _applyDataMap changed = self._updateRelationship(tobj, datamap) File "/usr/local/zenoss/zenoss/Products/DataCollector/ApplyDataMap.py", line 230, in _updateRelationship objchange, obj = self._createRelObject(device, objmap, rname) File "/usr/local/zenoss/zenoss/Products/DataCollector/ApplyDataMap.py", line 409, in _createRelObject return self._updateObject(remoteObj, objmap) or changed, remoteObj File "/usr/local/zenoss/zenoss/Products/DataCollector/ApplyDataMap.py", line 339, in _updateObject setter(*args) File "/usr/local/zenoss/zenoss/Products/ZenModel/IpInterface.py", line 295, in setIpAddresses self.addIpAddress(ip) File "/usr/local/zenoss/zenoss/Products/ZenModel/IpInterface.py", line 246, in addIpAddress ipobj = networks.createIp(ip, netmask) File "/usr/local/zenoss/zenoss/Products/ZenModel/IpNetwork.py", line 273, in createIp netobj = self.createNet(ip, netmask) File "/usr/local/zenoss/zenoss/Products/ZenModel/IpNetwork.py", line 184, in createNet netobj = netroot.getNet(netip) File "/usr/local/zenoss/zenoss/Products/ZenModel/IpNetwork.py", line 249, in getNet return self._getNet(ip) File "/usr/local/zenoss/zenoss/Products/ZenModel/IpNetwork.py", line 256, in _getNet if net.hasIp(ip): File "/usr/local/zenoss/zenoss/Products/ZenModel/IpNetwork.py", line 292, in hasIp return start <= numbip(ip) < end File "/usr/local/zenoss/zenoss/Products/ZenUtils/IpUtil.py", line 89, in numbip return ipToDecimal(ip) File "/usr/local/zenoss/zenoss/Products/ZenUtils/IpUtil.py", line 102, in ipToDecimal checkip(ip) File "/usr/local/zenoss/zenoss/Products/ZenUtils/IpUtil.py", line 78, in checkip raise IpAddressError( "%s is an invalid address" % ip ) IpAddressError: fe80::994:xxxx:70f6:3984 is an invalid address

As I understand it, Zenoss currently doesn't support IPv6. Is there a way to enumerate the network connections and just pull IPv4 addresses when creating Interfaces?

RE: WMI Windows Performance

community@zenoss.org — Tue, 13 Apr 2010 04:21:03 GMT

Egor thanks for the great set of ZenPacks

I am still having an issue with the Process monitoring. No matter what process I try and monitor it is always green both in /Process/<processname> and on the individual machines OS tab. I setup the monitoring the same way as you would the standard SNMP monitoring (meaning add the process to /Processes and then add it the machines instance to monitor it).

I decided to poke around a little in the OSProcesses template. I noticed that when I ran the test of the datasource (after hardcoding a process name since the tales expression for the process name was not available) test run when the process was running returned valid data. If I ran it when the process wasn’t running I got this error:

Get WMI Instance Win32_Process.Name="notepad.exe" from //<device>/root/cimv2 Traceback (most recent call last): File "/usr/local/zenoss/zenoss/ZenPacks/ZenPacks.community.WMIDataSource-2.2-py2.4.egg/ZenPacks/community/WMIDataSource/WMIClient.py", line 337, in ? results = WmiGet(url, query, properties) File "/usr/local/zenoss/zenoss/ZenPacks/ZenPacks.community.WMIDataSource-2.2-py2.4.egg/ZenPacks/community/WMIDataSource/WMIClient.py", line 306, in WmiGet return result['t'] KeyError: 't' DONE in 1 seconds

I know that when you put a where clause on a WQL query and the where clause causes a null result the WQL query comes back as a null set. Could this be causing this issue?

Everything else except the Software tab of the devices looks awesome but that could be a WMI Installer Provider issue so I assume the devices are setup okay. I haven’t had time to look at the Software list issue yet but I did notice the following error on during the device discovery:

2010-04-12 21:56:43ERROR zen.ZenDiscProblem while executing plugin community.wmi.ProductMap 2010-04-12 21:56:43ERROR zen.ZenDiscTraceback (most recent call last): File "/usr/local/zenoss/zenoss/Products/DataCollector/zenmodeler.py", line 612, in processClient datamaps = plugin.process(device, results, self.log) File "/usr/local/zenoss/zenoss/ZenPacks/ZenPacks.community.WMIPerf_Windows-1.97-py2.4.egg/ZenPacks/community/WMIPerf_Windows/modeler/plugins/community/wmi/ProductMap.py", line 62, in process om.setInstallDate = "%d/%02d/%02d %02d:%02d:%02d" % ( TypeError: unsubscriptable object

RE: WMI Windows Performance

community@zenoss.org — Tue, 13 Apr 2010 17:16:15 GMT

I just posted WMIPerf_Windows RC2. Thanks to gadams999, Jim Gilbert and ZenKeck for testing and feedbacks.

RE: WMI Windows Performance

community@zenoss.org — Tue, 13 Apr 2010 17:51:13 GMT

Confirming that an upgrade from RC1 (1.97) to RC2 (1.98), a restart of zenstack, and remodel of the devices with both IPv4 and IPv6 interfaces are working as expected. Software tab is now populating too, and I think I'll play round with the OS Processes bits.

RE: WMI Windows Performance

community@zenoss.org — Wed, 14 Apr 2010 19:19:40 GMT

The Software tab is now working great for me too. Thanks

OSProcess still appear to be in the same state. If the process is running (or multiple instances are running) everything works fine but the Status does not change if the process(es) are shut down and there is no alert sent.

I should be able to do a little more testing this afternoon.

RE: WMI Windows Performance

community@zenoss.org — Tue, 20 Apr 2010 20:44:59 GMT

I have 5 system that are having problem with the ProductMap. They are all getting this error, shown below. I verified that WMI Installer is installed and verified through wbemtest that Win32_Product is there also.

2010-04-20 10:46:17,239 ERROR zen.ZenModeler: Problem while executing plugin community.wmi.ProductMap
2010-04-20 10:46:17,239 ERROR zen.ZenModeler: Traceback (most recent call last):
File "/usr/local/zenoss/zenoss/Products/DataCollector/zenmodeler.py", line 612, in processClient
    datamaps = plugin.process(device, results, self.log)
File "/usr/local/zenoss/zenoss/ZenPacks/ZenPacks.community.WMIPerf_Windows-1.98-py2.4.egg/ZenPacks/community/WMIPerf_Windows/modeler/plugins/community/wmi/ProductMap.py", line 65, in process
    om.id = self.prepId(om.setProductKey)
File "/usr/local/zenoss/zenoss/Products/DataCollector/plugins/CollectorPlugin.py", line 53, in prepId
    return globalPrepId(id, subchar)
File "/usr/local/zenoss/zenoss/Products/ZenUtils/Utils.py", line 615, in prepId
    raise ValueError('Ids can not be None')
ValueError: Ids can not be None

RE: WMI Windows Performance

community@zenoss.org — Tue, 20 Apr 2010 20:54:49 GMT

Did you by any chance install the ZenPack but not restart Zenoss afterwards?

RE: WMI Windows Performance

community@zenoss.org — Tue, 20 Apr 2010 20:58:23 GMT

Zenoss was restarted. The 5 that I am having issues with our new systems. I have added about 30 new system and these 5 are having this issue. I have not been able to determine any relation to them because they very in OS.

RE: WMI Windows Performance

community@zenoss.org — Thu, 22 Apr 2010 19:41:19 GMT

I've update WMIDataSource (2.3) and WMIPerf_Windows (1.99) ZenPacks on SVN. Testers are welcomed!

Egor

RE: WMI Windows Performance

community@zenoss.org — Tue, 27 Apr 2010 06:50:52 GMT

Woops !

Double post of Allan M.

Best regards,

Tommy

~~I've been using this ZenPack for a week and it is working great for most of my servers.~~

~~I seems like I have a strange error at a couple of the servers when modeling them.~~

~~The ProductMap returns the following error, and no software shows up in the software tab.~~

~~----------------------~~

Traceback (most recent call last): File "/usr/local/zenoss/zenoss/Products/DataCollector/zenmodeler.py", line 612, in processClient datamaps = plugin.process(device, results, self.log) File "/usr/local/zenoss/zenoss/ZenPacks/ZenPacks.community.WMIPerf_Windows-1.98-py2.4.egg/ZenPacks/community/WMIPerf_Windows/modeler/plugins/community/wmi/ProductMap.py", line 65, in process om.id = self.prepId(om.setProductKey) File "/usr/local/zenoss/zenoss/Products/DataCollector/plugins/CollectorPlugin.py", line 53, in prepId return globalPrepId(id, subchar) File "/usr/local/zenoss/zenoss/Products/ZenUtils/Utils.py", line 615, in prepId raise ValueError('Ids can not be None') ValueError: Ids can not be None

~~----------------------~~

~~It seems that some of the software entrys doesn't return a value, I can't see which one it could be.~~

~~Can anybode give me hint on how to solve this ?~~

~~The servers that fails is running Windows server 2003 standard edition.~~

~~The Zenoss server is 2.5.1 and the ZenPacks is WMIDatasources 2.2 and WMIPerf_Windows 1.98.~~

~~Best regards,~~

~~Tommy~~

RE: WMI Windows Performance

community@zenoss.org — Tue, 27 Apr 2010 12:31:23 GMT

I've finally solved this (in my case at least) you could try the same.

In my case it was caused by a installer package that was "hidden", the packaged was "Veritas NetBackup 6.5.2A (6.5 Release Update 2A)" which instead of "ProductName" was listed as "HiddenProductName", which doesn't seems to be exposed to WMI, and with none of the other properties used by the ProductMap script.

I've created a workarround by simply browsing to the "HKEY_LOCAL_MACHINE\SOFTWARE\Classes\Installer\Products" in regedit, found the package and inserted a new string value named "ProdcutName".

To identify the package I first used wbemtest to query the "Win32_Product" this query gives a list where one of the packages don't have a name. You can use the ID from this to search the registry.

Maybe the ProductMap script should ignore values that returns <null>.

Best regards,

Tommy

RE: WMI Windows Performance

community@zenoss.org — Tue, 27 Apr 2010 15:59:19 GMT

Thanks for the information. That did correct all of my ProductName errors that I was receiving. It was either missing the ProductName key or its value was blank.

I only have one system that I am still getting an error and it happens to be with the InterfaceMap on a Windows 2000 Server. Anyone have an idea. I am thinking there is something wrong with WMI itself. When I used wbemtest and looked at the Win32_PerfRawData_Tcpip_NetworkInterface, it was blank as in no entries. Is there a repair that can be run or is there something I may have not installed with WMI on Windows 2000 Server?

2010-04-27 04:13:53,043 ERROR zen.ZenModeler: Received NT code 0x80041010 from query: SELECT * FROM Win32_PerfRawData_Tcpip_NetworkInterface
2010-04-27 04:13:53,043 INFO zen.ZenModeler: processing community.wmi.InterfaceMap for device lon-tstcal.global.sunrise.medical
2010-04-27 04:13:53,044 ERROR zen.ZenModeler: Problem while executing plugin community.wmi.InterfaceMap

2010-04-27 04:13:53,044 ERROR zen.ZenModeler: Traceback (most recent call last):
File "/usr/local/zenoss/zenoss/Products/DataCollector/zenmodeler.py", line 612, in processClient
datamaps = plugin.process(device, results, self.log)
File "/usr/local/zenoss/zenoss/ZenPacks/ZenPacks.community.WMIPerf_Windows-1.98-py2.4.egg/ZenPacks/community/WMIPerf_Windows/modeler/plugins/community/wmi/InterfaceMap.py", line 131, in process
instances = results["Win32_PerfRawData_Tcpip_NetworkInterface"]
KeyError: 'Win32_PerfRawData_Tcpip_NetworkInterface'

RE: WMI Windows Performance

community@zenoss.org — Wed, 28 Apr 2010 09:24:40 GMT

I think I've also resolved this, at least at one of my servers.

The server is running Windows server 2008 R2 on top of VMware ESX. It was originally installed with a virtual Intel netword adapter which was later switched to a VMXnet adapter. Windows keeps the cards in registry/device manager as a nonpresent adapter. After I've removed the nonpresent netadapter I can model the server.

Try take a look at this link which is for Server 2008, but might be some of the same for Server 2000.

http://forums.techarena.in/windows-server-help/1141789.htm

Best regards,

Tommy

RE: WMI Windows Performance

community@zenoss.org — Fri, 30 Apr 2010 12:46:30 GMT

After deploying 1.99, I've noticed that sysUpTime appears associated with type SNMP rather than WMI. Is it a isolated case?

RE: WMI Windows Performance

community@zenoss.org — Wed, 26 May 2010 12:05:22 GMT

After deploy v. 2.0 the CPU graps are getting blank with NAN values. I've tried a rrdtool fetch in the Processor_LoadPercentage_count.rrd file and all I get are nan values.

Any ideas?

RE: WMI Windows Performance

community@zenoss.org — Wed, 26 May 2010 13:02:17 GMT

same here..memory and paging are shown normally.

RE: WMI Windows Performance

community@zenoss.org — Wed, 26 May 2010 16:52:11 GMT

This problem caused by error in WMIDataSource ZenPack, this error is solved in SVN. Wait for next WMIDataSource release.

RE: WMI Windows Performance

community@zenoss.org — Wed, 26 May 2010 17:01:01 GMT

OK. Thanks Egor. I'll be waiting anxiously

RE: WMI Windows Performance

community@zenoss.org — Wed, 26 May 2010 20:34:47 GMT

I bet this is probably a Windows issue - when I remodel one of my servers it comes up with:

Error in community.wmi.FileSystemMap: NT_STATUS_CONNECTION_REFUSED

Server is set up exactly the same as another that works in the zProperties, etc. Any suggestions on what to look at? I've checked DCOM and WMI on the Windows side and both servers match.

RE: WMI Windows Performance

community@zenoss.org — Wed, 26 May 2010 20:49:18 GMT

Now this is odd - the zenosss.wmi.servicemap plugin reports no errors but the community.wmi.servicemap does (NT_STATUS_CONNECTION_REFUSED)

RE: WMI Windows Performance

community@zenoss.org — Thu, 27 May 2010 00:35:59 GMT

thanks Egor

RE: WMI Windows Performance

community@zenoss.org — Tue, 01 Jun 2010 12:38:35 GMT

For starters I'm a completely new to Zenoss. I just installed it last week and have been having issues installing this ZenPack which I need. It keeps giving me errors on install and i cant figure out why. Here is the information that it is giving me. I checked the links that it says are "blocked" and the first one is just gone it seems the page has been removed and the second link is just a page of links. Any help would be appreciated. Thanks in advance,

Output:

zenpack --install /tmp/ZenPacks.community.WMIPerf_Windows-2.0-py2.4.egg

Link to http://pypi.python.org/simple/ZenPacks.community.WMIDataSource/ ***BLOCKED*** by --allow-hosts

Couldn't find index page for 'ZenPacks.community.WMIDataSource' (maybe misspelled?)

Link to http://pypi.python.org/simple/ ***BLOCKED*** by --allow-hosts

No local packages or download links found for ZenPacks.community.WMIDataSource>=2.3

Couldn't find index page for 'ZenPacks.community.WMIDataSource' (maybe misspelled?)

No local packages or download links found for ZenPacks.community.WMIDataSource>=2.3

ERROR: zenpack command failed. Reason: SystemExit: error: Could not find suitable distribution for Requirement.parse('ZenPacks.community.WMIDataSource>=2.3')

Done installing ZenPack.

RE: WMI Windows Performance

community@zenoss.org — Tue, 01 Jun 2010 13:14:38 GMT

Hi mattw,

Some ZenPacks have dependencies for other ZenPacks, this one require you to first install the WMI data source ZenPack. If you check the requirements for the ZenPack you get the link.

Best regards,

Tommy

RE: WMI Windows Performance

community@zenoss.org — Tue, 01 Jun 2010 13:23:37 GMT

Thanks a ton that worked.

RE: WMI Windows Performance

community@zenoss.org — Fri, 04 Jun 2010 15:49:35 GMT

Thanks, we really appreciate all your work.

I'm having a similar issue with the CPU graphs, but instead of NAN values, I'm getting the following error:

"Missing RRD File: <server> Processor_LoadPercentage_avg" (Have been 48hrs with same result)

The other memory, paging, interfaces, disk graphs works perfect.

Any advice? I'll be glad to post the output of zenperfwmi

BTW: I was able also to fix the issue with the Win32_Products as explained some post above (Thanks!!)

RE: WMI Windows Performance

community@zenoss.org — Tue, 08 Jun 2010 08:26:34 GMT

Hi,

I'm having an issue with Process Monitoring.

I keep getting a "Could not get WMI Instance (Received NT code 0x80041017)" on a specific process.

Trying to run from command line with wmic produces the same error.

Full Error:

Could not get WMI Instance (Received NT code 0x80041017 from query: SELECT UserModeTime,PageFaults,PageFileUsage,Handle,CommandLine,VirtualSize,ThreadCount,KernelModeTime FROM Win32_Process WHERE CommandLine=""C:\\Program Files\\Microsoft SQL Server\\MSSQL10.MSSQLSERVER\\MSSQL\\Binn\\sqlservr.exe" -sMSSQLSERVER").

But, trying to run with "FROM Win32_Process WHERE ProcessId=1236" (the PID of the sqlservr in question) works fine.

any ideas on how i can fix\work around this?

Thanks,

Benji

(WMIDataSource 2.5, WMIPerf_Windows 2.1)

RE: WMI Windows Performance

community@zenoss.org — Tue, 08 Jun 2010 08:27:56 GMT

Was having the same issue, upgraded to the latest WMIDataSource (2.5) - all fixed.

RE: WMI Windows Performance

community@zenoss.org — Tue, 08 Jun 2010 13:11:27 GMT

Benji,

Having the same issue on a few processes and the same error code (0x80041017). The main process it's alerting on for me is VMwareTools.exe. Would there be any concerns in chaging the WHERE clause to PID vs. the CommandLine?

RE: WMI Windows Performance

community@zenoss.org — Tue, 08 Jun 2010 13:16:15 GMT

Only reason i would think that tracking the PID could be an issue would be that you would have to add some process logging logic behind it for when the process restarts, which might already be there but i'm too new to zenoss to know for sure.

at the moment, the query stays the same always.

RE: WMI Windows Performance

community@zenoss.org — Tue, 20 Jul 2010 17:16:30 GMT

Does this ZenPack update the standard Zenoss Performance reports (i.e. CPU, Memory, Availability, etc.)?

RE: WMI Windows Performance

community@zenoss.org — Tue, 20 Jul 2010 18:12:42 GMT

No, this Zenpack doesn't update any standard Zenoss performance reports, but it creates data points aliases which used by standard Zenoss reports.

Egor

RE: WMI Windows Performance

community@zenoss.org — Tue, 17 Aug 2010 16:06:56 GMT

I am new to Zenoss, but have read the documentation. I setup a test core linux box and have been playing around with it. I installed this zenpack along with the requirement pack. I have been messing around with it and see that I need to add the windows server into the /CIM/WMI/Win2000 folder, or I could create a folder under /CIM/WMI/ and just replicate the models and configuration for it to work.

The question I have is, how can I get this to work with other monitoring templates? For instance, I installed the WMI MSSQL Zenpack, but it will only allow you to add the templates to the /Server/Windows path. Is this typical to how custom monitoring Zenpacks work? Is there a way to apply the same configuration, models and templates from the /CIM/WMI location to say /Server/WMI/Windows? It just seems that having the configuration be unique in different portions of the tree make it difficult to apply unique monitoring combinations to a device without adding that device more than once to the tree.

Am I missing something? Please tell me I am missing something. If you need more information from me I'll be glad to provide. Using Zenoss 3.0.1 installed on Debian "Lenny" using the deb repository method.

RE: WMI Windows Performance

community@zenoss.org — Tue, 17 Aug 2010 20:20:59 GMT

Hi gcranshaw,

1. You can put Windows2003 servers in /CIM/WMI device class.

2. you can copy performance template from one device class to another (from /Server/Windows to /CIM/WMI)

Egor

RE: WMI Windows Performance

community@zenoss.org — Wed, 18 Aug 2010 15:49:53 GMT

1. Does it make a difference if I create a new device class under /CIM/WMI/ called say Windows2003?

2. Where can I copy the performance template from once device to another? I realize you can only add templates according to the default template bindings in the details view of a specific device class. I also tried manually adding a template to a specific device class under Advanced - Monitoring Templates, but that seemed to mess up some things and broke some graphs. The admin guide does not seem to cover this and I have not been able to find any documentation on it.

Thanks for your help.

RE: WMI Windows Performance

community@zenoss.org — Wed, 18 Aug 2010 19:35:46 GMT

To copy the template, go the Advanced->Monitoring Templates and select the one you want to copy (e.g. '/Server/Linux' for 'Device'). Then from the the Gear in the bottom left, select 'Override Template' and select where you want to copy it to (e.g. 'UPS in /Power'). Sorry for the confusion.

RE: WMI Windows Performance

community@zenoss.org — Wed, 18 Aug 2010 19:42:43 GMT

Hi gcranshaw,

you can create new device class Windows2003 under /CIM/WMI/, but it make no difference, because /CIM/WMI device class was created for Windows2003 servers.

Egor

RE: WMI Windows Performance

community@zenoss.org — Wed, 18 Aug 2010 19:43:49 GMT

OK, I did as you instructed. In my case, I copied the MSSQL2000/2005/2008 templates to the /CIM/WMI/ device class. Under that I have Win2000 and Windows. I have two servers in the Windows device calls. As soon as I copied those templates, I immediatly get the following event error on one of them;

Could not get WMI Instance (Received NT code 0x80041010 from query: SELECT FullScansPersec FROM Win32_PerfRawData_MSSQLSQLSERVER_MSSQLSQLSERVERAccessMethods)

WMI then seems to stop working and all the WMI graphs no longer update. What did do wrong, or, what did I not do?

RE: WMI Windows Performance

community@zenoss.org — Wed, 18 Aug 2010 19:49:01 GMT

OK, makes sense. Thanks for comfirming this for me.

RE: WMI Windows Performance

community@zenoss.org — Wed, 18 Aug 2010 19:56:16 GMT

OK, I moved the three MSSQL templates from selected to available for the /CIM/WMI/ device class, since that is where I copied the templates to. I'm guessing I would just add the template directly to the SQL server itself. Not sure what it would break the WMI data stream for all the effected servers though. Does that make sense? If one piece of WMI data is not working do they all fail?

RE: WMI Windows Performance

community@zenoss.org — Wed, 18 Aug 2010 20:43:30 GMT

1. you can bind or unbind performance templates per device basis.

2. normally, If one piece of WMI data is not working others doesn't fails?

RE: WMI Windows Performance

community@zenoss.org — Thu, 19 Aug 2010 15:12:40 GMT

Thanks for your help. Some of this I have already tried but I removed as I did not know if that was the proper way to do it. At least now I know more or less how to manage the templates across different device classes.

RE: WMI Windows Performance

community@zenoss.org — Fri, 20 Aug 2010 14:43:47 GMT

Hi folks!

The OSProcess template does monitor Virtual size and Pagefile usage in sum to estimate current process' memory consumption. Is there any possibility to add the Working Set Size without the need for changes to the ZenPack's python scripts itself?

RE: WMI Windows Performance

community@zenoss.org — Tue, 24 Aug 2010 10:36:32 GMT

Hi,

I'm rather new with zenoss. I've managed to build and setup zenoss on a Gentoo 64bit os. SNMP is working ok but I have a few window hosts that I want to monitor and I rather use WMI. The windows hosts allow access for wmi queries. I've tested it using wmic from the gentoo host.

I've installed wmidatasource and wmiperf_windows using zenpack and bound the WMIDevice template to /Device/Servers/Windows.

The performance graphs don't show anything so I ran a query from datasources. This is the error I'm getting:

Preparing Command...

Get WMI Instance SELECT AvailableKBytes, CacheBytes, PagesPerSec FROM Win32_PerfRawData_PerfOS_Memory from //192.168.78.2/root/cimv2

Traceback (most recent call last): File "/usr/local/zenoss/ZenPacks/ZenPacks.community.WMIDataSource-2.9.egg/ZenPacks/community/WMIDataSource/WMIClient.py", line 20, in import pysamba.twisted.reactor File "/usr/local/zenoss/lib/python/pysamba/twisted/reactor.py", line 15, in from pysamba.library import *

File "/usr/local/zenoss/lib/python/pysamba/library.py", line 18, in from ctypes import * File "/usr/local/zenoss/lib/python/ctypes/__init__.py", line 10, in from _ctypes import Union, Structure, Array ImportError: /usr/local/zenoss/lib/python/_ctypes.so: undefined symbol: PyUnicodeUCS2_FromEncodedObject

DONE in 0 seconds

Python version installed on system 2.6.

Has anyone encountered this error before?

RE: WMI Windows Performance

community@zenoss.org — Tue, 24 Aug 2010 12:26:03 GMT

Hi,

We are having problems with the netmask of devices not being correctly collected. We have several networks with /22 masks and when we model devices in these networks the interface appears with a /24 mask and a subnetwork with this mask is generated under the /22 network.

We have tried to change class of device to /Server/Windows, deleted the interface and the wrong network and remodel device. This time, modelling happens via SNMP and the interface appears with the correct /22 mask without generating the wrong /24 network.

I've been looking in source code in Zenoss original InterfaceMap.py as well in InterfaceMap.py of this Zenpack but I've been unable to find where the wrong behaviour is happening.

Some one else is having this issue?

Regards

RE: WMI Windows Performance

community@zenoss.org — Tue, 24 Aug 2010 18:40:56 GMT

Hi anc,

simply add new data point 'WorkingSetSize_sum' with alias ' WorkingSetSize' to OSProcess template and add this data point to graph.

RE: WMI Windows Performance

community@zenoss.org — Wed, 25 Aug 2010 10:10:55 GMT

Hi bigegor,

thanks a lot for your help and your great ZenPacks! WorkingSetSize works like a charm now.

RE: WMI Windows Performance

community@zenoss.org — Tue, 14 Sep 2010 07:05:34 GMT

Hi bigegor,

After we use the wmi windows performacne zenpack for our windows monitoring, we got some alerts like following:

Could not get WMI Instance (Received NT code 0x800700a4 from query: SELECT FreeSpace,DeviceID FROM Win32_LogicalDisk)

How can we resolve this kind of problems?

Regards

RE: WMI Windows Performance

community@zenoss.org — Tue, 21 Sep 2010 13:48:23 GMT

Can anyone confirm whether this ZenPack supports big filesystem sizes (ie. full 64-bit support) for many-terabyte systems?

Cheers,

Jane

RE: WMI Windows Performance

community@zenoss.org — Tue, 21 Sep 2010 13:58:16 GMT

Well, the real question is, does WMI support that? If so, perhaps it's in a different table than the one being queried in this pack. I'm going to be sending Egor a list of new select statements since there's a different table which provides the standard drive info, but also provides info for drives that are mounted in windows (which I've seen quite a few times).

RE: WMI Windows Performance

community@zenoss.org — Tue, 21 Sep 2010 14:38:05 GMT

We have several machines with volumes larger than 4TB and I can monitor these filesystems without problem.

RE: WMI Windows Performance

community@zenoss.org — Thu, 07 Oct 2010 11:55:24 GMT

MUST we put our server in the CIM/WMI class or can we use the community.wmi component elsewhere ? just asking, i'm all new to zenoss.

Thanks

RE: WMI Windows Performance

community@zenoss.org — Thu, 07 Oct 2010 13:56:50 GMT

Everyone I've seen try to move the plugins over to a different class (including myself) has failed. There appears to be something about the /CIM/WMI class which allows it to work. I'm sure this is by design.

RE: WMI Windows Performance

community@zenoss.org — Thu, 07 Oct 2010 14:04:29 GMT

I tried myself. I modified Device/Servers/Windows and removed all zenoss.snmp, replacing with community.wmi . And it works (remodel your device after creating it).

Then I re-enable snmp per device when I have a snmp service that allows me to monitor low level hardware (hp snmp agent, using corresponding modelers from zenpack).

RE: WMI Windows Performance

community@zenoss.org — Thu, 07 Oct 2010 14:07:48 GMT

That's all good and dandy until you run across that one server that you want to monitor with SNMP. You've just trashed your SNMP class so now you're stuck making a new class. I'll just leave it as /CIM/WMI. It's much more convenient.

RE: WMI Windows Performance

community@zenoss.org — Thu, 07 Oct 2010 14:10:46 GMT

no worries, that's just test purpose, I know I will be doing things wrong. Thanks for the practice, I think you are right.

edit:can someone PM me the default modelers for device/server/windows ?

RE: WMI Windows Performance

community@zenoss.org — Tue, 12 Oct 2010 18:10:15 GMT

I'm having issues getting data from 2K3 servers. When I model the device, I can check the security logs and see Zenoss authenticating to the server successfully. However, the modeling shows me the following results:

[librpc/rpc/dcerpc_connect.c:329:dcerpc_pipe_connect_ncacn_ip_tcp_recv()] failed NT status (c000023d) in dcerpc_pipe_connect_ncacn_ip_tcp_recv

[librpc/rpc/dcerpc_connect.c:790:dcerpc_pipe_connect_b_recv()] failed NT status (c000023d) in dcerpc_pipe_connect_b_recv

[librpc/rpc/dcerpc_connect.c:329:dcerpc_pipe_connect_ncacn_ip_tcp_recv()] failed NT status (c000023d) in dcerpc_pipe_connect_ncacn_ip_tcp_recv

[librpc/rpc/dcerpc_connect.c:790:dcerpc_pipe_connect_b_recv()] failed NT status (c000023d) in dcerpc_pipe_connect_b_recv

[librpc/rpc/dcerpc_connect.c:329:dcerpc_pipe_connect_ncacn_ip_tcp_recv()] failed NT status (c000023d) in dcerpc_pipe_connect_ncacn_ip_tcp_recv

[librpc/rpc/dcerpc_connect.c:790:dcerpc_pipe_connect_b_recv()] failed NT status (c000023d) in dcerpc_pipe_connect_b_recv

[librpc/rpc/dcerpc_connect.c:329:dcerpc_pipe_connect_ncacn_ip_tcp_recv()] failed NT status (c000023d) in dcerpc_pipe_connect_ncacn_ip_tcp_recv

[librpc/rpc/dcerpc_connect.c:790:dcerpc_pipe_connect_b_recv()] failed NT status (c000023d) in dcerpc_pipe_connect_b_recv

Can you provide any input on how I could further troubleshoot this?

RE: WMI Windows Performance

community@zenoss.org — Wed, 13 Oct 2010 04:13:12 GMT

I had the same issue when modeling one of my devices:

2010-10-12 23:12:03

ERROR

zen.ZenModeler

Traceback (most recent call last): File "/opt/zenoss/Products/DataCollector/zenmodeler.py", line 612, in processClient datamaps = plugin.process(device, results, self.log) File "/opt/zenoss/ZenPacks/ZenPacks.community.WMIPerf_Windows-2.3.egg/ZenPacks/community/WMIPerf_Windows/modeler/plugins/community/wmi/ProductMap.py", line 65, in process om.id = self.prepId(om.setProductKey) File "/opt/zenoss/Products/DataCollector/plugins/CollectorPlugin.py", line 53, in prepId return globalPrepId(id, subchar) File "/opt/zenoss/Products/ZenUtils/Utils.py", line 615, in prepId raise ValueError('Ids can not be None') ValueError: Ids can not be None

Any ideas?

RE: WMI Windows Performance

community@zenoss.org — Thu, 14 Oct 2010 00:55:05 GMT

Is there any resolution to the following. This appears to only happen for Windows 2000 servers.

Thanks!!

RE: WMI Windows Performance

community@zenoss.org — Thu, 14 Oct 2010 20:53:01 GMT

try last version (2.4) of this ZenPack.

RE: WMI Windows Performance

community@zenoss.org — Thu, 14 Oct 2010 21:33:52 GMT

Very similar results with 2.4. Getting the following below. Windows is showing a successful authentication of the zWinUser. Setup is using a distributed collector...

[librpc/rpc/dcerpc_connect.c:329:dcerpc_pipe_connect_ncacn_ip_tcp_recv()] failed NT status (c000023d) in dcerpc_pipe_connect_ncacn_ip_tcp_recv

[librpc/rpc/dcerpc_connect.c:790:dcerpc_pipe_connect_b_recv()] failed NT status (c000023d) in dcerpc_pipe_connect_b_recv

[librpc/rpc/dcerpc_connect.c:790:dcerpc_pipe_connect_b_recv()] failed NT status (c00000b5) in dcerpc_pipe_connect_b_recv

RE: WMI Windows Performance

community@zenoss.org — Thu, 14 Oct 2010 22:13:05 GMT

C00000B5 STATUS_IO_TIMEOUT

C000023D STATUS_HOST_UNREACHABLE

first of all check connectivity. I think you trying to model remote Windows server from Zenoss master server. Try to use last version of DistributedCollectors ZenPack, it supports remote modeling.

RE: WMI Windows Performance

community@zenoss.org — Fri, 15 Oct 2010 21:28:59 GMT

thanks!!! Looks like version 2.4 fixes the following issue!

RE: WMI Windows Performance

community@zenoss.org — Wed, 20 Oct 2010 15:40:18 GMT

New DistributedCollectors ZenPack fixed the authentication issues. Devices model just fine but I am not getting any data graphed. On my remote collector, it appears that it is collecting the data:

010-10-20 11:33:55,859 INFO zen.zenperfwmi: 2 devices processed (640 datapoints)
2010-10-20 11:33:55,860 INFO zen.collector.scheduler: Tasks: 2 Successful_Runs: 20 Failed_Runs: 0 Missed_Runs: 0
2010-10-20 11:38:55,864 INFO zen.zenperfwmi: 2 devices processed (704 datapoints)
2010-10-20 11:38:55,864 INFO zen.collector.scheduler: Tasks: 2 Successful_Runs: 22 Failed_Runs: 0 Missed_Runs: 0

Any ideas what would cause the graphs to not be populated?

RE: WMI Windows Performance

community@zenoss.org — Fri, 29 Oct 2010 21:58:40 GMT

Using Zenoss 3.0.3, WMI Data Source 2.9, and WMI Perf Windows 2.4 I'm getting the following errors in zenperfwmi.log

2010-10-29 14:49:53,340 ERROR zen.WMIClient: Received NT code 0x80041010 from query: SELECT SummaryActiveUsers FROM Win32_PerfRawData_SP2010R2Agent_SP2010R2Agent

2010-10-29 14:44:55,449 ERROR zen.zenperfwmi: Device 75.101.189.1: Received NT code 0x80041010 from query: SELECT ActiveUsers FROM Win32_PerfRawData_SP2010R2Agent_SP2010R2Agent

What's happening is I will successfully model the device and Zenoss will gather data for 10-15 minutes and then it will just stop gathering. Initially i thought this was a permissions problem but after fixing all permissions I'm still seeing this error.

When I test the WMI query from the Data Source interface I get the following error:

Preparing Command...

Get WMI Instance SELECT ActiveUsers FROM Win32_PerfRawData_SP2010R2Agent_SP2010R2Agent from //75.101.189.1/root/cimv2

[librpc/rpc/dcerpc_connect.c:329:dcerpc_pipe_connect_ncacn_ip_tcp_recv()] failed NT status (c00000b5) in dcerpc_pipe_connect_ncacn_ip_tcp_recv [librpc/rpc/dcerpc_connect.c:790:dcerpc_pipe_connect_b_recv()] failed NT status (c00000b5) in dcerpc_pipe_connect_b_recv

InstanceName: root/cimv2:Win32_PerfRawData_SP2010R2Agent_SP2010R2Agent.Name="0" ActiveUsers = 0

It seems to actually return the WMI query but I can't help but think that these NT codes have something to do with it. Any ideas?

RE: WMI Windows Performance

community@zenoss.org — Fri, 29 Oct 2010 22:51:10 GMT

The error 0x80041010 means "Invalid Class". Check wmi class name. Your problem is not related to WMI Windows Performance ZenPack.

RE: WMI Windows Performance

community@zenoss.org — Thu, 11 Nov 2010 20:33:38 GMT

Is there any way that we can move the /CMI/WMI/ templates? A lot of the ZenPacks that use the WMI Data Source don't install under the CIM/WMI class - this makes it impossible to do WMI monitoring using this zen pack and others that use the Data Souce. Thanks!

RE: WMI Windows Performance

community@zenoss.org — Thu, 11 Nov 2010 20:57:56 GMT

Just copy the templates in to /CIM/WMI. Copying is called "Override" in 3.0, which makes zero sense, but that's the word that they picked to replace "Copy" with.

RE: WMI Windows Performance

community@zenoss.org — Tue, 16 Nov 2010 22:59:51 GMT

When I restart Zenoss I get about 10 occurances of the error "Could not get WMI Instance" for each of the servers monitored by CMI. After about 10 minutes it stops generating the alerts. What might be causing this? Is there any way to avoid this?

Thanks!

Mike

Zenoss 3.0.3 on openSUSE 11

RE: WMI Windows Performance

community@zenoss.org — Wed, 15 Dec 2010 11:18:35 GMT

Hello,

very great work for this ZenPack!!

I encounter an issue with some servers.

After looking at the ZenPerfWMI debug log, I found that having accents on Network interface name crashes the collector process.

The problem is that Zenoss doesn't store accent well. Hence, when you use a WMI query with it, it crashes and the other perf counter are not taken.

Here his the log :

DEBUG zen.WMIClient: Query: SELECT PacketsOutboundErrors,Name,PacketsReceivedErrors,PacketsSentPersec,PacketsReceivedPersec,BytesSentPersec,PacketsReceivedUnicastPersec,BytesReceivedPersec,PacketsSentUnicastPersec FROM Win32_PerfRawData_Tcpip_NetworkInterface WHERE Name="Connexion rÃ�seau Intel[R] PRO_1000 MT - PÃ�riphÃ�rique filtre Ã�quilibrage de la charge rÃ�seau"

2010-12-15 12:12:48,732 ERROR zen.pysamba: NT code 0xc002001b

Traceback (most recent call last):

File "/usr/local/zenoss/zenoss/ZenPacks/ZenPacks.community.WMIDataSource-2.11.egg/ZenPacks/community/WMIDataSource/Query.py", line 257, in inner

yield deferred(qctx); driver.next()

File "/usr/local/zenoss/zenoss/Products/ZenUtils/Driver.py", line 64, in result

raise ex

WMIFailure: NT code 0xc002001b

2010-12-15 12:12:48,733 DEBUG zen.WMIClient: Exception collecting query: NT code 0xc002001b

2010-12-15 12:12:48,733 ERROR zen.zenperfwmi: Device S-35B-WEB-004: NT code 0xc002001b

2010-12-15 12:12:48,734 DEBUG zen.zenperfwmi: Queueing event {'manager': 'Zenoss.anamnis-asp.fr', 'monitor': 'localhost', 'severity': 4, 'device': 'S-35B-WEB-004', 'eventClass': '/Status/Wbem', 'component': 'zenperfwmi', 'message': 'Could not get WMI Instance (NT code 0xc002001b)', 'agent': 'zenperfwmi', 'summary': 'Could not get WMI Instance'}

2010-12-15 12:12:48,734 DEBUG zen.zenperfwmi: Total of 1 queued events

2010-12-15 12:12:48,735 DEBUG zen.collector.scheduler: Task S-35B-WEB-004 finished, result: <twisted.python.failure.Failure <class 'pysamba.twisted.callback.WMIFailure'>>

The correct name would be : "Connexion réseau Intel[R] PRO_1000 MT - Périphérique filtre équilibrage de la charge réseau"

And not "Connexion rÃ�seau Intel[R] PRO_1000 MT - PÃ�riphÃ�rique filtre Ã�quilibrage de la charge rÃ�seau"

If I remove the network interface from Zenoss, it works just fine.

Regards,

Richard

RE: WMI Windows Performance

community@zenoss.org — Wed, 15 Dec 2010 11:55:15 GMT

Ok, I've tried rename the Network adapter on one the windows server by removing all non ASCI char and It works for this server.

I cannot do that on all of them, so for now, I turned of monitoring for those interfaces on Zenoss, the other counters work fine for now.

Egor, do you think this is a Zenoss related problem, or do you have a way to handle this situation?

Thanks.

Richard

RE: WMI Windows Performance

community@zenoss.org — Wed, 15 Dec 2010 16:37:52 GMT

That traceback ends at twisted (which is the python library that Zenoss uses for things like ping, snmp, etc...). It looks like twisted is choking on those characters. In my personal experience, Zenoss does not perform well when non-english characters are involved.

RE: WMI Windows Performance

community@zenoss.org — Wed, 15 Dec 2010 18:02:00 GMT

i think libasync_wmi_lib (from pysamba) library has problem with non-english characters. IEnumWbemClassObject_SmartNext_recv function return incorrect encoded characters.

Egor

RE: WMI Windows Performance

community@zenoss.org — Thu, 16 Dec 2010 09:53:31 GMT

That was what I was afraid of .

It's a shame that nowadays, non ASCII characters handling is still an issue. I'll look at Zenoss forum and tickets to see if something is in the pipe.

Regards,

Richard Esteve

RE: WMI Windows Performance

community@zenoss.org — Mon, 17 Jan 2011 23:25:00 GMT

I'm looking to upgrade from version 2.4 of the WMI Performance Monitor to version 2.5 Do you have a recommended upgrade path? Also once I install 2.5 will I be able to monitor windows mount points, or will I need more tweeking?

RE: WMI Windows Performance

community@zenoss.org — Tue, 18 Jan 2011 18:24:14 GMT

Hi vmanutah,

there's nothing special here. Simply install it and replace community.wmi.FileSystemMap collector plugin with community.wmi.VolumeFileSystemMap collector plugin within zCollectorPlugins zProperty.

RE: WMI Windows Performance

community@zenoss.org — Thu, 27 Jan 2011 18:06:06 GMT

Egor,

I replaced the c.w.FileSystemMap with the VolumeFileSystemMap, remodeled, and now my percentage used (on the two devices which I am testing) shows as WAY over 100%. For instance, one device had 7 some odd GB used out of 10GB, now is showing 59.92 Used, and unknown free, and %Util is %599. This happens for all physical drives on both boxes. The boxes are Win2k3 VMs. Thoughts?

Thanks!

Tim

RE: WMI Windows Performance

community@zenoss.org — Thu, 27 Jan 2011 18:36:07 GMT

Hi Tim,

Win32_Volume and Win32_LogicalDisk classes returns different values for BlockSize property (Win32_LogicalDisk return None). Used size calculation based on actual BlockSize values.

Easiest way to solve your problem - is delete old .rrd files.

Egor

RE: WMI Windows Performance

community@zenoss.org — Thu, 27 Jan 2011 21:46:06 GMT

Thanks for the follow up Egor. I just checked the device I added yesterday, and it seems to have cleared up after 24 hours. I am waiting until tomorrow to verify the one I changed today, then I'll just go ahead and change the rest (seeing as I have no idea *which* rrd files to delete).

Tim

RE: WMI Windows Performance

community@zenoss.org — Tue, 01 Feb 2011 18:52:04 GMT

Egor,

I waited over the weekend to see if this cleaned up, and it did not. I'm not sure *why* the other server started working correctly, but this one is not. I have deleted the entire filesystems directory and remodeled, and the drives still show up incorrectly. Next, I deleted just the rrd files associated with usedBlocks for the drives, then remodeled, and still no-go. I'm not sure where to go next, as I believe I have done what you requested. Did I misunderstand, and if not, is there anything else I can do?

Thanks again,

Tim

RE: WMI Windows Performance

community@zenoss.org — Tue, 01 Feb 2011 19:01:45 GMT

I am using the volume filesystem plugin to monitor over 50 windows servers and I'm not seeing any problems at all with what is being reported for the filesystems on any of them. It sounds like the issue is WMI itself. Have you tried restarting the WMI service or rebooting the server? I had to reboot a server this morning to get WMI to respond again since it had stopped, and everything else I tried didn't work.

RE: WMI Windows Performance

community@zenoss.org — Thu, 03 Feb 2011 19:11:13 GMT

Hi Egor,

I've upgraded this ZenPack to v2.5 and am now seeing a lot of alerts 'Could not get WMI Instance'. When I look at the details I see a see there are from query's that are failing across many different machines.

For example:

2011-02-03 12:26:29,851 ERROR zen.WMIClient: Received NT code 0x80041010 from query: SELECT FreeSpace FROM C

2011-02-03 12:26:30,876 ERROR zen.WMIClient: Received NT code 0x80041010 from query: SELECT FreeSpace FROM C

2011-02-03 12:26:30,951 ERROR zen.WMIClient: Received NT code 0x80041017 from query: SELECT NetConnectionStatus FROM 9

2011-02-03 12:26:31,883 ERROR zen.WMIClient: Received NT code 0x80041010 from query: SELECT FreeSpace FROM C

2011-02-03 12:26:31,884 ERROR zen.WMIClient: Received NT code 0x80041010 from query: SELECT FreeSpace FROM C

2011-02-03 12:26:31,921 ERROR zen.WMIClient: Received NT code 0x80041017 from query: SELECT NetConnectionStatus FROM 9

2011-02-03 12:26:32,100 ERROR zen.WMIClient: Received NT code 0x80041017 from query: SELECT NetConnectionStatus FROM 13

I've made sure our WMI Data Source ZenPack is also up to date (v2.11) and have remodeled devices.

Any idea what may be causing this or how it could be fixed?

Thanks,

John

RE: WMI Windows Performance

community@zenoss.org — Thu, 03 Feb 2011 19:15:02 GMT

This is the WMI Data Source ZenPack we're using:

ZenPacks.community.WMIDataSource-2.11.egg.zip

Should we be using the following instead?

ZenPacks.community.WMIDataSource-2.5-py2.4.egg.zip

We are using python 2.4.4.

RE: WMI Windows Performance

community@zenoss.org — Thu, 03 Feb 2011 19:37:18 GMT

You restarted Zenoss after installing the pack I hope?

RE: WMI Windows Performance

community@zenoss.org — Thu, 03 Feb 2011 20:00:42 GMT

Hi johnnynoc,

WMIDataSource version is OK, it compatible with both Zenoss 2.5.2 and Zenoss 3.0.X.

try to run zenpwerfwmi daemon with DEBUG loging level:

zenperfwmi run -v10 -d yourwindowsserver

RE: WMI Windows Performance

community@zenoss.org — Mon, 07 Feb 2011 20:01:26 GMT

I'll try this when I can safely bring down Zenoss. The boss prefers after hours.

I did find a fix for some of the boxes exhibiting this behavior. There is additional software to install called "Windows Installer Provider".

http://msdn.microsoft.com/en-us/library/aa394523(v=VS.85).aspx

Queries were failing because of a lack of the W32_Product class and after this was installed they were returning successfully.

RE: WMI Windows Performance

community@zenoss.org — Thu, 10 Feb 2011 17:46:35 GMT

Ryan (and Egor),

I restarted the services and the server. I'm getting data, but it's not the correct data. My drives are continually showing more usage than they have capacity (so, 600% utilization). I have tried to delete the .rrd files for the volumes (from /opt/zenoss/perf/Devices/<deviceIP>/os/filesystems/<DriveLetter>/usedBlocks_usedBlocks.rrd), then re-model, but when the volume gets recreated, the same thing happens. Also, I have even tried removing the WHOLE filesystems directory (well, renaming it), then remodeling it, and the same thing happens - 600% utilization.

I have tried removing the volume from the Zenoss interface, then remodeling, nothing. I have tried removing the volumd from the Zenoss interface, removing the .rrd files, then remodeling, again, nothing. At this point, I'm stumped where it's getting it's data for the volume size, and why one box works, while another doesn't.

Any help would be GREATLY appreciated. Thanks again guys for all your hard work!

Tim

RE: WMI Windows Performance

community@zenoss.org — Thu, 10 Feb 2011 20:36:59 GMT

Hi Tim,

check zFileSystemSizeOffset zProperty.

Egor

RE: WMI Windows Performance

community@zenoss.org — Fri, 11 Feb 2011 16:18:46 GMT

It's set to 1.0 (on both the working device, and the non-working device). What should it be set to?

Tim

RE: WMI Windows Performance

community@zenoss.org — Fri, 11 Feb 2011 22:10:07 GMT

Hi Tim,

do you have community.wmi.FileSystemMap modeler plugin in zCollectorPlugins list?

Egor

RE: WMI Windows Performance

community@zenoss.org — Mon, 14 Feb 2011 18:33:36 GMT

Egor,

No, only the community.wmi.VolumeFileSystemMap.

Tim

RE: WMI Windows Performance

community@zenoss.org — Mon, 14 Feb 2011 19:52:57 GMT

Hi Tim,

check with ZMI the real values of attributes 'blockSize' and 'totalBlocks' for your volumes.

Egor

RE: WMI Windows Performance

community@zenoss.org — Thu, 17 Feb 2011 16:09:29 GMT

Egor,

Removing the community.wmi.FileSystemMap plugin seemed to fix an issue I was having with disks being reported as overutilized.

However the disk utilization stilll doesn't seem to match.

I was poking through the Zope Interface hoping to find these attributes but they were nowhere to be found. Am I looking in the wrong place?

Thanks,

John

RE: WMI Windows Performance

community@zenoss.org — Thu, 17 Feb 2011 16:34:12 GMT

Hi John,

yourDevice/os/filesystems/filesystemName

Egor

RE: WMI Windows Performance

community@zenoss.org — Thu, 24 Feb 2011 12:32:24 GMT

We are experiencing the same problem as cdb0788.
Maybe we should explain our test-enviroment:

We have a Hub which only acts as a hub

Then we have a RemoteCollector which collects the data from a WindowsSBS2003. (For the DistributedSetup we have used Egor Puzanovs ZenPack "Distributed Collectors ver. 1.7" on Zenoss 3.1.0)

Our zenperfwmi.log file on the RemoteCollector seems to collect the data correct but nothing is being populated...

2011-02-24 13:59:39,268 INFO zen.collector.scheduler: Tasks: 1 Successful_Runs: 0 Failed_Runs: 0 Missed_Runs: 0
2011-02-24 14:04:39,272 INFO zen.zenperfwmi: 1 devices processed (52 datapoints)
2011-02-24 14:04:39,273 INFO zen.collector.scheduler: Tasks: 1 Successful_Runs: 1 Failed_Runs: 0 Missed_Runs: 0
2011-02-24 14:09:39,278 INFO zen.zenperfwmi: 1 devices processed (78 datapoints)
2011-02-24 14:09:39,279 INFO zen.collector.scheduler: Tasks: 1 Successful_Runs: 2 Failed_Runs: 0 Missed_Runs: 0
2011-02-24 14:14:39,284 INFO zen.zenperfwmi: 1 devices processed (104 datapoints)
2011-02-24 14:14:39,284 INFO zen.collector.scheduler: Tasks: 1 Successful_Runs: 3 Failed_Runs: 0 Missed_Runs: 0

Any insight on this issue??

RE: WMI Windows Performance

community@zenoss.org — Thu, 24 Feb 2011 21:17:59 GMT

Hi Lumskekonig,

is this issue wmi specific?

If yes - try to run zenperfwmi daemon with debug option:

zenperfwmi run -v 10 -d yourserver

if not - then check zenrender daemon connectivity.

Egor

RE: WMI Windows Performance

community@zenoss.org — Fri, 25 Feb 2011 09:32:09 GMT

HeyEgor...

Thanks for your support...

I solved by running a debug on the zenrender... i realized that it was listening on the wrong port.. so i change the render port (RenderURL) and updated the remote collector and it worked...

Thank you...

RE: WMI Windows Performance

community@zenoss.org — Sun, 06 Mar 2011 23:57:24 GMT

I'm finding that the Windows Event log is not monitored for some reason. If I move the devices back to the "/Server/Windows/8 Processors" (ZenPacks.Nova.Windows.SNMPPerfMonitor 1.6) I see a bunch of Event log alerts coming in.

Is the "ZenPacks.community.WMIPerf_Windows" supposed to support the Windows Event log? We really want to use this ZenPack since it does not require SNMP.

Thanks!

RE: WMI Windows Performance

community@zenoss.org — Mon, 07 Mar 2011 05:18:43 GMT

Event log monitoring is built in to Zenoss core, it is not a function of this ZenPack. You need to go in to the zProperties and toggle zWinEventlog to true.

RE: WMI Windows Performance

community@zenoss.org — Fri, 11 Mar 2011 22:29:11 GMT

Egor, this zenpack is awesome,

But I tried it on a big horse of a windows server that has some huge logical drives mounted like so:

c:\A

\bigdrive mounted here

\another bigdrive mounted here

But the wmi reports the size of the c: drive but not including the logical drives.

Snmp does the same thing, is there a possible solution?

RE: WMI Windows Performance

community@zenoss.org — Fri, 11 Mar 2011 22:33:13 GMT

Look under Change History on this page: "2.5 added VolumeFileSystemMap (windows mount points) collector plugin. Thanks Ryan Matte for the suggestion."

You need to use the VolumeFileSystemMap collector plugin instead of the FileSystemMap plugin.

RE: WMI Windows Performance

community@zenoss.org — Fri, 11 Mar 2011 22:42:21 GMT

woohoo! thanks ryan.

I will try that next week.

RE: WMI Windows Performance

community@zenoss.org — Fri, 11 Mar 2011 22:47:23 GMT

No problem.

RE: WMI Windows Performance

community@zenoss.org — Mon, 14 Mar 2011 19:55:33 GMT

Hi,

I am using Zenoss 3, WMIDataSource 2.11 and WMIPerfWindows 2.5.

For all my devices i have problem monitoring used disk space. I can se the size of the disk, but i cant se how much space is used or available.

I have tried to change the collector from FileSystemMap to VolumeFileSystemMap, but that doesn't seams to help or i do something wrong.

Also if i select a device and click on Graphs the browser dies and the following information show up in the log file event.log: ERROR zen.MinMaxCheck User-supplied Python expression (here.totalBlocks * .9) for maximum value caused error: ['usedBlocks_usedBlocks']. I guess this error is related to my first problem...

What else can i do to solve the problem. Please help.

Thanks

RE: WMI Windows Performance

community@zenoss.org — Mon, 14 Mar 2011 22:45:12 GMT

Hi slingeland,

do NOT bind components templates to device.

Egor

RE: WMI Windows Performance

community@zenoss.org — Tue, 15 Mar 2011 10:48:30 GMT

Hi Egor,

I am very new to Zenoss..

I installed the two Zenpacks and and added my devices to /CIM/WMI. Is that wrong?

What do you mean with not binding templates to devices?

Thanks

RE: WMI Windows Performance

community@zenoss.org — Tue, 15 Mar 2011 15:31:01 GMT

If you are very new to Zenoss then I would suggest reading the Zenoss Admin Guide top to bottom (which is what I did when I started with Zenoss and it was extremely useful). Also, he wasn't saying don't bind templates in general, he was saying don't bind component templates to devices. For example, the FileSystem template is a component template as it applies to filesystems, not to the device itself. The UI should make it impossible to bind anything except device templates to devices. All component templates get bound automatically to the component to which they are relevant.

RE: WMI Windows Performance

community@zenoss.org — Tue, 15 Mar 2011 18:43:26 GMT

Ryan,

Thanks for the advice. I know i should read the admin guide. Guess i was a little bit lazy this time :-)

I uninstalled the Zenpacks and installed them again and now it works great.

Thanks

RE: WMI Windows Performance

community@zenoss.org — Mon, 28 Mar 2011 22:08:47 GMT

Zenoss 3.1.0 on SUSE 11

I'm also getting these errors (see below) after installing 2.5 (I did restart zenoss after installing the zenpack)

Has this changed from 2.4 to 2.5? Is "Windows Installer Provider" mentioned by johnnynoc now used by the 2.5 release where it was not used by the 2.4 release? In the 2.4 release I would get these messages occasionally, but now it's every minute on many (not all) /CIM/WIM servers.

When I use my WMI Explorer, I don't see WMI classes like "C", "D", etc.. on any of my servers.

2011-03-28 15:02:53,310 ERROR zen.WMIClient: Received NT code 0x80041010 from query: SELECT FreeSpace FROM C
2011-03-28 15:02:56,588 ERROR zen.WMIClient: Received NT code 0x80041010 from query: SELECT FreeSpace FROM D
2011-03-28 15:02:58,359 ERROR zen.WMIClient: Received NT code 0x80041010 from query: SELECT FreeSpace FROM C
2011-03-28 15:02:58,368 ERROR zen.WMIClient: Received NT code 0x80041010 from query: SELECT FreeSpace FROM D
2011-03-28 15:03:01,581 ERROR zen.WMIClient: Received NT code 0x80041017 from query: SELECT NetConnectionStatus FROM 1

RE: WMI Windows Performance

community@zenoss.org — Wed, 30 Mar 2011 00:58:48 GMT

I downgraded (restored) back to the 2.4 release and the problem above is not occuring anymore. Any idea why this is happening?

RE: WMI Windows Performance

community@zenoss.org — Mon, 04 Apr 2011 08:14:13 GMT

i was getting some servers not responding to different queries e.g. invalid class errors after some googling i found that if run these commands from the server it should "rebuild your wmi structure"

wmiadap /f

winmgmt /clearadap

this worked for me within about 10 minutes data started showing up on the graphs.

RE: WMI Windows Performance

community@zenoss.org — Fri, 29 Apr 2011 22:59:45 GMT

So I can set "zSnmpMonitorIgnore = TRUE" at the /CIM/WMI level without losing any monitoring features? I notice that it's FALSE by default. Any harm turning it on for all /CIM/WMI devices?

RE: WMI Windows Performance

community@zenoss.org — Mon, 02 May 2011 13:30:34 GMT

It's common sense. If it's disabled, no SNMP monitoring will be permitted for that class. If it's enabled, it will be permitted.

RE: WMI Windows Performance

community@zenoss.org — Mon, 02 May 2011 16:59:28 GMT

I was asking if setting zSnmpMonitorIgnore = TRUE at the /CMI/WMI level, will any functionality or monitoring be effected for /CIM/WMI monitored servers. Just wanted to make sure before making the change.

RE: WMI Windows Performance

community@zenoss.org — Mon, 02 May 2011 17:01:04 GMT

No, all WMI monitoring will continue to function normally (I've done this type of thing on several servers that I'm monitoring and it's worked fine).

RE: WMI Windows Performance

community@zenoss.org — Mon, 02 May 2011 17:10:01 GMT

Great! Thanks!!

RE: WMI Windows Performance

community@zenoss.org — Wed, 11 May 2011 19:39:46 GMT

Hi All,

I am monitoring the WINDOWS server 2000 through WMI . I put the device in /CIM/WMI/2000 . I gave the WMI username and Passwd as well . When i am modeling the device i am getting the Ip address error . Even i deleted the device and checked the IP address if i am using the same ip for the other device, but i am not using the same ip . I am using the IP address only for the devices which i am modeling . This is happening for only Win2000 devices . I am using Zenoss 3.10 . I am using the below WMI datasource . Please help me out in this .Its an emergency requirement for me

ZenPacks.community.WMIDataSource (/opt/zenoss/ZenPacks/ZenPacks.community.WMIDataSource-2.11.egg)
ZenPacks.community.WMIPerf_Windows (/opt/zenoss/ZenPacks/ZenPacks.community.WMIPerf_Windows-2.5.egg).

Error:

File "/opt/zenoss/Products/ZenHub/PBDaemon.py", line 75, in inner
return callable(*args, **kw)
File "/opt/zenoss/Products/ZenHub/zenhubworker.py", line 116, in remote_execute
return runOnce()
File "/opt/zenoss/Products/ZenHub/zenhubworker.py", line 111, in runOnce
res = m(*args, **kw)
File "/opt/zenoss/Products/ZenHub/PBDaemon.py", line 83, in inner
traceback.format_exc())
RemoteException: Remote exception: : IP address conflict for IP: Traceback (most recent call last):
File "/opt/zenoss/Products/ZenHub/PBDaemon.py", line 75, in inner
return callable(*args, **kw)
File "/opt/zenoss/Products/ZenHub/services/ModelerService.py", line 109, in remote_applyDataMaps
if adm._applyDataMap(device, map):
File "/opt/zenoss/Products/DataCollector/ApplyDataMap.py", line 198, in _applyDataMap
changed = self._updateRelationship(tobj, datamap)
File "/opt/zenoss/Products/DataCollector/ApplyDataMap.py", line 261, in _updateRelationship
objchange, obj = self._createRelObject(device, objmap, rname)
File "/opt/zenoss/Products/DataCollector/ApplyDataMap.py", line 445, in _createRelObject
return self._updateObject(remoteObj, objmap) or changed, remoteObj
File "/opt/zenoss/Products/DataCollector/ApplyDataMap.py", line 370, in _updateObject
setter(*args)
File "/opt/zenoss/Products/ZenModel/IpInterface.py", line 302, in setIpAddresses
self.addIpAddress(ip)
File "/opt/zenoss/Products/ZenModel/IpInterface.py", line 241, in addIpAddress
ipobj = networks.findIp(ip)
File "/opt/zenoss/Products/ZenModel/IpNetwork.py", line 463, in findIp
raise IpAddressConflict( "IP address conflict for IP: %s" % ip )
IpAddressConflict: IP address conflict for IP:
Traceback (most recent call last):
File "/opt/zenoss/Products/DataCollector/zenmodeler.py", line 658, in processClient
if driver.next():
File "/opt/zenoss/Products/ZenUtils/Driver.py", line 64, in result
raise ex
RemoteException: Traceback (most recent call last):
File "/opt/zenoss/Products/ZenHub/PBDaemon.py", line 75, in inner
return callable(*args, **kw)
File "/opt/zenoss/Products/ZenHub/zenhubworker.py", line 116, in remote_execute
return runOnce()
File "/opt/zenoss/Products/ZenHub/zenhubworker.py", line 111, in runOnce
res = m(*args, **kw)
File "/opt/zenoss/Products/ZenHub/PBDaemon.py", line 83, in inner
traceback.format_exc())
RemoteException: Remote exception: : IP address conflict for IP: Traceback (most recent call last):
File "/opt/zenoss/Products/ZenHub/PBDaemon.py", line 75, in inner
return callable(*args, **kw)
File "/opt/zenoss/Products/ZenHub/services/ModelerService.py", line 109, in remote_applyDataMaps
if adm._applyDataMap(device, map):
File "/opt/zenoss/Products/DataCollector/ApplyDataMap.py", line 198, in _applyDataMap
changed = self._updateRelationship(tobj, datamap)
File "/opt/zenoss/Products/DataCollector/ApplyDataMap.py", line 261, in _updateRelationship
objchange, obj = self._createRelObject(device, objmap, rname)
File "/opt/zenoss/Products/DataCollector/ApplyDataMap.py", line 445, in _createRelObject
return self._updateObject(remoteObj, objmap) or changed, remoteObj
File "/opt/zenoss/Products/DataCollector/ApplyDataMap.py", line 370, in _updateObject
setter(*args)
File "/opt/zenoss/Products/ZenModel/IpInterface.py", line 302, in setIpAddresses
self.addIpAddress(ip)
File "/opt/zenoss/Products/ZenModel/IpInterface.py", line 241, in addIpAddress
ipobj = networks.findIp(ip)
File "/opt/zenoss/Products/ZenModel/IpNetwork.py", line 463, in findIp
raise IpAddressConflict( "IP address conflict for IP: %s" % ip )
IpAddressConflict: IP address conflict for IP:
2011-05-11 14:22:17,500 ERROR zen.ZenModeler: Client ATL518 finished with message: [Failure instance: Traceback (failure with no frames): : Traceback (most recent call last):
File "/opt/zenoss/Products/ZenHub/PBDaemon.py", line 75, in inner
return callable(*args, **kw)
File "/opt/zenoss/Products/ZenHub/zenhubworker.py", line 116, in remote_execute
return runOnce()
File "/opt/zenoss/Products/ZenHub/zenhubworker.py", line 111, in runOnce
res = m(*args, **kw)
File "/opt/zenoss/Products/ZenHub/PBDaemon.py", line 83, in inner
traceback.format_exc())
RemoteException: Remote exception: : IP address conflict for IP: Traceback (most recent call last):
File "/opt/zenoss/Products/ZenHub/PBDaemon.py", line 75, in inner
return callable(*args, **kw)
File "/opt/zenoss/Products/ZenHub/services/ModelerService.py", line 109, in remote_applyDataMaps
if adm._applyDataMap(device, map):
File "/opt/zenoss/Products/DataCollector/ApplyDataMap.py", line 198, in _applyDataMap
changed = self._updateRelationship(tobj, datamap)
File "/opt/zenoss/Products/DataCollector/ApplyDataMap.py", line 261, in _updateRelationship
objchange, obj = self._createRelObject(device, objmap, rname)
File "/opt/zenoss/Products/DataCollector/ApplyDataMap.py", line 445, in _createRelObject
return self._updateObject(remoteObj, objmap) or changed, remoteObj
File "/opt/zenoss/Products/DataCollector/ApplyDataMap.py", line 370, in _updateObject
setter(*args)
File "/opt/zenoss/Products/ZenModel/IpInterface.py", line 302, in setIpAddresses
self.addIpAddress(ip)
File "/opt/zenoss/Products/ZenModel/IpInterface.py", line 241, in addIpAddress
ipobj = networks.findIp(ip)
File "/opt/zenoss/Products/ZenModel/IpNetwork.py", line 463, in findIp
raise IpAddressConflict( "IP address conflict for IP: %s" % ip )
IpAddressConflict: IP address conflict for IP:
]

Thanks,

Subbu

RE: WMI Windows Performance

community@zenoss.org — Mon, 23 May 2011 12:39:49 GMT

Hi guys...

i'm a bit confused, does this zenpack provide Installed software list (the Software tab) or not?

I'm managing a Win2003 server (via zenoss 3.1.0), WMI only (no SNMP monitoring) and of course everything is working okay, except Software list is empty.

Did i miss a statement somewhere that this zenpack doesn't acquire a installed SW list?

Cause if it should, all my win servers have an empty SW list, how to enable it then?

regards

marko

RE: WMI Windows Performance

community@zenoss.org — Wed, 25 May 2011 17:34:14 GMT

Egor, I am curious what got changed here. I got an email saying you changed stuff on this page, but it didnt say what.

-Nick Yeates-

Community Support Engineer

Zenoss Inc.

nyeates@zenoss.com

RE: WMI Windows Performance

community@zenoss.org — Wed, 25 May 2011 17:47:03 GMT

This was the update Nick:

"2.6 BETA1 (2.5.80) WinService Performance monitoring, bugfixes."

RE: WMI Windows Performance

community@zenoss.org — Wed, 25 May 2011 17:51:49 GMT

It shows installed software for me.

RE: WMI Windows Performance

community@zenoss.org — Thu, 26 May 2011 08:31:58 GMT

What is the required Modeller Plugin to get the sowftware list? I want to double-check it.

RE: WMI Windows Performance

community@zenoss.org — Thu, 26 May 2011 14:09:36 GMT

I don't think it has anything to do with modeler plugins. I've looked through all modeler plugins including the snmp ones and there are none listed for software (or anything similar). It's basically just a report that is run when you click on the tab (or if you use any of the reports in the reports section). Egor probably coded a custom one for the /CIM/WMI class.

RE: WMI Windows Performance

community@zenoss.org — Thu, 26 May 2011 17:41:51 GMT

Hi Marko,

community.wmi.ProductMap

RE: WMI Windows Performance

community@zenoss.org — Fri, 27 May 2011 07:55:29 GMT

OK, thank Egor, this modeller plugin is attached to my device. When modelling this device i get:

2011-05-27 09:47:08,036 INFO zen.ZenModeler: Connecting to localhost:8789

2011-05-27 09:47:09,220 INFO zen.ZenModeler: Connected to ZenHub

2011-05-27 09:47:10,076 INFO zen.ZenModeler: Collecting for device X.X.X.X

2011-05-27 09:47:10,328 INFO zen.ZenModeler: No WMI plugins found for X.X.X.X

2011-05-27 09:47:10,344 INFO zen.ZenModeler: Python collection device X.X.X.X

2011-05-27 09:47:10,344 INFO zen.ZenModeler: plugins: community.wmi.NewDeviceMap, community.wmi.DeviceMap, community.wmi.ProcessorMap, community.wmi.InterfaceMap, community.wmi.FileSystemMap, community.wmi.ProcessMap, community.wmi.RouteMap, community.wmi.DiskDriveMap, community.wmi.WinServiceMap, community.wmi.ProductMap, community.wmi.VolumeFileSystemMap

2011-05-27 09:47:10,373 INFO zen.ZenModeler: No command plugins found for X.X.X.X

2011-05-27 09:47:10,374 INFO zen.ZenModeler: SNMP monitoring off for X.X.X.X

2011-05-27 09:47:10,387 INFO zen.ZenModeler: Portscan collector method for device X.X.X.X

2011-05-27 09:47:10,387 INFO zen.ZenModeler: plugins: zenoss.portscan.IpServiceMap

2011-05-27 09:47:11,081 INFO zen.ZenModeler: Running 2 clients

2011-05-27 09:47:42,303 INFO zen.PortscanClient: portscan client finished collection for X.X.X.X

2011-05-27 09:47:42,305 INFO zen.ZenModeler: processing Ip Services for device X.X.X.X

2011-05-27 09:47:42,611 INFO zen.ZenModeler: No change in configuration detected

2011-05-27 09:47:42,828 ERROR zen.WMIClient: Received NT code 0x80041010 from query: SELECT * FROM Win32_Product

2011-05-27 09:47:44,078 INFO zen.PythonClient: Python client finished collection for X.X.X.X

2011-05-27 09:47:44,081 INFO zen.ZenModeler: processing community.wmi.RouteMap for device X.X.X.X

2011-05-27 09:47:44,086 ERROR zen.ZenModeler: Received NT code 0x80041010 from query: SELECT * FROM Win32_Product

2011-05-27 09:47:44,088 INFO zen.ZenModeler: processing community.wmi.VolumeFileSystemMap for device X.X.X.X

2011-05-27 09:47:44,093 INFO zen.ZenModeler: processing community.wmi.ProcessMap for device X.X.X.X

2011-05-27 09:47:44,100 INFO zen.ZenModeler: processing community.wmi.DiskDriveMap for device X.X.X.X

2011-05-27 09:47:44,102 INFO zen.ZenModeler: Processing WinServices for device X.X.X.X

2011-05-27 09:47:44,126 INFO zen.ZenModeler: processing community.wmi.InterfaceMap for device X.X.X.X

2011-05-27 09:47:44,130 INFO zen.ZenModeler: processing community.wmi.NewDeviceMap for device X.X.X.X

2011-05-27 09:47:44,131 INFO zen.ZenModeler: processing community.wmi.DeviceMap for device X.X.X.X

2011-05-27 09:47:44,133 INFO zen.ZenModeler: processing community.wmi.FileSystemMap for device X.X.X.X

2011-05-27 09:47:44,136 INFO zen.ZenModeler: processing community.wmi.ProcessorMap for device X.X.X.X

2011-05-27 09:47:46,999 INFO zen.ZenModeler: Changes in configuration applied

2011-05-27 09:47:47,147 INFO zen.ZenModeler: Scan time: 37.07 seconds

2011-05-27 09:47:47,168 INFO zen.ZenModeler: Daemon ZenModeler shutting down

So, there is an Error:

2011-05-27 09:47:42,828 ERROR zen.WMIClient: Received NT code 0x80041010 from query: SELECT * FROM Win32_Product

2011-05-27 09:47:44,086 ERROR zen.ZenModeler: Received NT code 0x80041010 from query: SELECT * FROM Win32_Product

I don't know why since WMI access to this device is otherwise OK (all other WMI info is gathered).

Any ideas?

RE: WMI Windows Performance

community@zenoss.org — Tue, 31 May 2011 13:00:37 GMT

We are getting the "Could not get WMI Instance" messages all the time too, and we are using WMIDataSource version 2.9. Any suggestions about how to get rid of these messages?

RE: WMI Windows Performance

community@zenoss.org — Tue, 31 May 2011 13:44:30 GMT

dbman, as someone who has a lot of experience with monitoring WMI I can tell you that in general, WMI is an unreliable protocol. Different servers behave differently based on the version of Windows/WMI that they are running. There is a WIDE range of possible causes for you receiving those errors. I've seen various causes such as WMI intermittently not responding, WMI counters suddenly disappearing (and only returning when a reboot is performed), WMI ceasing to function and never recovering no matter what I tried, etc... The best method that we've found is to use SNMP to monitor pretty much everything and then use WMI to monitor very specific things such as Windows Services, and performance counters for things like Exchange when required. It's much more stable and much more reliable to do it that way.

RE: WMI Windows Performance

community@zenoss.org — Tue, 31 May 2011 14:17:33 GMT

Thanks, Ryan, for your quick response. We are now discussing how to parse out the stuff accordingly: snmp for most stuff, and use WMI for a small subset of specific stuff.

Curious, could the SQL response time from SQL Server be a part of the problem that might be fixed via recreating indexes and tuning the tables better?

RE: WMI Windows Performance

community@zenoss.org — Tue, 31 May 2011 17:10:59 GMT

Are you referring to MS SQL being monitored via WMI?

RE: WMI Windows Performance

community@zenoss.org — Tue, 31 May 2011 19:06:03 GMT

Yes, specifically, the TERMSESSIONS datasource:
SELECT ActiveSessions,InactiveSessions,TotalSessions FROM Win32_PerfFormattedData_LocalSessionManager_TerminalServices from //10.1.17.26/<userid>/<password>

RE: WMI Windows Performance

community@zenoss.org — Tue, 31 May 2011 19:13:16 GMT

That is very possible, I remember one of our Windows engineers mentioning that sort of problem at some point in the past.

RE: WMI Windows Performance

community@zenoss.org — Tue, 21 Jun 2011 06:40:16 GMT

The version 2.5.8 return "Invaild class" for Windows 2000 and Windows 2003 by below WQL,

"SELECT UserModeTime,PageFaults,VirtualSize,CommandLine,KernelModeTime,ThreadCount,PageFileUsage FROM Win32_Process". It's fine for Windows 2008.

As I found the CommandLine is not exists in Windows 2000 and Windows 2003.

How could I edit the template for monitoring Windows 2000 and Windows 2003?

Thanks.

RE: WMI Windows Performance

community@zenoss.org — Tue, 21 Jun 2011 17:25:59 GMT

Hi kimalto,

first of all, Windows 2003 Win32_Process class has CommandLine Property and Templates was tested against Windows 2003 R2 Servers.

1. copy /CIM/WMI/OSProcess template to /CIM/WMI/Win2000 device class

2. replace WQL query with: Win32_Process.Name="${here/procName}"

3. copy /CIM/WMI/WinService template to /CIM/WMI/Win2000 device class

4. replace WQL query with: Win32_Process.Name="${here/procName}"

Egor

RE: WMI Windows Performance

community@zenoss.org — Wed, 22 Jun 2011 01:15:14 GMT

Hi Egor,

Thank you for your reply, I've found the CommandLine Property in other Windows Server 2003, sorry for that.

I've copied the template to /CIM/WMI/Win2000 and create new template OSProcess(create data source, data points one by one)

and got this error for Test Against a Device

=======================================================

Preparing Command...

exception while executing command

type: value: exceptions must be classes or instances, not NoneType

Exception while performing command:

Traceback (most recent call last):

File "/opt/zenoss/Products/ZenUI3/browser/command.py", line 120, in stream

self.reportError)

File "/opt/zenoss/ZenPacks/ZenPacks.community.WMIDataSource-2.11.egg/ZenPacks/community/WMIDataSource/datasources/WMIDataSource.py", line 206, in testDataSourceAgainstDevice

write('DONE in %s seconds' % long(time.time() - start))

UnboundLocalError: local variable 'start' referenced before assignment

=======================================================

The Data Source of OSProcess is:

Namespace: root/cimv2

Query:Win32_Process.Name="${here/procName}"

Thanks.

RE: WMI Windows Performance

community@zenoss.org — Wed, 22 Jun 2011 17:51:25 GMT

SAVE data source before testing.

RE: WMI Windows Performance

community@zenoss.org — Sun, 26 Jun 2011 18:17:46 GMT

Hi!

For others, if you are getting DCERPC_FAULT_ACCESS_DENIED error, try this: http://community.zenoss.org/docs/DOC-3392#comment-4096 .

RE: WMI Windows Performance

community@zenoss.org — Mon, 27 Jun 2011 16:15:43 GMT

Hi Egor,

Finally, I've copied the OSProcess and WinService templates by editing the objects.xml the same error occurs.

Thanks for Help.

=======================================================

Preparing Command...

exception while executing command

type: value: exceptions must be classes or instances, not NoneType

Exception while performing command:

Traceback (most recent call last):

File "/opt/zenoss/Products/ZenUI3/browser/command.py", line 120, in stream

self.reportError)

File "/opt/zenoss/ZenPacks/ZenPacks.community.WMIDataSource-2.11.egg/ZenPacks/community/WMIDataSource/datasources/WMIDataSource.py", line 206, in testDataSourceAgainstDevice

write('DONE in %s seconds' % long(time.time() - start))

UnboundLocalError: local variable 'start' referenced before assignment

=======================================================

I've got another error when I viewing the zenperfwmi, some of the Windows 2003 servers shown

error like this:

ERROR zen.zenperfwmi: Device MACHINENAME: Received NT code 0x80041017 from query: SELECT UserModeTime,PageFaults,VirtualSize,PageFileUsage,CommandLine,ThreadCount,KernelModeTime FROM Win32_Process WHERE CommandLine=""C:\\Program Files\\Dell\\SysMgt\\iws\\bin\\win32\\dsm_om_connsvc32.exe""

All errors are 0x80041017 but not all server pointing to same .exe, other error are CA Uniagent.exe, or Tomcat6.

How can I resolve this?

Many thanks.

RE: WMI Windows Performance

community@zenoss.org — Mon, 27 Jun 2011 16:42:55 GMT

this problem already solved here http://community.zenoss.org/people/bigegor/blog/2011/06/14/sqldatasource-zenpack-20-beta4

RE: WMI Windows Performance

community@zenoss.org — Mon, 27 Jun 2011 17:03:49 GMT

I confuse that the solution is.

I need to install the SQLDataSource or downgrade the WMI DataSource to 2.83 from the link of your blog or change the WINRM?

Thanks.

RE: WMI Windows Performance

community@zenoss.org — Mon, 27 Jun 2011 19:48:42 GMT

this is how BETA4 differ from BETA3. carefully read last 3 posts from here http://community.zenoss.org/people/bigegor/blog

RE: WMI Windows Performance

community@zenoss.org — Tue, 28 Jun 2011 02:48:35 GMT

Hi Egor,

The zenperfwmi is missing in the Zenoss Daemon after install the SQLDataSource 1.96 and WMIDataSource 2.83

RE: WMI Windows Performance

community@zenoss.org — Wed, 29 Jun 2011 19:38:54 GMT

its Ok. New WMIDataSource ZenPack used zenperfsql daemon from SQLDataSource ZenPack.

RE: WMI Windows Performance

community@zenoss.org — Wed, 13 Jul 2011 23:12:54 GMT

Hi,

i've been testing wmidatasource via sqldatasource, and when zenpersql tries to fetch performance values from some monitored winservices, it fails as an invalid query. Just for the record, changing the " character for apostrophe is working perfect on the PROCESS DataSource from WinServices applied on CIM/WMI, i mean, changing:

Win32_Process.CommandLine="${here/pathName}"

for

Win32_Process.CommandLine='${here/pathName}'

But still i have a problem:

I'm trying to get Windows PerfData through SNMP and winservices data through WMI... is there any way to do it? /Server/Windows has its own SNMP template, asking for the metrics we want, and the winservice template, copied from /CIM/WMI. As modeler we added community.wmi.winservice, not the zenoss.wmi.winservice.

On this situation there are two issues I don't get to understand..

I get a "Could not fetch data from source (Received error (Connect on x.x.x.x (WERR_ACCESS_DENIED)) from query: SELECT LoadPercentage FROM Win32_Processor)", when wmic works like a charm with the credentials on zwinuser/pass and there's other data retrieved that is working. Also "SELECT LoadPercentage FROM Win32_Processor" is a CPU metric that should not be asked via wmiservice, but WMIDevice, which i'm not using and it's not assigned anywhere!

So, briefing:

1. Is it possible to use /Server/Windows, asking via SNMP the main metrics and modeling/collecting via winservice every winservice found at the host?

2. If question 1's answer is no... is it possible to speak only-wmi with that host, still at /Server/Windows?

3. Winservice::Process Datasource should work okay with the original definition Win32_Process.CommandLine="${here/pathName}", instead of the one with apostrophes?

Thanks in advance for your answer, Egor

RE: WMI Windows Performance

community@zenoss.org — Thu, 14 Jul 2011 10:32:49 GMT

Hi!

I've used your recommendation. Every monitorized windows on zenoss is now on /Server/Windows/WMI, and every template applied to /CIM/WMI has ben copied to /Server/Windows/WMI.

I don't know if i should call them bugs or is there any misconfiguration causing the behaviour of zenoss. My doubt is on the datasources:

1. On WinService/Process (as i "misplacedly" commented before), the original query is: Win32_Process.CommandLine="${here/pathName}", but it only works correctly if i change it to Win32_Process.CommandLine='${here/pathName}'. Otherwise it returns WBEM_E_INVALID_QUERY, on queries which commandline has spaces as in WHERE CommandLine=""C:\Program Files\Symantec\Backup Exec\RAWS\beremote.exe".

2. On FileSystem/usedBlocks: The query is the content of ${here/snmpindex}, which is always going to return stuff like SELECT FreeSpace FROM 2, which returns an ugly WBEM_E_INVALID_QUERY

    3. Win32DiskDrive/PerfDisk (similar to 2)

4. Win32DiskDrive/usedBlocks (similar to 2)

5. Filesystem/usedBlocks (similar to 2)

My zenpacks stack is:

SQLDatasource 1.96

WMIDatasource 2.83

WMIPerf_Windows 2.5.80

Is there any WMIPerf_Windows 2.5.83 (according to WMIDS version) with other Datasource queries for the points i've noticed before?

And at last, and I believe it's a thing non-wmiperf related, when I add any device on this /Server/Windows/, there's every 5 minutes zenwin running over this device returns an error 0xc002001b. Could I disable zenwin running anything on this device???

Thanks a lot!

RE: WMI Windows Performance

community@zenoss.org — Thu, 14 Jul 2011 17:32:09 GMT

Hi,

1. this problem already solved (i hope) in Git repository (https://github.com/epuzanov/ZenPacks.community.WMIPerf_Windows/commit/31368b9045fdcff332650142c31b415c00b1808c)

2,3,4,5 it seems like you have modeled your server with zenoss.snmp.FileSystemMap modeler plugin. So, delete file systems components and remodel your sever (check zCollectorPlugins zProperty, it must be the same as on /CIM/WMI).

Egor

RE: WMI Windows Performance

community@zenoss.org — Tue, 26 Jul 2011 18:47:04 GMT

Thanks for this Zenpack. Only issue I have is that it slightly conflicts with you Dell Monitor Pack. I can either monitor the performance of the logical disk via WMI or the status of the physical disk via the dell snmp pack. Would I be possible to change the "Hard Disk" entry in the WMI package to logical disk so that people can do both.

RE: WMI Windows Performance

community@zenoss.org — Tue, 09 Aug 2011 11:11:57 GMT

Hi BigEgor,

I'm having a trouble with the os.interfaces from the modeled devices with wmiperf at version 2.5.80, WMIDatasource at 2.83 and SQLDatasource 1.96.

The fact is (and it's double checked): on the hosts to be modeled, the interface has an address with a /22 netmask. When modeling any of them, the route map is perfect, but the ipaddress of the interface shows as w.x.y.z/24. This fact leads zenoss to create a /24 subnet of my w.x.y'.0/22 net, so every host mapped to this /24 subnet won't be able to get shown into the Network Map (which i've been modifying to adapt it to our interests, but i've not been able to draw an edge from network to network).

When modeling the same scenario with snmp, the ipaddress is correctly retrieved as w.x.y'.0/22, no new subnet is created under the network tab, all perfect.

Here is a pastebin link for the debug of the modelling, conviniently censored, with the debug output of processing InterfaceMap and RouteMap: http://pastebin.com/pNdDSu3K

Thanks in advance (and sorry not to answer the MySQL but this has much more priority)

RE: WMI Windows Performance

community@zenoss.org — Tue, 09 Aug 2011 13:17:22 GMT

See: http://dev.zenoss.org/trac/ticket/7702

This is a known bug in Zenoss Core which has not yet been fixed.

RE: WMI Windows Performance

community@zenoss.org — Tue, 09 Aug 2011 19:12:29 GMT

Hi, Ryan,

i'm not sure about it. In this case, WMIPerf, we are collecting the interface configuration (in wmi/InterfaceMap.py) from the WMI view/table Win32_NetworkAdapterConfiguration, and in this modeler we are not getting the IPSubnet field.

i've been checking and there's no trouble retrieving an interface on a /22 netmask correctly via snmp, so i conclude the model suports it.

i'm modifying the InterfaceMap.py, and the ip/netmask that i'm passing is (ip + '/' + str(self.maskToBits(instance['IPSubnet'][0]))), taking in account that i modified the table for the net configuration query as follows:

"Win32_NetworkAdapterConfiguration": (

"Win32_NetworkAdapterConfiguration",

None,

"root/cimv2",

{

'Description':'interfaceName',

'Index':'snmpindex',

'InterfaceIndex':'ifindex',

'IPAddress':'_setIpAddresses',

'IPEnabled':'_ipenabled',

'MTU':'mtu',

'IPSubnet':'IPSubnet',

}

Well, i'm not afraid of playing with python but i'm a newbie, and the string I get is always "IP /22", with the space I draw. I've been applying strip everywhere but this is where i'm stuck :S:S

The thing is:

1. I believe the model supports it, otherwise snmp modeling wouldn't work as it is working for me.

2. We are simply not retrieving the IPSubnet field, which converted via maskToBits will be our netmask.

3. I'm pathetic, i'm not able to concatenate correctly the IP field with the netmask to finish testing this

The script just as I modified is in this pastebin: http://pastebin.com/ywqM5qpw

And thanks a lot for the fast reply!!

RE: WMI Windows Performance

community@zenoss.org — Tue, 09 Aug 2011 19:51:43 GMT

No, luizzmiz, you are wrong!!! (cronicle of a retractation)

Third point is false, I was really passing the ip/net correctly, but not outputting via logger :S.

Via zendmd, i tried playing with the device.os.ipinterface[index].setIpAddresses and then getIpAddresses, and finally saw what u were telling me, sorry!

Well i'm gonna have a look to the setIpAddress definition and also try to understand why is it fu...ing working on snmp modeling :S

Thanks a lot!

RE: WMI Windows Performance

community@zenoss.org — Wed, 10 Aug 2011 00:28:21 GMT

Well, finally, not completely wrong, not completely right.

As i wrote two posts ago, adding the IPSubnet to the WMI query and forming IP/netmask on the setIpAddresses, should do the trick --> YES, it does it, BUT

In my case, the first modelations of the device associated an IpAddress, which is also an object, hanging from a Network object. For that reason, my device IP was scanned as w.x.y'.z/24, and got on a Network w.x.y'.z/24, belonging to a w.x.y.z/22. Playing with zendmd and changing the interface address of the device with setIpAdressess like follows:

i=find('devicename').os.interfaces()[ifacenumber]

i.setAddress('w.x.y.z/22')

i.getAddress()

w.x.y.z/24

but when setting another random address with /22, the time used to do the operation was like half a second, meanwhile the above setaddress was an instant.

i.setAddress('w.x.y.R/22')

i.getAddress()

w.x.y.R/22

commit()

So, i was thinking maybe when i'm specifying w.x.y.z/22, zenoss finds that w.x.y.z already exists so it uses it, nevermind the netmask. Maybe? Let's delete this IP, (via zendmd, using deleteUnusedIps over the unloved /24 network) and then let's model once again the device, having the modifications over InterfaceMap, retreiving and using IPSubnet: CHORDS! It works!

Well, my modification is not perfect at all, but i wanted to give the functional idea to Egor. The problem i have is being a little newbie here... the way i did it, if the interface has several addresses, the netmask applied is the first found (instance['IPSubnet'][0])

I would like to publish this info on the ticket Ryan wrote, but i don't have access... so, if anyone who agrees that this info is useful could attach it to the ticket, or could tell me a way to have access (without beeing customer), i would thank it!

Well... good night!

PS: The InterfaceMap.py stays like the pastebin I wrote before: http://pastebin.com/ywqM5qpw

RE: WMI Windows Performance

community@zenoss.org — Tue, 16 Aug 2011 19:43:28 GMT

Hi luizzmizz,

thanks for pointing that out. Fixed in GIT (2.5.82)

Egor

RE: WMI Windows Performance

community@zenoss.org — Tue, 04 Oct 2011 03:04:21 GMT

Hi Egor,

I just noticed something relevant regarding the NewDeviceMap modeller.

To set a device's serial number, the modeller queries SerialNumber from Win32_SystemEnclosure. This causes an incorrect serial number to be allocated for blade servers, as the serial for the blade enclosure rather than the blade itself is modelled.

I have edited NewDeviceMap.py to query SerialNumber from Win32_BIOS instead. This results in the individual blade's serial number to be allocated. For regular servers, the serial number allocated is the same as is reported by Win32_SystemEnclosure.

So I think Win32_BIOS might be a better source for SerialNumber.

Cheers,

James

RE: WMI Windows Performance

community@zenoss.org — Wed, 11 Jan 2012 18:35:01 GMT

Hi Egor.

First thank you for this effort that gives the Zenoss community, with these zenpacks, the reason for my query is that I can not send emails alerts from those templates, other devices can do it without problems, but since the device classes CIM, no send alerts.

I have zenoss 3.2.1 on a centos 5.7 64 bitss both.

RE: WMI Windows Performance

community@zenoss.org — Mon, 16 Jan 2012 12:21:19 GMT

Hi, I am sorry to say I am not using Zenoss any more. So I can not help

you.

Regards...

RE: WMI Windows Performance

community@zenoss.org — Mon, 16 Jan 2012 14:00:23 GMT

@Diorgenes: I'm not trying to be rude here, but there's no point in posting just to say that you can't help. You're just cluttering the comments.

RE: WMI Windows Performance

community@zenoss.org — Wed, 29 Feb 2012 13:45:13 GMT

Hi Egor, hi all!

First of all thanks for your great zenpacks.

But i have 2 questions about this zenpack.

1. I have 16 windows 2k8 r2 servers. After installing your zenpack everything looks good. but after two hours 5 servers get an error with: "could not get wmi instances". So i searched via google and tried all solutions for the login to fix this problem but it is still there. So does anyone has an idea, how i can fix this problem?

2. Can anyone tell me where i can change the threshold for the disk util?

Thx for your help

Best regards from austria

Khelben

RE: WMI Windows Performance

community@zenoss.org — Wed, 29 Feb 2012 15:10:06 GMT

Hi Khelben.

I have had a similar problem 2 weeks ago and it was because the CPU usage was 100%, such as CPU usage was at the limit, Zenoss to make the query to server could not attend because it was full .

For change the threshold of disk is this link:

http://community.zenoss.org/blogs/zenossblog/2008/04/23/tip-of-the-month-creating-property-driven-thresholds

If only want change the threshold for one device:

Select Device server.
Select File System.
In the horizotal line there a option 'Display' is a options menu, there select Template (default is Graph).
In threshold selec + add trheshold.
In select add --> usedBlocks_usedBlocks
In max value -- > here.totalBlocks * .95 (disk usage 95 %)
Save

RE: WMI Windows Performance

community@zenoss.org — Wed, 29 Feb 2012 15:23:56 GMT

You are witnessing the very reason why we moved to using SNMP instead of WMI to perform Windows monitoring. We only use WMI in a very targetted fashion for monitoring applications such as Exchange, IIS, etc... It was at the point where we had so many "Could not get wmi instance" events that we couldn't possibly spend the time continuously resolving the issue on all servers. In many cases we tried everything we could think of and still couldn't fix it. WMI is just a flakey protocol.

I created a few ZenPacks to enhance the SNMP monitoring capabilities of Zenoss for windows...

Windows SNMP Service Monitor

Windows SNMP Performance Monitor (Advanced)

Windows SNMP Performance Monitor (Simple)

RE: WMI Windows Performance

community@zenoss.org — Thu, 01 Mar 2012 10:15:41 GMT

@Lord_King: hello and thx for the answer but my server didn't had an 100% useage. I get this messeage instantly after the devices was discovered. I also tried to change the logon credentials, but nothing changes. Do you have any ideas for me?

@Ryan: Thx for your answer but my CIO want to wmi version, so your zenpacks informations is very nice but a can't use it. Do you have another idea for me?

Please let me know!

Thx and best regards

Khelben

RE: WMI Windows Performance

community@zenoss.org — Thu, 01 Mar 2012 14:53:13 GMT

Hi Khelben.

Try checking the permissions WMI or SNMP active but only send the traps snmp to the Zenoss server and verify if its work.

These credentials can make query WMI? if not, join the user Administrator to the zenoss.

Other option is restart the zenoss service.

RE: WMI Windows Performance

community@zenoss.org — Thu, 01 Mar 2012 15:06:15 GMT

Unfortunately SNMP is the only way around the horror that is WMI. If your CIO is forcing you to use it then you're going to have to suffer through it. In my experience WMI will work fine on some servers and just refuse to work properly on others. We've had servers where WMI would stop working, we'd fix it, it would stop working again a few days later, we'd fix it, it would stop working again, and so on. In the end it turned out to be way too unreliable and tedious for us to use. I'm amazed that after 10+ years since WMI was first developed it is still as flakey as it is. We've also noticed a pattern that it runs better on Windows Server 2003 than it does on 2008. We've also noticed that it runs better on Windows Server 2008 with SP2 installed than it does on Windows Server 2008 SP1.

RE: WMI Windows Performance

community@zenoss.org — Thu, 17 May 2012 10:26:32 GMT

Hi Egor,

Thanks for a great ZenPack.

We've been using this for several months without any problems, however we have just updated the WMIDataSource to v3.0 (which uses the SQLDataSource), and this ZenPack now produces an error on our Windows Servers:

RouteMap modelling:

File "/usr/local/zenoss/zenoss/ZenPacks/ZenPacks.community.WMIPerf_Windows-2.5.80.egg/ZenPacks/community/WMIPerf_Windows/modeler/plugins/community/wmi/RouteMap.py", line 88, in mapSnmpVal value = map[value-1] TypeError: tuple indices must be integers, not float

This appears to be because the SQLDataSource is returning data as 'float's not 'integer's. My crude fix is to simply update line 88 to "mapSnmpVal value = map[int(value)-1]".

I hope this my help anyone else who is having this issue and perhaps a permenent fix can be incorporated into the next release?

Thanks,

Nick.

RE: WMI Windows Performance

community@zenoss.org — Sat, 16 Jun 2012 13:54:22 GMT

Hi egor,

Thanks for this zenpack. here is a bug report for version 3.0 (ZenPacks.community.WMIDataSource-3.0.egg).

Description:

WMIPerf zenpack crashes on ubuntu Ubuntu 12.04 LTS when using WMIDataSource-3.0.egg

This does not happen when reverting to WMIDataSource-2.11.egg.

other zenpack is use

ZenPacks.community.SQLDataSource-2.2.egg

ZenPacks.community.WMIPerf_Windows-2.5.80.egg

zenoss versions:

------------

Zenoss 3.2.1

OS Linux (i686) 3.2.0 (Linux honey 3.2.0-24-generic-pae #39-Ubuntu SMP Mon May 21 18:54:21 UTC 2012 i686)

Zope Zope 2.12.1

Python Python 2.6.2

Database MySQL 5.0.45 (Ver 5.0.45)

RRD RRDtool 1.3.9

Twisted Twisted 8.1.0

NetSnmp NetSnmp 5.4.1

PyNetSnmp PyNetSnmp 0.29.13

WMI Wmi 1.3.13

crash log while running zenmodeler on a /CMI/WMI machine:

zenoss@honey:/usr/local/zenoss/zenoss/bin$ zenmodeler run -d 192.168.56.101 -v 10

2012-06-16 16:52:26,375 DEBUG zen.ZenModeler: Run in foreground, starting immediately.

2012-06-16 16:52:26,382 DEBUG zen.pysamba: client ntlmv2 auth is now no

2012-06-16 16:52:26,382 DEBUG zen.ZenModeler: Starting PBDaemon initialization

2012-06-16 16:52:26,383 INFO zen.ZenModeler: Connecting to localhost:8789

2012-06-16 16:52:26,383 DEBUG zen.ZenModeler: Logging in as admin

2012-06-16 16:52:26,393 INFO zen.ZenModeler: Connected to ZenHub

2012-06-16 16:52:26,394 DEBUG zen.ZenModeler: Setting up initial services: EventService, ModelerService

2012-06-16 16:52:26,395 DEBUG zen.ZenModeler: Chaining getInitialServices with d2

2012-06-16 16:52:26,399 DEBUG zen.ZenModeler: Loaded service EventService from zenhub

2012-06-16 16:52:26,399 DEBUG zen.ZenModeler: Loaded service ModelerService from zenhub

2012-06-16 16:52:26,400 DEBUG zen.ZenModeler: Queueing event {'severity': 0, 'component': 'zenmodeler', 'agent': 'zenmodeler', 'summary': 'started', 'manager': 'honey', 'device': 'localhost', 'eventClass': '/App/Start', 'monitor': 'localhost'}

2012-06-16 16:52:26,400 DEBUG zen.ZenModeler: Total of 1 queued events

2012-06-16 16:52:26,401 DEBUG zen.ZenModeler: Calling connected.

2012-06-16 16:52:26,401 DEBUG zen.ZenModeler: fetching monitor properties

2012-06-16 16:52:26,414 DEBUG zen.ZenModeler: Getting threshold classes...

2012-06-16 16:52:26,417 DEBUG zen.ZenModeler: Loading classes ['Products.ZenModel.MinMaxThreshold']

2012-06-16 16:52:26,417 DEBUG zen.ZenModeler: Fetching default RRDCreateCommand...

2012-06-16 16:52:26,418 DEBUG zen.ZenModeler: Getting collector thresholds...

2012-06-16 16:52:26,432 DEBUG zen.thresholds: Updating threshold ('high event queue', ('localhost', ''))

2012-06-16 16:52:26,433 DEBUG zen.thresholds: Updating threshold ('zenmodeler cycle time', ('localhost', ''))

2012-06-16 16:52:26,433 DEBUG zen.thresholds: Updating threshold ('zenperfsnmp cycle time', ('localhost', ''))

2012-06-16 16:52:26,433 DEBUG zen.thresholds: Updating threshold ('zenping cycle time', ('localhost', ''))

2012-06-16 16:52:26,434 DEBUG zen.thresholds: Updating threshold ('zenprocess cycle time', ('localhost', ''))

2012-06-16 16:52:26,434 DEBUG zen.ZenModeler: Getting collector plugins for each DeviceClass

2012-06-16 16:52:26,460 DEBUG zen.ZenModeler: Starting collector loop...

2012-06-16 16:52:26,460 INFO zen.ZenModeler: Collecting for device 192.168.56.101

2012-06-16 16:52:26,551 DEBUG zen.ZenModeler: Loaded plugin community.wmi.ProcessorMap

2012-06-16 16:52:26,551 DEBUG zen.ZenModeler: Loaded plugin community.wmi.FileSystemMap

2012-06-16 16:52:26,552 DEBUG zen.ZenModeler: Loaded plugin community.wmi.DiskDriveMap

2012-06-16 16:52:26,552 INFO zen.ZenModeler: No WMI plugins found for 192.168.56.101

2012-06-16 16:52:26,552 DEBUG zen.ZenModeler: Loaded plugin community.wmi.ProcessorMap

2012-06-16 16:52:26,552 DEBUG zen.ZenModeler: Loaded plugin community.wmi.FileSystemMap

2012-06-16 16:52:26,552 DEBUG zen.ZenModeler: Loaded plugin community.wmi.DiskDriveMap

2012-06-16 16:52:26,553 DEBUG zen.ZenModeler: Using community.wmi.ProcessorMap on 192.168.56.101

2012-06-16 16:52:26,553 DEBUG zen.ZenModeler: Using community.wmi.FileSystemMap on 192.168.56.101

2012-06-16 16:52:26,553 DEBUG zen.ZenModeler: Using community.wmi.DiskDriveMap on 192.168.56.101

2012-06-16 16:52:26,554 INFO zen.ZenModeler: Python collection device 192.168.56.101

2012-06-16 16:52:26,554 INFO zen.ZenModeler: plugins: community.wmi.ProcessorMap, community.wmi.FileSystemMap, community.wmi.DiskDriveMap

2012-06-16 16:52:26,554 DEBUG zen.PythonClient: Running collection for plugin community.wmi.ProcessorMap

2012-06-16 16:52:26,554 DEBUG zen.SQLClient: Running collection for plugin community.wmi.ProcessorMap

2012-06-16 16:52:26,555 DEBUG zen.PythonClient: Running collection for plugin community.wmi.FileSystemMap

2012-06-16 16:52:26,555 DEBUG zen.SQLClient: Running collection for plugin community.wmi.FileSystemMap

2012-06-16 16:52:26,555 DEBUG zen.PythonClient: Running collection for plugin community.wmi.DiskDriveMap

2012-06-16 16:52:26,556 DEBUG zen.SQLClient: Running collection for plugin community.wmi.DiskDriveMap

2012-06-16 16:52:26,556 DEBUG zen.ZenModeler: Loaded plugin community.wmi.ProcessorMap

2012-06-16 16:52:26,557 DEBUG zen.ZenModeler: Loaded plugin community.wmi.FileSystemMap

2012-06-16 16:52:26,557 DEBUG zen.ZenModeler: Loaded plugin community.wmi.DiskDriveMap

2012-06-16 16:52:26,557 INFO zen.ZenModeler: No command plugins found for 192.168.56.101

2012-06-16 16:52:26,557 INFO zen.ZenModeler: SNMP monitoring off for 192.168.56.101

2012-06-16 16:52:26,557 DEBUG zen.ZenModeler: Loaded plugin community.wmi.ProcessorMap

2012-06-16 16:52:26,558 DEBUG zen.ZenModeler: Loaded plugin community.wmi.FileSystemMap

2012-06-16 16:52:26,558 DEBUG zen.ZenModeler: Loaded plugin community.wmi.DiskDriveMap

2012-06-16 16:52:26,558 INFO zen.ZenModeler: No portscan plugins found for 192.168.56.101

2012-06-16 16:52:26,558 DEBUG zen.ZenModeler: Running 1 clients

2012-06-16 16:52:26,559 DEBUG zen.ZenModeler: Collection slots filled

2012-06-16 16:52:26,559 DEBUG zen.ZenModeler: Running 1 clients

2012-06-16 16:52:27,560 DEBUG zen.ZenModeler: Running 1 clients

2012-06-16 16:52:28,574 DEBUG zen.ZenModeler: Running 1 clients

2012-06-16 16:52:28,809 DEBUG zen.pysamba: OK: 192.168.56.101 - Connect

2012-06-16 16:52:28,815 DEBUG zen.pysamba: OK: 192.168.56.101 - ExecQuery

2012-06-16 16:52:29,575 DEBUG zen.ZenModeler: Running 1 clients

2012-06-16 16:52:29,864 DEBUG zen.pysamba: OK: 192.168.56.101 - Retrieve result data.

2012-06-16 16:52:29,870 DEBUG zen.pysamba: OK: 192.168.56.101 - Retrieve result data.

Segmentation fault (core dumped)

running zenmodeler with strace

stat64("/etc/localtime", {st_mode=S_IFREG|0644, st_size=2197, ...}) = 0

write(2, "2012-06-16 16:55:55,276 DEBUG ze"..., 642012-06-16 16:55:55,276 DEBUG zen.ZenModeler: Running 1 clients

) = 64

gettimeofday({1339854955, 278450}, NULL) = 0

gettimeofday({1339854955, 278590}, NULL) = 0

epoll_wait(4, 2012-06-16 16:55:55,526 DEBUG zen.pysamba: OK: 192.168.56.101 - Retrieve result data.

2012-06-16 16:55:55,532 DEBUG zen.pysamba: OK: 192.168.56.101 - Retrieve result data.

+++ killed by SIGSEGV (core dumped) +++

Segmentation fault (core dumped)

RE: WMI Windows Performance

community@zenoss.org — Sun, 08 Jul 2012 10:58:10 GMT

Hi and thanks for a great ZenPack.

I have done a fresh install and added one device to monitor, but as soon as i do a modeling of my device the zenperfsql daemon die. I can start the zenperfsql daemon and it keeps running until a do a modeling.

The log file zenperfsql.log gives me nothing, but in the message log i get the following error

Jul 8 12:52:33 xxx kernel: [332446.704883] .python.bin[19401] general protection ip:b5906c29 sp:b5906ac8 error:0

Jul 8 12:52:34 xxx kernel: [332448.296505] .python.bin[19413]: segfault at ffffffa0 ip b6b98a3f sp b57d21f4 error 4 in _ctypes.so[b6b8e000+12000]

Environment:

openSUSE 11.4 (i586)

Zenoss 3.2.1

SQLDataSource-2.2

WMIDataSource-3.0

WMIPerf_Windows-2.5.80

HPProcurve-1.1

Do you know what the problem is?

Thanks

Magnus

RE: WMI Windows Performance

community@zenoss.org — Sun, 08 Jul 2012 11:09:07 GMT

Zenpack: ZenPacks.community.WMIPerf_Windows, ZenPacks.community.WMIDataSour

Server: Ubuntu 12.04

Zenoss: 3.2.1ce, ZenPacks.community.SQLDataSource

Hi Guys I have had a few Server 2008 servers running under the cim/wmi class for a few days now. Zenoss is able to connect to all the machines and pulls down info on the server like cpu type but none of the graphs are populating. I checked under /perf/devices and no rrd files are being created there.

The graphs do populate if I use snmp with the extra client installed on the servers but I would rather use wmi.

Any thoughts?

RE: WMI Windows Performance

community@zenoss.org — Sun, 08 Jul 2012 14:49:22 GMT

You don't need an extra client installed on the servers to monitor with SNMP. Check out my Windows SNMP ZenPacks. I've released packs to monitor CPU, memory and services via the standard windows SNMP agent. WMI is horribly buggy/unstable. We used to monitor servers with WMI but it was endless problems. We only use WMI when we need to monitor a specific application like Exchange.

RE: WMI Windows Performance

community@zenoss.org — Thu, 26 Jul 2012 21:22:25 GMT

Hi Egor.

When I see the file interface of a device, in the graph Packets, show 04 ítems inbound, inboud unicast, outbound, outbound unicast.

What's the difference between inbound and inbound unicast?

RE: WMI Windows Performance

community@zenoss.org — Thu, 26 Jul 2012 21:26:42 GMT

A unicast packet is a regular IP packet that has a single destination IP address. This is opposed to a multicast or broadcast packet that is destined for more than one device.

RE: WMI Windows Performance

community@zenoss.org — Thu, 26 Jul 2012 21:44:34 GMT

Thanks Ryan.

Now, i want know in the graph packet/sec have relation with Throughput,

What is the size of the packet that show me Zenoss?

RE: WMI Windows Performance

community@zenoss.org — Thu, 26 Jul 2012 21:46:42 GMT

Packet sizes vary, hence the separate graph.

RE: WMI Windows Performance

community@zenoss.org — Thu, 26 Jul 2012 22:07:42 GMT

Thanks. I understood.

My last question: In the option detail of the File system see:

Total Files:

Available Files:

Capacity Files:

Maximum Name Length:

What is this data?

RE: WMI Windows Performance

community@zenoss.org — Thu, 26 Jul 2012 22:09:20 GMT

Those are just properties available for any filesystem components within Zenoss by default. They aren't populated via WMI. I believe those are only used in special cases.

RE: WMI Windows Performance

community@zenoss.org — Thu, 26 Jul 2012 22:12:07 GMT

Thanks for all.

RE: WMI Windows Performance

community@zenoss.org — Tue, 20 Nov 2012 10:39:54 GMT

I have a problem with Windows Services monitoring. Performance graphs are empty. The Data Source query value Win32_Process.CommandLine='${here/pathName}' cause incorrect query because in select statement the WHERE clause get value like eg. CommandLine='C:\Program Files\xxxxx.exe'.

WMI expect value like CommandLine='C:\\Program Files\\xxxxx.exe' for proper query execution. This return from the other hand empty result (tried CommandLine=\"C:\\\\Program Files\\\\xxxxx.exe\" using wmic) perhaps because C:\\Program Files\\xxxxx.exe != C:\Program Files\xxxxx.exe' during the WHERE comparison. Anyone can help or got solution?