I've sorted out the issue with the unmonitored servers (had an improper 
monitor assigned). I am now returning to trying to identify the source 
of the segmentation faults.

I have grouped my windows servers into a handful of classes, based on 
authentication method and whether there is a network firewall in-between.

DOMAIN1 - Local Network Domain
WORKGROUP1 - Local workgroup
DMZ-FW - Local DMZ workgroup behind firewall
DOMAIN2-FW - Local network, secondary domain behind firewall
DOMAIN3-WAN - Remote Domain accessed across MPLS

The user accounts are configured as members of the Domain Admins or 
local Administrators groups as appropriate. Manually running wmic from 
the command line for all servers completes successfully.

Here are the results of enabling WMI monitoring for each class.

CLASS        USERNAME            ZENWIN RESULT
DOMAIN1      DOMAIN1\zusername   success
WORKGROUP1   .\zusername         segfault
DMZ-FW       .\zusername         segfault
DOMAIN2-FW   DOMAIN2\zusername   segfault
DOMAIN3-WAN  DOMAIN3\zusername   success

The segfault appears to be related to either configuring a local user 
account and/or when traversing a firewall.

jamesroman wrote:
> OK. So I have two of my servers functioning properly. I divided my servers 
> up, based on whether I had to go through a firewall or not. It looks like the 
> root of the issue may have to do with wmi attempts going through a firewall. 
> My immediate concern is a new problem.
>
> I currently only have the Zenwin process enabled (zWmiMonitorIgnore = false) 
> and zeneventlog enabled. No WMI plugins. Working with just the servers in my 
> local AD domain, I've got zenwin and zeneventlog monitoring 2 of the 4 
> servers in the class properly. The other two just get skipped. 
>
> If I manually run "zenwin run --debug -v 10 -d servername" on the two servers 
> that are not being monitored, it does not seem to even attempt to log in 
> using the WMIClient:connect. If I manually run wmic from the command line 
> using the same username and password I have configured, the select statement 
> works on all 4 servers. The two servers that are not being scanned do take 
> longer to complete the manual wmic select.
>
> The zenwin process is behaving similar to if I had a bad authentication event 
> for the two skipped servers, only there are no " DEBUG:zen.zenwin:Ignoring 
> devices []". 
>
> Anyone have an idea how to get the remaining servers to be scanned by zenwin? 
>  This looks like the behavior some are experiencing in this thread: 
> http://forums.zenoss.com/viewtopic.php?t=8058
>
> Zenos 2.3.3 on Cent 5.3
> Previously (2.2.4) all my windows devices were scanned properly.
>
>
>
>
> -------------------- m2f --------------------
>
> Read this topic online here:
> http://forums.zenoss.com/viewtopic.php?p=34684#34684
>
> -------------------- m2f --------------------
>
>
>
> _______________________________________________
> zenoss-users mailing list
> [email protected]
> http://lists.zenoss.org/mailman/listinfo/zenoss-users
>   

_______________________________________________
zenoss-users mailing list
[email protected]
http://lists.zenoss.org/mailman/listinfo/zenoss-users

Reply via email to