cluther wrote:
> There are a couple of ways you could tackle this problem. The quick hack fix 
> would be to open up $ZENHOME/Products/ZenRRD/zencommand and find the "def 
> error" method. You'll see that it is this method that is sending the event 
> with the sensitive information in them.
> 
> 
> Code:
> 
> msg = "Command timed out on device %s: %s" % (dc.device, cmd.command)
> 
> 
> 
> 
> You could change this so that it didn't include the command that timed out. 
> Of course it will be very difficult to know what is timing out at this point.
> 
> 
> Code:
> 
> msg = "Command timed on on device %s." % (dc.device,)
> 
> 
> 
> 
> The other way to tackle this problem would be to leverage the "-A" parameter 
> to the wmic utility that allows passing credentials via a file instead of the 
> command line. This has the added bonus that someone who can run ps on your 
> system couldn't see the passwords either. Of course this would require 
> modifying the community WMI ZenPack to use -A and somehow setting up the 
> authentication files ahead of time.


Well, mucking with the ZenPack code is a bit more than I want to tackle this 
week.  But I wonder if it would be doable to scan the string in cmd.command to 
look for the presence of the password string and do a substitution that masks 
it out before passing it to the message?  I can visualize doing that in Perl, 
but my Python skills are mui poquito.




-------------------- m2f --------------------

Read this topic online here:
http://forums.zenoss.com/viewtopic.php?p=26296#26296

-------------------- m2f --------------------



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

Reply via email to