headers
WaltB | 1 Nov 2010 10:43

Can't un-monitor an interface


Hi all,
 
I'm monitoring in a very HP-centric environment. Loads of Proliant BL460 blades and EVA's, and on theese
blades I have Expansion-cards that is not being used.
They make a lot of unnessesary noice as they are in status "other" so I would like to set these interfaces to
not-monitores but I can't, The option "Monitoring..." is grayed out.
 
I'm running the 3.0.3 core stack install, and latest versions of the HPmon, Proliant MIBS, the EVAmon and
latest versions of the prereqs to theese ZenPacks.
 
Does any of you have any idears of where I could look for at hint as to why the option is grayed out, or even
better a solution ;-)
 
WaltB

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

To post a reply to the thread message, either reply to this email (recommended) or visit the message page: http://community.zenoss.org/message/54253#54253

--end--
Permalink | Reply | 0 comments |
headers
bluefox | 1 Nov 2010 15:10

Two questions on monitoring device state and performance


Hi,
 
I'm a new Zenoss user, and I have to say I like the product, after having tried Cacti and Zabbix (great
performance graphs, but lousy event handling, especially standard SNMP traps).
 
I have two important questions that I can't find the answer to, and some smaller ones.
 
1) I am running Zenoss from a completely different network, to ensure Zenoss tells me if my main network is
completely out. One of the metric that is critical to me is the response time (ping response time to keep it
simple) from server MAIN_1 to some_IP_A, some_IP_B and some_IP_C.  Is there an easy way to add this in a
monitoring template? Or do I need to run a custom script and have custom OIDs?
 
2) How can I add some sort of heartbeat to an application? I have an important application running on MAIN_1,
and I obviously want to know if MAIN_1 is down (ping heartbeat), but I would also like to have Zenoss run a
custom script, send some custom command to the application on that server and wait for a response (STATUS
OK for example).  How do I do this?
 
 
Less important: I added a pfSense device to my list, and it's classified under "Network".  I want to change
it and put it under "Network/Routers".  Can I do that, or is it hardcoded into Zenoss?

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

To post a reply to the thread message, either reply to this email (recommended) or visit the message page: http://community.zenoss.org/message/54234#54234

--end--
Permalink | Reply | 17 comments |
headers
tgdaero9 | 1 Nov 2010 15:12

ZFS Filesystem Monitoring: Generic attempt


ZFS cannot be properly monitored & graphed by Zenoss due to limitations of NETSNMP.
 
Now my attempt is to exclude known ZFS via 'zFileSystemMapIgnoreNames' unfortunately a manual task to
avoid false alerts and wrong FS graphs.
Still we need to monitor the ZFS apart from the zpool status. Therefore i may create a script with key/value
pairs to gather correct data using for example 'df -F zfs'.
 
But this results in individual keys dependent from the amount of ZFS on a server and also the naming of these
keys is not determined.
 
Somehow i need a way to feed this dynamically into Zenoss while modelling a device.
It is preferred to show this information also in the 'File System' section of the 'OS' tab.
 
Can someone give me hint how to feed this dynamically into Zenoss and make it dependent from modeling?
 
Thanks for any answer

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

To post a reply to the thread message, either reply to this email (recommended) or visit the message page: http://community.zenoss.org/message/54235#54235

--end--
Permalink | Reply | 0 comments |
headers
jmp242 | 1 Nov 2010 15:24

Re: Two questions on monitoring device state and performance


You may want to look at the fping zenpack. There is a FAQ entry about setting up custom heartbeats in Zenoss:
http://community.zenoss.org/docs/DOC-2445#HowdoIuseHeartbeatstylemonitoring
 
Granted, this was written for an older version, but may still apply.
 
For moving the device class, you drag it to the new class (same as the FAQ for locations):
http://community.zenoss.org/docs/DOC-4724#In_Zenoss_3x_how_do_I_assign_a_device_to_a_location
 
 
 
--
James Pulver
Information Technology Area Supervisor
LEPP Computer Group
Cornell University

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

To post a reply to the thread message, either reply to this email (recommended) or visit the message page: http://community.zenoss.org/message/54254#54254

--end--
Permalink | Reply | 0 comments |
headers
bluefox | 1 Nov 2010 15:54

Re: Two questions on monitoring device state and performance


Thanks James, I don't know why dragging didn't cross my mind before. Device Class is just a way to organise
things right?  I mean, if I put weird Device Class, will it have negative impact on my monitoring?
 
As for the heartbeat, I saw the page, and maybe I'm just too new to Zenoss, but I don't understand how sending a
trap will alert me to something being down.  The application in question doesn't send a trap, I want to
send some telnet command (or something else) regularly and see if it responds, and I want Zenoss to create
an event if it doesn't
 
The trap isn't there to start with. I wish it was ;-)

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

To post a reply to the thread message, either reply to this email (recommended) or visit the message page: http://community.zenoss.org/message/54266#54266

--end--
Permalink | Reply | 0 comments |
headers
jmp242 | 1 Nov 2010 16:16

Re: Two questions on monitoring device state and performance


Device Class is sort of a container, but you use them to apply monitoring templates. Sort of like OUs and GPOs
in AD if that makes more sense. Fully general containers (that don't do anything to the devices, but that
you would use for alerting rules etc) are systems and groups...
 
--
James Pulver
Information Technology Area Supervisor
LEPP Computer Group
Cornell University

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

To post a reply to the thread message, either reply to this email (recommended) or visit the message page: http://community.zenoss.org/message/54256#54256

--end--
Permalink | Reply | 0 comments |
headers
themactech | 1 Nov 2010 16:56

Re: Two questions on monitoring device state and performance


First thing, on what platform is your application running, windows, linux, OS X?  This will give a better
idea of what tools we can use to query for the status or your application.
 
Manuel

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

To post a reply to the thread message, either reply to this email (recommended) or visit the message page: http://community.zenoss.org/message/54257#54257

--end--
Permalink | Reply | 0 comments |

Gmane