
Hi – Update, found that bit now, for future users it was: [cid:image002.png@01D0655A.F9E01A60] So check conditions were status_event notmatch ok thus providing a hit on anything that is bad, or semi-bad… Cheers for your help Tom!
Robert Williams Custodian Data Centre Email: Robert@CustodianDC.com http://www.CustodianDC.com From: observium [mailto:observium-bounces@observium.org] On Behalf Of Robert Williams Sent: 23 March 2015 10:55 To: Observium Network Observation System Subject: Re: [Observium] Critical alert not matching alerts checker
Hi Tom,
Ah ok thanks, found it! Didn’t realise it was something you selected when you first create a checker, rather than something which is filtered out by the match criterial.
Right, one final question remains now then - what do I use to ‘match’ a failed status now?
Currently I have: “sensor_event match down”
Obviously that won’t work if it’s not a ‘sensor’ any more - so do I use something like “status_event match critical” or just “status match critical” ?
Cheers!
Robert Williams Custodian Data Centre Email: Robert@CustodianDC.commailto:Robert@CustodianDC.com http://www.CustodianDC.com From: observium [mailto:observium-bounces@observium.org] On Behalf Of Tom Laermans Sent: 23 March 2015 09:21 To: Observium Network Observation System Subject: Re: [Observium] Critical alert not matching alerts checker
Hi Robert,
"Status" is not a sensor (used to be entity "senso"r of class "state") but a separate entity type ("status") now.
Tom
On 03/23/2015 09:49 AM, Robert Williams wrote:
Hi Tom,
Thanks for that, however I thought that with an (Entity Match = *) that it would match all types of sensor regardless?
[cid:image003.png@01D0655A.F9E01A60]
Is there an updated bit of documentation I'm missing? Or maybe someone can give me the exact syntax it needs to match this if it has some specific requirements?
Cheers!
Robert Williams Custodian Data Centre Email: Robert@CustodianDC.commailto:Robert@CustodianDC.com http://www.CustodianDC.com
-----Original Message-----
From: observium [mailto:observium-bounces@observium.org] On Behalf Of
Tom Laermans
Sent: 22 March 2015 20:04
To: Observium Network Observation System
Subject: Re: [Observium] Critical alert not matching alerts checker
Quick reply from my cellphone, you need an alert checker for entity type
status instead of sensor since a few weeks...
Tom
On Mar 22, 2015 7:55 PM, Robert Williams <Robert@CustodianDC.commailto:Robert@CustodianDC.com> wrote:
Hi All,
I have a device a Cisco 2960G with a failed internal PSU that has
switched to it’s backup RPS supply. Observium is correctly identifying
this as ‘critical’ under the Status Indicators for the device in
question (image below):
There is a “Hardware Fault” event I have created which /should/ have
matched this, I believe:
(I’ve added the match for value = critical as a test, it wasn’t there
originally)
This alert checker correctly matches a who load of devices and status
checkers on many devices (1691 in total, all green):
However, for this particular type of sensor, it does not see any
sensors which match the “Hardware Fault” checker. As you can see, I have
already removed all the “Device Match” and “Entity Match” criteria from
the checker itself to rule that out. So essentially this checker is
matching every type of sensor that there is (except, obviously, the one
which I want it to).
When I click the link on the graph for “Sw1, PS1 Critical, RPS Normal”
sensor I get this URL http://observium-http://observium-server/device/device=343/tab=health/metric=status/
server/device/device=343/tab=health/metric=status/http://observium-server/device/device=343/tab=health/metric=status/ which may be of use
in helping diagnose this.
After doing some more checking, I’ve found that it is not matching any
other kind of sensor with a similar class, for example it does not match
the WS-CAC-4000W-IN power supplies in any of the Cat6k chassis either.
In fact, I think it’s actually failing to match sensors which have
‘normal’ and ‘critical’ type statuses, instead of ‘true’ and ‘false’
So the following sensors highlighted in yellow are also being missed
from matching in this generic alerting config:
So I believe it’s a wider issue and not just related to this one
device.
Can someone confirm what I’m doing wrong and/or maybe check to see if
your alerting configurations can match sensors which have an OK state of
‘normal’ instead of ‘true’?
Cheers!
Robert Williams
Custodian Data Centre
Email: Robert@CustodianDC.commailto:Robert@CustodianDC.com
observium mailing list
observium@observium.orgmailto:observium@observium.org
http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
_______________________________________________
observium mailing list
observium@observium.orgmailto:observium@observium.org
http://postman.memetic.org/cgi-bin/mailman/listinfo/observium