- observium - lists.observium.org

Alert for port-channel summary?
by Laszlo Nemeth 20 Jul '16

20 Jul '16

Hello, We have Cisco Nexus switches with VPC configuration. On the switches are configured a portchannel (in VPC with 2x10G ethernet) to the same 3. device. The problem is: I would like to create an alert if the traffic to the 3. device more than 8G I have to get an alert. Now because the 2 nexus just show only the own port-channel traffic, if one portchannel traffic on switch#1 is 6G and on the switch#2 is 5G the 8G alert doesn’t work. But because of the redundancy in this case I have to get an alert because the aggregated traffic is 11G (and if one of the 10G goes down this traffic is overloaded). The alert is: Test: ifOutOctets_perc gt 80 Device Match / Entity Match Hostname match *switch* ifAlias match *to 3. device* So it cover the switch#1 and switch#2 and the port-channel interfaces with description “2x10G to 3. device” but this alert check the port-channels like a line not an aggregated traffic. Is there any option to add two device’s interfaces and create an alert for this summary? Or can I create an alert for a port group aggregated traffic? This would be great in this case too. Thanks BR Laszlo

2 1

Frequent drop outs from WebUI
by Roland Schwingel 20 Jul '16

20 Jul '16

Hi... I am running Observium for quite a while now. But there is one thing that bugs me. I first thought this is a bug and would be fixed at some time but I am updating Observium frequently from svn yet with no change. I am frequently dropped out of Observium with an authentication error. I log in (with LDAP authbackend) just fine. But sometimes after 30 minutes, sometimes even after a few seconds I am dropped out of Observium. When I click the "Remember me" (Cookie) button this lasts a little longer. Sometimes I am immediately kicked directly after logging in. This is also logged in Observiums "Authentication log" as "Authentication Failure" in the very same second as the "Logged in" event occured. Is anything known about this? Or is it just a nifty PHP config variable about session handling I forgot to set? Thanks for your help, Roland

1 0

JunOS autodiscovery where SNMP is in a routing-instance
by Jason Lixfeld 20 Jul '16

20 Jul '16

Hi, I’ve been reading up on how to interact with a JunOS device where the SNMP agent is contained in a VRF (Routing Instance). In order to poll it properly, I’m required to use '<Routing_Instance_Name>@commmunity' instead of just ‘community’. I think I’ve covered this off in the config.php by adding the equivalent to the proper array: $config['snmp']['community'] = array(“comm",”MANAGEMENT@comm”); While reading up on the autodiscovery process itself in the documentation, it says: "Devices must return valid, resolvable sysName” When I poll it manually, sysName returns <Routing_Instance_Name>@hostname: SNMPv2-MIB::sysName.0 = STRING: MANAGEMENT(a)lab01.juniper Does that mean MANAGEMENT(a)lab01.juniper needs to resolve, or will Observium truncate MANAGEMENT@ when trying to resolve the rest of what is in sysName? Thanks!

1 0

Turn off the map?
by Sean Drury 20 Jul '16

20 Jul '16

Hi All, I've looked through http://www.observium.org/docs/config_options/ a few times now and can see options for maps - I don't need to use the map feature so how do I turn it off? That is, remove it from the front page? Regards, Sean. This email has been scanned by the ES1000.

2 2

Re: [Observium] New Community Edition Date
by ron＠rjr-services.com 19 Jul '16

19 Jul '16

New community editions are released not less than 60 days after the last person to ask about a community release. Your inquiry has reset the 60 day counter. ;-) (just kidding!!!!!) Sent from my Verizon 4G LTE smartphone -------- Original message -------- From: Andrew White <andywhite(a)gmail.com> Date: 7/19/16 2:41 PM (GMT-06:00) To: observium(a)observium.org Subject: [Observium] New Community Edition Date -- This message has been scanned by E.F.A. Project and is believed to be clean.

1 0

New Community Edition Date
by Andrew White 19 Jul '16

19 Jul '16

Hi, Is there a date for the next community edition ? All I can see is that there should be one every 6 months, and last was 26th Jan. Thanks ! Loving Observium ! Andy

2 1

Google map does not show
by Nixon, Anthony S. 19 Jul '16

19 Jul '16

Worked during installed. Now map does not show with "Oops! Something went wrong. The page didn't load Google Maps correctly. See the JavaScript console for technical details." Is this an api issue? How to fix or update to get working? Using Observium CE 0.16.1.7533 ----------------------------------------- This message, including any attachments, is intended only for the use of the addressee and contains information that is PROPRIETARY and CONFIDENTIAL. It may be used only by the addressee and may not be divulged without the express consent of the sender. If you are not the intended recipient, you are hereby notified that any dissemination of this communication is strictly prohibited. If you have received this communication in error, please erase all copies of the message and its attachments and notify us immediately. Thank you.

3 10

Port description
by MTiVi＠google 19 Jul '16

19 Jul '16

Hello, all! If I use port description like "Peering: something [40Mbit]" how can I use speed in Observium? Does interface speed (from IfAlias) saved to DB? I expect that I can see traffic graph and select scale to this speed. http://observium.org/docs/port_descr_parsing/

3 4

Making an alert for mapped interfaces
by Heston Kan 18 Jul '16

18 Jul '16

Hey All, I want to create an alert if any interface between two managed/discovered devices in Observium. This includes switchers and routers, servers...etc. I'd also like these to be based on Observium's device Mapping that is available per device. In the checker conditions I'd put hostname matches * but under the checker device and entity association. There's no entity for mapped devices whether it be via CDP/LLDP or detected from observium itself. Perhaps I'm missing the obvious. Thanks! Heston

2 2

Subscription Version not showing "Storage" correctly?
by Dano, Leo 18 Jul '16

18 Jul '16

Hello all, I have two instances of Observium running with two different hostnames. One is Community and one is the Subscription version. My Community version on is showing storage correctly. My Subscription version is not. All the partitions are showing as 100% free and green when I know / can't be 0bytes as well as all other partitions. Is this a bug? Thanks, ----------------------- Leo Dano

2 3

How does SYSLOG Alerting work?
by Сережка Хомяков 18 Jul '16

18 Jul '16

Hello! Thanks for higly awaited feature with SYSLOG alerting! Just playing with it, I was unable to make it working (of course, with knowledge of PCRE). The example of my test Alert was based on *SYS-5-CONFIG_I:Configured* message from IOS router. Can you advice on how the regexp should look like and how does SYSLOG Alerter work - on a poll or triggered on a log receipt? -- Best, Sergey

2 2

alert 'time of recovery' reports recovery of previous recovery...
by Henrik Cednert (Filmlance) 18 Jul '16

18 Jul '16

Hello This was touched upon in another discussion but as a sidetrack no solution was found. Pretty much all my alerts 'time or recovery' reports the time of this entities PREVIOUS recovery. So the time reported are always pretty bonkers. I can see that with the help of timestamp on the emails , slack notifications and the timestamp of previous recovery. And for the very first recovery of a device/sensor the duration is list as "unknown". Any idea of what's going on and what I can kick on to make it work properly? =) Cheers and thanks -- Henrik Cednert cto | compositor Filmlance International | www.filmlance.se mobile [ + 46 (0)704 71 89 54 ] skype [ cednert ]

2 2

Alert on device status = degraded
by Daniel Horth 18 Jul '16

18 Jul '16

Howdy all I have a steelhead that is showing “Degraded” status in Observium due to a power supply issue. In Observium I can see the status “degraded” at this location: https://observium/device/device=12/tab=health/metric=status/id=123/ I can also see that the status changed 8 days ago. DB sample:  <database name="observium">  <table name="status"> <column name="status_id">123</column> <column name="status_deleted">0</column> <column name="device_id">12</column> <column name="poller_type">snmp</column> <column name="status_oid">.1.3.6.1.4.1.17163.1.1.2.7.0</column> <column name="status_mib">NULL</column> <column name="status_object">NULL</column> <column name="status_index">systemHealth.0</column> <column name="status_type">steelhead-system-state</column> <column name="status_descr">System Health</column> <column name="entPhysicalIndex">NULL</column> <column name="entPhysicalClass">chassis</column> <column name="entPhysicalIndex_measured">NULL</column> <column name="measured_class">NULL</column> <column name="measured_entity">NULL</column> <column name="status_ignore">0</column> <column name="status_disable">0</column> </table>  <table name="status-state"> <column name="status_id">123</column> <column name="status_value">30000</column> <column name="status_polled">1468819524</column> <column name="status_last_change">1468104632</column> <column name="status_event">alert</column> <column name="status_name">degraded</column> </table> </database> I am trying to set up an alert in Observium such that when a device status / System Health changes I am notified. Is there a way to access values from status-state table in setting up alerts? Thanks

2 2

pps metic
by Matthias Cramer 16 Jul '16

16 Jul '16

Hi There Is there a metric for normal packets per second and not only unicast packets (ifInUcastPkts_rate) and also a metric for non unicast packets per second. According to the doc at http://www.observium.org/docs/entities/#metrics_1 this is not existent... I would like to make an alert on high pps and if possible on high non unicast pps. Any hint is appreciated. Regards Matthias -- Matthias Cramer / mc322-ripe Senior Network & Security Engineer iway AG Phone +41 43 500 1111 Badenerstrasse 569 Fax +41 44 271 3535 CH-8048 Zurich http://www.iway.ch/ GnuPG 1024D/2D208250 = DBC6 65B6 7083 1029 781E 3959 B62F DF1C 2D20 8250

2 3

Tripp-lite UPS Alerts
by Johnston, Tom 15 Jul '16

15 Jul '16

I’m trying to set up alerting for our tripplite when it goes on battery. I cannot get it to work properly for some reason. Entity Type: Status Check Conditions: upsOutputSource.0 ne normal Device Match: os eq poweralert Entity Match: * Observium Version: 0.16.7.7983 (professional) -A debug from observium yields a value of: upsOutputSource.0 = normal (the normal changes to state of “Battery” when on battery) -Even if I change my check conditions to eq normal to test the alert it does not trigger an alert. -I also tried check conditions of * and that did not help. Also we lost power and the event log showed the following: -Also tried just upsOutputSource (without the .0) Source of Output Power<http://epsobservium.isd273.org/device/device=337/tab=health/metric=status/i…> Status Alert: hs-6509-ups.isd273.org Source of Output Power entered ALERT state: battery (previous: normal) Any help would be appreciated -Tom

2 1

Setting up LDAP Auth
by Dano, Leo 15 Jul '16

15 Jul '16

Hello all, Anyone care to share your config settings? I am having a hard time setting this up. Please remove your domain info in the reply. /opt/observium/config.php Thanks, ----------------------- Leo Dano

8 20

Toner History Graphs
by Richard Franks 15 Jul '16

15 Jul '16

I have a bit of an odd one. I have two printers in my observium install. One has been in for ages (an HP laser jet), the other I added a couple of weeks ago (A sharp MFD photocopier thingy). By the looks of things, the HP has been collecting data correctly for ages, and graphing it nicely. I can see the pretty line that shows that the toner has reduced over time. Oddly, it stopped collecting on the 7th July, and hasn’t graphed anything since then. The Sharp, which has been in less time (since Thursday 7th), has never graphed its toner properly. They are both showing the correct level of toner (well, the Sharp is showing two of them at -3, but I’ve checked the snmp output, and it’s returning -3, so that’s nothing to do with observium). Even more weirdly, the toner level on the weekly graph is showing the correct number. The rest show -nan% The Sharp: The HP: I don’t know much about how the graphing side of things works, so I don’t have much knowledge with which to do any debugging, but happy to send more debug info if required! Thanks, Richard

1 0

Exim Volume Count
by Tiago Reis 15 Jul '16

15 Jul '16

Hi, There is anyone here that already create on the exim service for Observium something that reads the volume of sends an not the mail queue? Kind Regards, Tiago Reis

1 0

Incorrect free physical memory for Unix hosts
by Oliver Baker 15 Jul '16

15 Jul '16

Hi, Loving Observium in general but the only issue I'm having is that for all our Unix-like hosts the free memory is reported too literally to be useful. The cached memory pool is included in the "used memory" count, causing almost all our hosts to appear as having less than 5% free physical memory. Is there a way to get free memory not including the cached memory pool? Currently the alert checker I'm using is flagging all our hosts as "out of memory" Many thanks in advance Oliver B

3 8

Hello world
by Daniel Horth 14 Jul '16

14 Jul '16

Hi all I have some questions to ask however wanted to search archives first. Checking the archives page: http://postman.memetic.org/pipermail/observium/ seems to be cactus at the moment. Is this a known issue. Are there archives available anywhere else? Thanks Dan -- Daniel Horth AU +61 458 248 588 USA + 1 505 552 2074 Denmark + 45 50 22 89 55

3 4

Request for HWg-PWR MIB Support
by Gumilar Siegfried 12 Jul '16

12 Jul '16

Hello, could you please add support for the HWg-PWR Devices? http://www.hw-group.com/products/HWg-PWR/hwg-pwr25_M-bus_gateway_en.html You are already supporting another device from this manufacturer: http://jira.observium.org/browse/OBSERVIUM-512 :) Enclosed I am sending you the MIB File, snmpwalk output and the OID Table directly exported from the device GUI. Thanks in advance! Siegfried

1 0

Request for Nutanix MIB support
by Fredrik Winäs 12 Jul '16

12 Jul '16

Hi It would be awesome if there could be support for Nutanix on Observium. And their Hypervisor Acropolis (KVM based) https://portal.nutanix.com/#/page/docs/details?targetId=Web_Console_Guide-A… When I add my Nutanix CVM's (the controller VM that handles everything)it shows as a generic Linux machine. What can I do to help? Med Vänlig Hälsning Fredrik Winäs tel +46 31 762 02 43 fredrik.winas(a)cgit.se<mailto:fredrik.winas@cgit.se> www.cgit.se<http://www.cgit.se/> CGit AB Taljegårdsgatan 11C SE-431 53 Mölndal

2 3

Adding devices by IP address instead of hostname
by Darcy Marshall 09 Jul '16

09 Jul '16

Is there a way of adding devices to Observium by IP address (without having a FQDN with DNS for it)? If not, could I do it by editing the values in the database? — Spydar007

2 1

Any Email alerting on the CE edition
by Paul Marchant 08 Jul '16

08 Jul '16

Hi all , is there any email alerting on the CE edition ? If anyone has got to work could you please forward me some settings. Looked around the net and all I can see is for the Pro edition. Any help appreciated. Paul Marchant Senior IT Technician ________________________________ E: PMarchant(a)cmatrust.net W: stamfordwellandacademy.org<http://www.stamfordwellandacademy.org/> Reception: +44 (0)1780 761 000 [cid:SWA_tweet.jpg]<http://twitter.com/stamfordwelland> [cid:cmat-schools.gif] The content of this e-mail (including any attachments) is strictly confidential. If you are not, or believe you may not be, the intended recipient, please advise the sender immediately by return e-mail, delete this e-mail and destroy any copies. Cambridge Meridian Academies Trust a Company limited by guarantee, registered in England & Wales. No. 7552498. Registered Office: Cambridge Meridian Academies Trust (CMAT), Gibraltar Lane, Swavesey, Cambridgeshire CB24 4RS www.cmatrust.co.uk<http://www.cmatrust.co.uk/>

3 2

Arch -> extension (since 16.04 upgrade)
by Robert Williams 08 Jul '16

08 Jul '16

Hi, Just completed a release upgrade on Observium itself (14.04 to 16.04) and now it is reporting its own architecture as 'extension'. Wouldn't be an issue really, but every single time it polls it updates the event log like this: I've attached a debug output of the poller as this may help to track it down. Interestingly, a lot of our other servers run 16.04 now but it only reports this for 'itself' - not the others. Anyone have any ideas? Cheers! Robert Williams Custodian Data Centre Email: Robert(a)CustodianDC.com http://www.CustodianDC.com

5 12