- observium - lists.observium.org

Distro Script not working
by Gary Tang 30 Oct '14

30 Oct '14

I have added the distro script as per http://www.observium.org/wiki/Linux_Distribution_Detection Running the distro script produces the right output, but observium does not seem to be detecting the distro correctly (still showing as Linux ***-generic) Snmpd has been restarted, and the OS being detected is Ubuntu 14.04. Any idea what could be going wrong? Thanks!

3 2

Re: [Observium] Wireless signal graphs
by Ron Marosko, Jr. 29 Oct '14

29 Oct '14

I quasi-hacked my community edition release to provide basic support for canopy, but it didn't give any RF details, just basic network utilization. I see where there is a patch to provide better support in the subscription version, which I am hoping will show up in the next community release. If the data looks good for canopy on that, we'll be converting to the subscription version very shortly thereafter... I just have to show positive results to the person who writes the checks, so to speak, and get their go ahead first. ...Ron Sent from my Verizon Wireless 4G LTE DROID Jeremy Lukan <jeremy(a)GLMobile.com> wrote: >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

1 0

Wireless signal graphs
by Travis Kreikemeier 29 Oct '14

29 Oct '14

Any plans to add wireless signal metric graphs to Observium? Such as signal strength and SNR. Example device being Cambium Canopy CPE's. Is there a place to put feature bounties and payments for completion bounties? Thanks TRAVIS KREIKEMEIER

8 7

Graph problems after update
by Jerome SCHEVINGT 29 Oct '14

29 Oct '14

Hi after a update from 0.13.11.48.05 to 0.14.9.5778, we have lost all graphics. Now we have "Draw error" anyone know this problems ? Thanks jerome

3 4

ADSL Sync Speed change alert
by Tom Halapacz 29 Oct '14

29 Oct '14

Hi I am trying to set the Alert checker to notify me on any sync speed changes on the adsl port. Currently I can see all the changes on the Event log as below: 1d 19h 58m adsl Interface changed: [ifSpeed] 5152000 -> 5184000 4d 12h 48m adsl Interface changed: [ifSpeed] 5120000 -> 5152000 6d 4h 23m adsl Interface changed: [ifSpeed] 5344000 -> 5120000 How to set the check conditions to alert on any change in Sync speed such as "ifSpeed ne old_speed" Checker Type:Port Check Conditions: ifSpeed ?? Associations Device match: hostname equals *hostname* Entity match: ifType equals adsl Any help is much appreciated. Regards ---- Tomasz Halapacz

1 0

rancid integration view previous configs
by Justin Miller 29 Oct '14

29 Oct '14

I got it working, but when viewing the config under the device, it has a “Latest” button but nothing else. Is it possible to view previous config? If not is this functionality planned? Justin

4 4

help with an alarm
by Brett Jones 29 Oct '14

29 Oct '14

Hi there, Just got your product up and working with a few back end modifies ect we have a good monitoring system hear. We paid the extra to have the alert functionality as we are an ISP. I have a question though for you. How on earth do you make a packet loss alarm? IE if a CPE hits over 30% then it will alarm? I cant find an attribute from the guide to do this. I have set an alert as device_ping gt 250 type_match * * This works when a cpe hits over 250mms thus showing latency. But how do I do a packet loss? I have tried packet_loss gt 30 but as this is not in the list it dosnt work! Any help would be most appreciated. Kind regards Brett Jones 2nd/3rd line support engineer Tel: 01202 612000 Fax: 01202 612111 W:www.voip-unlimited.net E:brett.jones@voip-unlimited.net<mailto:brett.jones@voip-unlimited.net> E:support@voip-unlimited.net<mailto:support@voip-unlimited.net> [Description: Description: Description: Description: Description: Description: Description: voipunlimited]<http://www.voip-unlimited.net/> [cid:image002.jpg@01CFF36A.05032990] [http://www.brandonu.ca/careerplanning/files/2010/08/linkedin-logo.png]<http://www.linkedin.com/company/2290787?trk=prof-0-ovw-curr_pos> [Description: Description: Description: Description: Description: Description: Description: http://www.wfamedia.co.uk/demo-images/logo_twitter_withbird_1000_allblue.png] <https://twitter.com/VoIPUnlimited> [cid:image005.png@01CFF36A..05032990] Registered Office: Co Carter & Coley, 3 Durrant Road, Bournemouth, Dorset BH2 6NE Registered in England No: 05225497 The information contained in this email and any attachments is confidential and may be subject to copyright or other intellectual property protection. If you are not the intended recipient, you are not authorized to use or disclose this information, and we request that you notify us by reply mail or telephone and delete the original message from your mail system.

1 0

Support for Nimble storage devices
by Robin Cook 29 Oct '14

29 Oct '14

Hello, I appreciate a fair bit of work is involved to add support for new devices but I wanted to test the waters and see if it would be possible to add support for nimble SANs. I have access to the MIBs if required. Many thanks, Robin --------------------------------------------------------------------------------------- This email has been scanned for email related threats and delivered safely by Mimecast. For more information please visit http://www.mimecast.com ---------------------------------------------------------------------------------------

2 1

F5 device O/S identification failures
by Colin Stubbs 29 Oct '14

29 Oct '14

Hi All I've got a support case open with F5 regarding an issue whereby their net-snmp implementation stops reporting F5 specific sysObjectID values. This may be a recent 11.5+ version issue, as I've not run across it before, and it only seems to occur after making SNMP related configuration changes. e.g. on virtual editions, SNMPv2-MIB::sysObjectID.0 = OID: F5-BIGIP-SYSTEM-MIB::bigipVirtualEdition Starts to become at some point, SNMPv2-MIB::sysObjectID.0 = OID: NET-SNMP-TC::linux Observium goes back to treating the system as a Linux box. The F5 specific OID's are still pollable, so the best fix seems to be a quick poll against F5-BIGIP-SYSTEM-MIB::sysGeneralChassisSerialNum.0 to determine if it's an F5 system. Patch is simply, --- observium-trunk/includes/discovery/os/linux.inc.php 2014-10-21 16:36:00.000000000 +1000 +++ equate_observium-trunk/includes/discovery/os/linux.inc.php 2014-10-29 19:32:46.000000000 +1000 @@ -39,6 +39,7 @@ if (strpos(trim(snmp_get($device, "dot11manufacturerProductName.5", "-Osqnv", "IEEE802dot11-MIB", mib_dirs())), "UAP") !== FALSE) { $os = "unifi"; } elseif (strpos(trim(snmp_get($device, "dot11manufacturerProductName.2", "-Osqnv", "IEEE802dot11-MIB", mib_dirs())), "UAP") !== FALSE) { $os = "unifi"; } } + elseif (snmp_get($device, "1.3.6.1.4.1.3375.2.1.3.3.3.0", "-OQv", "F5-BIGIP-SYSTEM-MIB") != '') { $os = "f5"; } } if ($os == "linux") { I hope that helps. -Colin

1 0

Observium not updating
by Jurè Wennekes 28 Oct '14

28 Oct '14

My observiun is stuck in 5902, svn is working and giving me 5918 but te site is stil giving 5902, what am I missing here? :/opt/observium$ sudo svn update At revision 5918. :/opt/observium$ sudo ./discovery.php -h none Observium v0.14.10.5902 Discovery Current Revision : 5902 New Revision : 5918 watch@watch:/opt/observium$ Met vriendelijke groet, Jurè Wennekes Loko IT-specialisten Mast 13 Postbus 146 3890 AC Zeewolde T. 036-521 93 20 W. http://www.loko.nl [cid:image715f40.JPG@5cd3bdf8.418ffbca] Meer informatie op www.ikstapoverop.nl.

2 1

- green point on the map
by Paolo De Michele 27 Oct '14

27 Oct '14

hi everybody, I've added a "normal user" by observium installation I customized it adding a new device (just for test) when I'm doing the login I see the world map and the device why can I not see the green point? (usually indicates machine up&running) how can I solve that? let me know, thanks in advance have a good weekend cheers

1 1

Unix agent bind issue
by Tim Schroepfer 27 Oct '14

27 Oct '14

I got the unix-agent running on one of my main dns server and when i test it telnet to port 36602 it displays results but my observium server isn't displaying any information for that server. Any ideas? Tim S Field Tech/Ast. System Admin E-vergent.com

3 5

Reduce your storage IO with rrdcached
by Markus Klock 27 Oct '14

27 Oct '14

There was some discussion on IRC the other day about rrdcached and how it can be used with Observium to reduce your storage IO a lot when polling a huge amount of interfaces. However, not a lot of documentation exists yet on how to get rrdcached to work with observium. I did some trial and error and found a working method and wrote a quick and dirty guide. You can find it here: http://blog.best-practice.se/2014/10/using-rrdcached-with-observium.html Feedback is highly appreciated! /Markus

5 7

Re: [Observium] Several broken graphs
by Adam Armstrong 26 Oct '14

26 Oct '14

1 0

Several broken graphs
by Robert Williams 26 Oct '14

26 Oct '14

Hi, A number of our hosts have suddenly got broken graphs for items like: * Ping Response * SNMP Response * Poller Duration * Running Processes * Users Logged In * Context Switches * System Interrupts When in debug mode, the graph is showing these errors (depending which graph): command returned 1 (ERROR: String ends after the = sign on 'CDEF:ping_min=' ) command returned 1 (ERROR: String ends after the = sign on 'CDEF:val_min=' ) command returned 1 (ERROR: String ends after the = sign on 'CDEF:value_min=' ) command returned 1 (ERROR: String ends after the = sign on 'CDEF:users_min=' ) command returned 1 (ERROR: String ends after the = sign on 'CDEF:procs_min=' ) An example of a full string for a broken graph is: graph /tmp/0H1FmUUbABhyOl7U.png --alt-autoscale --rigid --start 1414240174 --end 1414326574 --width 1159 --height 300 -c BACK#EEEEEE00 -c SHADEA#EEEEEE00 -c SHADEB#EEEEEE00 -c FONT#000000 -c CANVAS#FFFFFF00 -c GRID#a5a5a5 -c MGRID#FF9999 -c FRAME#5e5e5e -c ARROW#5e5e5e -R normal --font LEGEND:8:'DejaVuSansMono' --font AXIS:7:'DejaVuSansMono' --font-render-mode normal -E COMMENT:' ' COMMENT:' Now' COMMENT:' Avg' COMMENT:' Min' COMMENT:' Max' COMMENT:' Total' COMMENT:'\l' DEF:procs=/opt/observium/rrd/hostname.was.here/hr_processes.rrd:procs:AVERAGE CDEF:procs_min= CDEF:procs_max= AREA:procs#E881B1:"Processes " LINE1.5:procs#744058 GPRINT:procs:LAST:%6.1lf%S GPRINT:procs:AVERAGE:%6.1lf%S GPRINT:procs_min:MIN:%6.1lf%S GPRINT:procs_max:MAX:%6.1lf%S COMMENT:'\l' Anyone else seeing this or has our install spontaneously broken itself? We are at 5910 on Ubuntu 14.04.1 - cheers! Robert Williams Custodian Data Centre Email: Robert(a)CustodianDC.com http://www.CustodianDC.com

1 0

Ports not found on older PowerConnect switches
by Dan Ratzlaff 25 Oct '14

25 Oct '14

Hello. I'm having trouble getting Observium (CE 0.14.4.5229) to recognize ports on my Dell PowerConnect 3048 and 3024 switches. It appears these switches only support SNMP v1. Observium connects and grabs SNMP contact, location, etc. but doesn't see ports. Here's output from add and discovery for my 3024 switch: ----- # ./add_device.php switch-ne Observium v0.14.4.5229 Add Device(s) Try to add switch-ne: Trying v2c community public ... No reply on community public using v2c. Could not reach switch-ne with given SNMP community using v2c. Trying v1 community public ... Now discovering switch-ne (id = 29) switch-ne 29 powerconnect Ports : XXXXXXXXXXXXXXXXXXXXXXXXXX Discovered in 0.498 seconds switch-ne 29 powerconnect IPv4 Addresses : Discovered in 0.092 seconds switch-ne 29 powerconnect IPv6 Addresses : Discovered in 0.189 seconds Added device switch-ne (29). Devices added: 1. ----- # ./discovery.php -h 29 Observium v0.14.4.5229 Discovery switch-ne 29 powerconnect OLD: powerconnect, NEW: powerconnect --- Ports : XXXXXXXXXXXXXXXXXXXXXXXXXX Port Stacks: Processors : Dell-Vendor-MIB hrDevice: Memory : Dell-Vendor-MIB HOST-RESOURCES-MIB IPv4 Addresses : IPv6 Addresses : Sensors: FASTPATH-BOXSERVICES-PRIVATE-MIB + Dell-Vendor-MIB Storage : Inventory: Physical Inventory : hrDevice : Discovery protocols: LLDP-MIB CISCO-CDP-MIB OSPF Neighbours: ARP/NDP Tables : BGP Sessions: No BGP on host VLANs: Q-BRIDGE-MIB UCD Disk IO : Discovered in 2.012 seconds Current Revision : 5229 New Revision : 5909 MySQL: Cell[2/0s] Row[19/0.01s] Rows[33/0.02s] Column[0/0s] Update[1/0s] Insert[5/0.02s] Delete[0/0s] ----- I can do a snmpwalk on these switches from the same machine that Observium is running on and can see all the interface information. Is there something I need to tweak in Observium to get these switches discovered correctly? Thanks! -- Dan

2 5

allow graphs to be viewed by anyone / config.php vs web configuration
by Massimo Sala 25 Oct '14

25 Oct '14

Hi ! Observium CE 0.14.4.5229 Following the docs, in config.php I set : $config['auth_mechanism'] = "mysql"; $config['allow_unauth_graphs'] = 1; $config['allow_unauth_graphs_cidr'] = array( "10.1.0.0/16" ); I create a user "guest" ... but it cannot see any graph. I login to Observium as admin, I edit the user "guest" properties : Device Access None Configured Grant access to new device Interface Access None Configured Grant access to new interface I see different configuration : config.php vs web configuration. Is there any mismatch ? How can I grant a full "read-only" access to a "guest" user ? Best regards, Sala

2 1

observium customizations on PHP code
by Massimo Sala 24 Oct '14

24 Oct '14

Observium CE 0.14.4.5229 PHP 5.4.4-14+deb7u14 MySQL 5.5.40-MariaDB-1~wheezy-log SNMP NET-SNMP version: 5.4.3 RRDtool 1.4.7 I have modified one source file : /opt/observium/includes/alerts.inc.php I restarted web server and php-fastcgi ( no opcode cache ), but Observium doesn't see the update. Is there an internal function caching ? Am I missing something ? Best regards, Sala

3 2

How can I unsubscribe?
by Franck Burgaudeau 24 Oct '14

24 Oct '14

Hi, I would unsubscribe from this miling list but when I enter my mail in this page, I don't receive anything and I continue to receive all mail... Thank you for your reply. Regards *Franck Burgaudeau* f.burgaudeau(a)graphique-alliance.com <http://www.graphique-alliance.com/> *Attention !!* *Graphique Alliance a déménagé.Les téléphones, fax et e-mails sont inchangés.* *Nouvelle Adresse : * *50 rue Laudinat - 33130 Bègles**Graphique**Alliance* Intégration, développement et formation pour les industries graphiques 50 rue Laudinat 33130 Bègles Standard : +33 (0)5 56 67 55 78 Téléphone : +33 (0)1 85 08 64 79 Portable : +33 (0)6 75 61 75 73 Fax : +33 (0)5 57 54 12 77

1 0

Re: [Observium] polling issues with 5897
by Rutger Bevaart 24 Oct '14

24 Oct '14

A strange issue indeed… Based on Mike’s input I ran on the old-style poller for a day, then all is fine. I’m now running on poller-wrapper for another day, with concurrency set to 2, and that seems to be fine as well. I had it at 8, then all goes to h*ll. Will try ‘4’ today ;-) Regards, Rutger > > Netscreen firewall > > [https://observium.exegy.net/graph.php?type=device_processor&device=3&to=141…] > > Arista Switch > [https://observium.exegy.net/graph.php?type=device_processor&device=108&to=1…] > > Cisco Router > [https://observium.exegy.net/graph.php?type=device_uptime&device=70&to=14140…] > > > I can send you a debug poller.php run privately if you'd like. >

1 0

Multiple tun0 interfaces showing up in graphs
by Gary Tang 24 Oct '14

24 Oct '14

For some reason, I am getting multiple tun0 interfaces in graphs, even though I only have one via openVPN. Where should I start looking to debug this? Graph: http://postimg.org/image/b876gqbq5/ Thanks for looking, Gary

1 0

Re: [Observium] Alerts in "Other" state
by Adam Armstrong 24 Oct '14

24 Oct '14

I have no idea. How long is a piece of string? Guys, I'm not here to tickle you until you actually give some information. Give as much information as you have about a problem or you won't get it fixed. Life is too short to chase you all for screen shots and more than 3 word descriptions of your problems. adam. ------ Original Message ------ From: "Ben Steele" <ben(a)bensteele.org> To: "Adam Armstrong" <adama(a)memetic.org>; "Observium Network Observation System" <observium(a)observium.org> Sent: 10/23/2014 4:02:08 PM Subject: Re: [Observium] Alerts in "Other" state >What might cause that? > >The alerts that haven't been checked aren't new and as previously >mentioned even something simple like is device up/down has these >non-checked alerts so it's not like the alert isn't applicable to the >device. > >I've checked a device that has a non-checked alert and its graphs and >polling look fine. > >Cheers > >On 24/10/2014 3:33 am, "Adam Armstrong" <adama(a)memetic.org> wrote: >>It means the alert has never been checked. >> >>adam. >> >>------ Original Message ------ >>From: "Ben Steele" <ben(a)bensteele.org> >>To: "Observium Network Observation System" <observium(a)observium.org> >>Sent: 10/22/2014 8:15:36 PM >>Subject: [Observium] Alerts in "Other" state >> >>>Hi, >>> >>>I have quite a few alerts stuck in the "Other" state, any idea on >>>what causes this? >>> >>>I had this problem a while ago and then it seemed to disappear after >>>creating a few more alerts and several svn updates so i'm not really >>>sure what cleared it but after adding more devices it's back again. >>>I've tried regenerating the alerts a few times but that makes no >>>difference, is there some debug that can be turned on to work out >>>what is going on here? >>> >>>It's across multiple alert checkers and even something as simple as >>>"device status eq 0" for a device down check has multiple alerts >>>stuck in "other" state. Hovering over the alert shows an empty >>>status: >>> >>> >>> >>> >>> >>>Cheers >> >>_______________________________________________ >>observium mailing list >>observium(a)observium.org >>http://postman.memetic.org/cgi-bin/mailman/listinfo/observium >>

2 1

Re: [Observium] error on svn update
by Adam Armstrong 24 Oct '14

24 Oct '14

2 3

Re: [Observium] polling issues with 5897
by Rutger Bevaart 23 Oct '14

23 Oct '14

Hi all, I can confirm that the issue is gone in 5903! Thanks Mike & Joe for the excellent input (and fix)! Let me know if any additional input, debugging or testing. Regards, Rutger On 22 Oct 2014, at 00:35, observium-request(a)observium.org wrote: > Date: Wed, 22 Oct 2014 02:35:27 +0400 > From: Mike Stupalov <mike(a)observium.org> > To: Observium Network Observation System <observium(a)observium.org> > Subject: Re: [Observium] polling issues with 5897 > Message-ID: > <CAL9uo4jwRDeDhxw6eMAxWk03zxmw-gQv3gGESyiJ5hwTkf3Row(a)mail.gmail.com> > Content-Type: text/plain; charset="utf-8" > > I make small change in r5901 (revert global var removed in 5884). > Please check with this revision. > > (But if this will fix your problem, I'll be shocked) :)

4 4

Alerts in "Other" state
by Ben Steele 23 Oct '14

23 Oct '14

Hi, I have quite a few alerts stuck in the "Other" state, any idea on what causes this? I had this problem a while ago and then it seemed to disappear after creating a few more alerts and several svn updates so i'm not really sure what cleared it but after adding more devices it's back again. I've tried regenerating the alerts a few times but that makes no difference, is there some debug that can be turned on to work out what is going on here? It's across multiple alert checkers and even something as simple as "device status eq 0" for a device down check has multiple alerts stuck in "other" state. Hovering over the alert shows an empty status: [image: Inline image 2] [image: Inline image 1] Cheers

2 2