- observium - lists.observium.org

SVN commit comments, overall usability of SVN code
by Sebastian Wiesinger 17 Jun '13

17 Jun '13

Hello, I'm testing out Observium and I like a lot of the things that Observium does. But what really irks me is the way and style of releasing new versions and committing to the SVN repository. 1) Commit messages. To quickly get a feel what changed between revisions I consult the svn log. That is the primary source of information. In the Observium SVN this often leads me to read things like this: r4147 - CODE, Y U NO LOOK RITE r4140 - Me need your soul, more souls. >:-> r4139 - you're not ready for this. :) r4125 - disable whatever the fuck that is r4101 - add some crappy errors And on and on. Things like that make me question the code quality that is committed to the repository. What kind of project do you want to be with these kind of commit messages? Mind you, I know that many people use cheeky or funny commit messages but normally they include some kind of explanation/context to see what has changed. That's missing here. This hurts the project. It projects an immature image. 2) Follow up to this: How is code quality review done? What are the procedures to decide when code is added to the repository and how do you make sure that the repository is stable for the majority of the users? Also I noticed that you do not separate code and HTML output which is another source of potential problems and accidental breakage. Is it planned to rewrite this to use a templating system? Regards Sebastian -- GPG Key: 0x93A0B9CE (F4F6 B1A3 866B 26E9 450A 9D82 58A2 D94A 93A0 B9CE) 'Are you Death?' ... IT'S THE SCYTHE, ISN'T IT? PEOPLE ALWAYS NOTICE THE SCYTHE. -- Terry Pratchett, The Fifth Elephant

1 0

Observium Map Bug
by observium＠velder.li 17 Jun '13

17 Jun '13

Hello there One of my locations is down (Frankfurt am Main, Germany). But observium shows my atlanta location offline. Mouseover still shows Frankfurt. Bug? Regards patrick

4 5

Re: [Observium] JPGraph, accurate bills, image link broken
by Adam Armstrong 12 Jun '13

12 Jun '13

Did you disable the stuff at the bottom of the config? Bart Teunis <bteunis(a)a2b-internet.com> wrote: >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

1 0

JPGraph, accurate bills, image link broken
by Bart Teunis 12 Jun '13

12 Jun '13

To all, I updated my observium and found out that jpgraph is not coming with the SVN anymore. So I installed this according the instructions I found in my observium installation. Some how the link to the accurate img remains broken, so I could use some help with this, two questions: 1. Is this a known issue 2. How can I solve this Regards, Bart Teunis Bart Teunis | A2B Internet BV | +31 85 90 20 410 ( Office ) | +31 6 57 88 48 47 ( Dutch cell ) | <mailto:bteunis@a2b-internet.com> bteunis(a)a2b-internet.com |

1 0

AP Statistics from Cisco WLC
by Fletcher Haynes 11 Jun '13

11 Jun '13

Hello, Observium has been awesome so far...I am wondering if there are plans to support AP discovery and statistics from Cisco WiSMs? We have several hundred APs in lightweight mode we would like to monitor. If there are no current development efforts in that area, I may take a crack at it...

1 0

Inversing in and out
by Benjamin Abadie 10 Jun '13

10 Jun '13

Hi, I'm looking for a way to reverse in and out position in port graphs. That is, display positive values for outbound, and negative ones for inbound. It can be interesting for a CDN, which mostly send its data to other ASes (booooo ! ;] ) I found a variable "$inverse" which *seems* to do what I want to do : > /opt/observium 0 # grep --exclude-dir=".svn" -n -r "\$inverse" > html/includes/graphs/generic_multi_data.inc.php:48: if ($inverse) { $in = 'out'; $out = 'in'; } else { $in = 'in'; $out = 'out'; } > html/includes/graphs/generic_multi_bits.inc.php:39: if ($inverse) { $in = 'out'; $out = 'in'; } else { $in = 'in'; $out = 'out'; } > html/includes/graphs/generic_data.inc.php:28:if ($inverse) { $in = 'out'; $out = 'in'; } else { $in = 'in'; $out = 'out'; } > html/includes/graphs/multiport/bits_trio.inc.php:17: if (strstr($inverse, "a")) { $in = "OUT"; $out = "IN"; } else { $in = "IN"; $out = "OUT"; } > html/includes/graphs/multiport/bits_trio.inc.php:37: if (strstr($inverse, "b")) { $in = "OUT"; $out = "IN"; } else { $in = "IN"; $out = "OUT"; } > html/includes/graphs/multiport/bits_trio.inc.php:57: if (strstr($inverse, "c")) { $in = "OUT"; $out = "IN"; } else { $in = "IN"; $out = "OUT"; } > html/includes/graphs/multiport/bits_duo.inc.php:45:if ($inverse) { $in = 'out'; $out = 'in'; } else { $in = 'in'; $out = 'out'; } But setting it to true does not seem to change anything. Am I completely mistaken ? If not, how should I use this ? I know it's not an official feature (or I would have found a config option) but I'm just messing around a little with observium's code... ;) Regards, Benjamin Abadie

3 5

IPMI broken again?
by observium＠velder.li 10 Jun '13

10 Jun '13

Hi Is anyone also seeing problems with IPMI? (NA at Fanspeed, flapping fans etc..) Regards

3 2

Re: [Observium] backend DB
by Adam Armstrong 09 Jun '13

09 Jun '13

The post from the ex-mysql security guy from a few months ago detailing oracles lack of fucks? Alex Pressé <alex.presse(a)gmail.com> wrote: >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

1 0

Negative values in the ICMP graphics
by Javier Lopez 08 Jun '13

08 Jun '13

Hello, I'm running Observium 0.13.6.4115 updated today and I see negative values in the ICMP graphics, it also fails to draw the preview images. I update Observium weekly from svn. http://i.imgur.com/xGCoaHJ.png http://i.imgur.com/3Hmtj2A.png The same Draw error is visible in the ICMP Statistics, ICMP Informational Statistics, IP Statistics and UDP Statistics graphics. Any clue to get it fix it is welcome, thanks in advance. Cheers

5 9

Setup problem
by Richard Davies 07 Jun '13

07 Jun '13

Hi I am looking to implement snmp monitoring for the servers and network devices we support. I am looking at opennms and observium, and have them both running, monitoring some of our internal servers and devices. Observium is running on debian 7 - Wheezy. I believe I followed the debian specific installation instructions exactly, and Observium is up and running. I have configured it to monitor 2 of our Windows servers, 2 LINUX servers, a cisco router and sonicwall firewall. The initial discovery seems to have worked. However the olling does not seem to be working from cron. Looking at the observium.log > /opt/observium/discovery.php new June 5, 2013, 15:25 - 0 devices > discovered in 0. > 006 secs > /opt/observium/discovery.php new June 5, 2013, 15:30 - 0 devices > discovered in 0. > 007 secs > /opt/observium/discovery.php new June 5, 2013, 15:35 - 0 devices > discovered in 0. > 005 secs > ... discovery is running from cron every 5 mins as expected, but there is no mention in the log of poller.php running. I am sure I have been stupid and am missing something simple. All the graphs have a title Error Drawing Graph, guess that is because there is no data. Any help and advice would be appreciated, where to look to resolve the issues. Richard

11 22

Re: [Observium] Setup problem
by Adam Armstrong 07 Jun '13

07 Jun '13

Its largely unreliable because it relies on parts of the poller code working properly when run as the apache process. We rarely test it, and a lot of that code is horrific to debug. Adam. Alex Pressé <alex.presse(a)gmail.com> wrote: >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

1 0

frontpage event status > 24h?
by Rob VanHooren 07 Jun '13

07 Jun '13

hi, is there a quick way to adjust the frontage status to show events longer than 24h? i.e., tweak the time period for this config.php:$config['frontpage']['device_status']['ports'] = true; // Show the down ports (Last 24 hours) to be longer. dont' care so much about ports, but more the larger transient events (e.g., device rebooted). anyone? thanks! R. (ps adam, pls. nuke the nonmember @gmail dupe post that's queued up for moderation)

3 2

Observium Ubuntu Server. Ports>All ports blank page
by Dhendup 07 Jun '13

07 Jun '13

Dear Team, Please do help me solve the following issues: Recently I added few more devices into the observium. To my surprise, when I click on Ports---> All Ports, the page show is blank. Before the addition of the additional devices, the interface work fine. I tried to dig the issues by putting debug=yes into the URL: I get the following output. SQL[SELECT sensor_class,COUNT(sensor_id) AS c FROM sensors GROUP BY sensor_class ORDER BY sensor_class ] SQL[SELECT COUNT(`app_id`) FROM `applications`] devices WHERE status = '1' AND `ignore` = '0' AND `disabled` = '0'] SQL[SELECT COUNT(*) FROM devices WHERE status = '0' AND `ignore` = '0' AND `disabled` = '0'] SQL[SELECT COUNT(*) FROM devices WHERE `ignore` = '1' AND `disabled` = '0'] SQL[SELECT COUNT(*) FROM devices WHERE `disabled` = '1'] SQL[SELECT COUNT(*) FROM ports WHERE deleted = '0'] SQL[SELECT COUNT(*) FROM ports AS I, devices AS D WHERE I.deleted = '0' AND I.ifOperStatus = 'up' AND I.ignore = '0' AND I.device_id = D.device_id AND D.ignore = '0'] SQL[SELECT COUNT(*) FROM ports AS I, devices AS D WHERE I.deleted = '0' AND I.ifOperStatus = 'down' AND I.ifAdminStatus = 'up' AND I.ignore = '0' AND D.device_id = I.device_id AND D.ignore = '0'] SQL[SELECT COUNT(*) FROM ports AS I, devices AS D WHERE I.deleted = '0' AND I.ifAdminStatus = 'down' AND I.ignore = '0' AND D.device_id = I.device_id AND D.ignore = '0'] SQL[SELECT COUNT(*) FROM ports AS I, devices AS D WHERE I.deleted = '0' AND D.device_id = I.device_id AND (I.ignore = '1' OR D.ignore = '1')] SQL[SELECT COUNT(*) FROM ports AS I, devices AS D WHERE I.deleted = '0' AND D.device_id = I.device_id AND (I.ignore = '0' OR D.ignore = '0') AND (I.ifInErrors_delta > '0' OR I.ifOutErrors_delta > '0')] SQL[SELECT COUNT(service_id) FROM services] SQL[SELECT COUNT(service_id) FROM services WHERE service_status = '1' AND service_ignore ='0'] SQL[SELECT COUNT(service_id) FROM services WHERE service_status = '0' AND service_ignore = '0'] SQL[SELECT COUNT(service_id) FROM services WHERE service_ignore = '1'] SQL[SELECT COUNT(service_id) FROM services WHERE service_disabled = '1'] SQL[SELECT COUNT(sensor_id) FROM sensors] SQL[SELECT COUNT(sensor_id) FROM sensors WHERE sensor_status = '1' AND sensor_ignore ='0'] SQL[SELECT COUNT(sensor_id) FROM sensors WHERE sensor_status = '0' AND sensor_ignore = '0'] SQL[SELECT COUNT(sensor_id) FROM sensors WHERE sensor_ignore = '1'] Device Alerts SQL[SELECT * FROM `devices` AS D WHERE D.status = 0 AND D.ignore = 0 ORDER BY D.hostname ASC] SQL[SELECT * FROM `devices` AS D WHERE D.status = 1 AND D.uptime > 0 AND D.uptime < 84600 AND D.ignore = 0 ORDER BY D.hostname ASC] SQL[SELECT * FROM `ports` AS I LEFT JOIN `devices` AS D ON I.device_id = D.device_id WHERE I.ifOperStatus = 'down' AND I.ifAdminStatus = 'up' AND I.ignore = 0 AND I.deleted = 0 AND D.ignore = 0 AND I.ifLastChange >= DATE_SUB(NOW(), INTERVAL 24 HOUR) ORDER BY I.ifLastChange DESC, D.hostname ASC, I.ifDescr * 1 ASC ] SQL[SELECT * FROM `services` AS S LEFT JOIN `devices` AS D ON S.device_id = D.device_id WHERE S.service_status = 'down' AND S.service_ignore = 0 AND D.ignore = 0 ORDER BY D.hostname ASC] SQL[SELECT * FROM `devices` AS D LEFT JOIN bgpPeers AS B ON B.device_id = D.device_id WHERE (bgpPeerAdminStatus = 'start' OR bgpPeerAdminStatus = 'running') AND bgpPeerState != 'established' AND D.ignore = 0 ORDER BY D.hostname ASC] Your help will be appreciated. Regards

3 2

24h && DWIM
by Rob VanHooren 06 Jun '13

06 Jun '13

hrrrumph. who'da thought a grep for 86400 isn't the same as a grep for 84600. $ sudo yum install do-what-I-mean /grin/ Thanks Paul. R. On 2013-06-06, at 6:59 PM, observium-request(a)observium.org wrote: > root@localhost:/opt/observium# grep uptime includes/defaults.inc.php > $config['uptime_warning'] = "84600"; // Time in > seconds to display a "Device Rebooted" Alert. 0 to disable warnings.

2 1

Re: [Observium] Add ignore iftype
by Adam Armstrong 06 Jun '13

06 Jun '13

In that case it needs to be matched using the ifdescr filters not the iftype filters. :-D Adam. Nikolay Shopik <shopik(a)inblock.ru> wrote: >looks like its "other" > >On 05.06.2013 21:31, Adam Armstrong wrote: >> You managed to find the ifDescr in an snmpwalk, try looking for the ifType. >> >> Adam. >> >> Kody <kklorres(a)nexeosolutions.com> wrote: >> >>> Adam Armstrong <adama@...> writes: >>> >>>> >>>> >>>> What ifType? >>>> >>> >>> Adam, >>> If the above is not what your are looking for and EFXS is not the ifType. >>> Then I do not know what you are looking for. >>> >>> >>> _______________________________________________ >>> observium mailing list >>> observium(a)observium.org >>> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium >> _______________________________________________ >> observium mailing list >> observium(a)observium.org >> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium >> >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

2 1

cannot add hosts anymore
by Ted 06 Jun '13

06 Jun '13

I am running version 0.13.6.4133 and noticed that I cannot add hosts anymore. The web interface doesn't error out but when I try the command line I get: PHP Fatal error: Call to undefined function getHostOS() in /opt/observium/includes/functions.php on line 497 Can anyone else reproduce that...or advise on how to fix it. Thank you.

2 1

Re: [Observium] cannot add hosts anymore
by Adam Armstrong 06 Jun '13

06 Jun '13

I'm sure comrade landy will fix it shortly :-) Ted <theodore.dd(a)gmail.com> wrote: >I am running version 0.13.6.4133 and noticed that I cannot add hosts anymore. >The web interface doesn't error out but when I try the command line I get: > > PHP Fatal error: Call to undefined function getHostOS() in >/opt/observium/includes/functions.php on line 497 > > >Can anyone else reproduce that...or advise on how to fix it. Thank you. > >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

1 0

Billing started giving unrealistic P95s
by Hynek Med 06 Jun '13

06 Jun '13

Probably somewhere between r3465 to r3981 (yes, I know) billing started producing wrong results. now at 4102, the situation remains the same. I do run the discovery script after each update. Graphs (quick graphs) look fine with real traffic accounted, but the calculated P95 is unrealistically low (something like 1/3 of the real P95, but this ratio is different for each case). Both with single VLAN and multiple VLAN bills. One of the bills (with id=1) works completely fine though. Any ideas where to look at / what to compare at the working case with the non-working case? Thanks, Hynek -- Hynek Med PRO-ZETA a.s. Tiskarska 10/257 108 00 Praha 10 Czech Republic

2 6

ExtremeXOS cpu and temp
by Marco Pirovano 06 Jun '13

06 Jun '13

Hello, I'm running Observium 0.13.6.4110 on FreeBSD 8.3 In our network we have a lot of ExtremeXOS switch, so I tried to write a code for CPU e Temperature monitoring. For Temperature OID = 1.3.6.1.4.1.1916.1.1.1.8.0 "Current temperature in degrees celcius measured inside device enclosure". For CPU OID = 1.3.6.1.4.1.1916.1.32.1.4.1.9.0 "The CPU utilization by system resources. This is the utilization of the system resources in the last 5 minutes". I test the script with: - BD8810 ExtremeXOS version 12.6.2.10 - X460 ExtremeXOS version 12.5.4.5 - X450e ExtremeXOS version 12.6.4.1 - X450a ExtremeXOS version 12.3.3.6 - X440 ExtremeXOS (Stack) version 15.1.4.3 For Memory Observium displayed the status without doing anything. This is my first tentative, so test and let me know. Thanks. Ciao Marco -- Marco Pirovano Infrastrutture e Tecnologie Area Sistemi Informatici e Telematici Universita' Bocconi via Gobbi, 5 - 20136 Milano Tel. +39 02 5836.3173 Fax. +39 02 5836.3160 --- Se credi nei giovani di talento, investi sul loro futuro. Dai il tuo 5x1000 alla Bocconi cf: 80024610158. E' importante. E' per loro. www.unibocconi.it/5x1000 Please note that the above message is addressed only to individuals filing Italian income tax returns. ---

1 0

Re: [Observium] Add ignore iftype
by Adam Armstrong 05 Jun '13

05 Jun '13

You managed to find the ifDescr in an snmpwalk, try looking for the ifType. Adam. Kody <kklorres(a)nexeosolutions.com> wrote: >Adam Armstrong <adama@...> writes: > >> >> >> What ifType? >> > >Adam, >If the above is not what your are looking for and EFXS is not the ifType. >Then I do not know what you are looking for. > > >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

2 1

Add ignore iftype
by Kody 05 Jun '13

05 Jun '13

I have many of the following ports on Cisco voice routers with PRI voice lines. I would like add them to the ignore list but everything I try will keep finding them. Am I missing something? This is from the snmpwalk. ifDescr.536 = STRING: EFXS 50/0/400 *** this one router has 400+ of them and they all show up in the router *** ifDescr.537 = STRING: Voice Encapsulation (POTS) Peer: 20208 *** Same here many show up on each router *** Some of the lines I have added to see if it will ignore them. But no luck. $config['bad_iftype'][] = "EFXS"; $config['bad_iftype'][] = "EFXS 50/0/1"; $config['bad_if_regexp'][] = "EFXS.*"; Thanks for any help.

4 12

IP SLAs
by Robert Williams 05 Jun '13

05 Jun '13

Hi guys, I’ve just noticed that Observium is showing up Cisco IP SLA targets (didn’t know the feature was there!) – however it doesn’t actually log any values. All the entries IDs and targets are in there (correctly) though: [cid:image003.png@01CE61C9.7B3A9280] It would be cool if they worked (now I’ve spotted it!). They are all recording NaN so I’m not sure if there is simply something not setup correctly on the Cisco side in order for Observium to be able to read it; or maybe if there a ‘supported’ setup of the IPSLA entries on the Cisco I’m meant to follow? The existing config for them is pretty uninteresting: ip sla 139 icmp-echo 10.0.5.139 frequency 10 ip sla schedule 139 life forever start-time now ip sla reaction-configuration 139 react timeout threshold-type immediate action-type trapOnly If a debug would help then I’ll post one, but first I wanted to check if it’s just not fully supported yet (or a known issue). Just curious as I’ve only just noticed it, but it’s probably been there forever ☺ Cheers! Robert Williams Custodian Data Centre Email: Robert(a)CustodianDC.com http://www.CustodianDC.com

2 2

High latency poller performance
by Adam Armstrong 05 Jun '13

05 Jun '13

Hi All, I've been testing some improvements to the Poller when deployed in high latency situations (polling hosts on different continents, for example). It involved some SNMP tweaks to pull more OIDs per round-trip, but has varying support in different OSes, so we're making it a hardcoded per-OS attribute. It seems to work particularly well on Cisco IOS, IOS-XR and Arista devices. http://alpha.memetic.org/~adama/snaps/fwy8vmmvii.png Shows the performance gains on a selection of Cisco IOS devices. I'll be committing the changes later today, disabled as default. If anyone wants to test them for us before we enable them as default, give us a shout. adam.

1 0

Re: [Observium] eigrp b0rken?
by Adam Armstrong 05 Jun '13

05 Jun '13

Eigrp was committed by mistake a while back. One day it might get finished, time permitting :-) Rob VanHooren <rob_vanhooren(a)mac.com> wrote: >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

1 0

Observium Integration with Munin
by Scott A. Davis 04 Jun '13

04 Jun '13

Hello! First of all, thanks for a great monitoring tool! Second, apologies in advance if my topic has been covered elsewhere. I scanned the mailing list and documentation but did not find what I need. I have Observium and Munin functioning separately on the sane node. I would like to import some of the graphs that Munin has created into Observium. I am not exactly sure how to even start that project. - Is there a simple way to do it? - Are there step-by-step instructions that a semi-novice can follow? Thanks in advance! [SD]

1 0

eigrp b0rken?
by Rob VanHooren 04 Jun '13

04 Jun '13

hi, all. as at today's r4116… who can say whether EIGRP is in a workable state? have an example c3845 (advsecurity-15.1(4)M5, iirc) don't seem to get much (debugs below) … nothing shows up under routing tab. is a bug? or is eigrp coverage a work-in-progress? thoughts? thanks, R. === ! appears that there are bits for it… ! # grep -R eigrp * includes/functions.inc.php: case "eigrp": includes/.svn/text-base/functions.inc.php.svn-base: case "eigrp": pages/device.inc.php: $device_routing_count['eigrp'] = dbFetchCell("SELECT COUNT(*) FROM `eigrp_ports` WHERE `device_id` = ?", array($device['device_id'])); pages/device.inc.php: if ($device_routing_count['eigrp']) { $routing_tabs[] = 'eigrp'; } pages/device/routing/eigrp.inc.php: $navbar['options'][$type]['url'] = generate_url(array('page' => 'device', 'device' => $device['device_id'], 'tab' => 'routing', 'proto' => 'eigrp', 'view' => $type )); pages/device/routing/eigrp.inc.php: include("eigrp_ports.inc.php"); pages/device/routing/.svn/all-wcprops:eigrp.inc.php pages/device/routing/.svn/all-wcprops:/svn/!svn/ver/3884/observer/trunk/html/pages/device/routing/eigrp.inc.php pages/device/routing/.svn/entries:eigrp.inc.php pages/device/routing/.svn/text-base/eigrp.inc.php.svn-base: $navbar['options'][$type]['url'] = generate_url(array('page' => 'device', 'device' => $device['device_id'], 'tab' => 'routing', 'proto' => 'eigrp', 'view' => $type )); pages/device/routing/.svn/text-base/eigrp.inc.php.svn-base: include("eigrp_ports.inc.php"); pages/.svn/text-base/device.inc.php.svn-base: $device_routing_count['eigrp'] = dbFetchCell("SELECT COUNT(*) FROM `eigrp_ports` WHERE `device_id` = ?", array($device['device_id'])); pages/.svn/text-base/device.inc.php.svn-base: if ($device_routing_count['eigrp']) { $routing_tabs[] = 'eigrp'; } includes/functions.inc.php: return "EIGRP"; includes/.svn/text-base/functions.inc.php.svn-base: return "EIGRP"; pages/device/routing/eigrp.inc.php:$navbar['brand'] = "EIGRP"; pages/device/routing/.svn/text-base/eigrp.inc.php.svn-base:$navbar['brand'] = "EIGRP"; # grep eigrp config.php $config['enable_eigrp'] = 1; // Enable EIGRP session collection and display THEROUTER#sho ip eigrp neigh EIGRP-IPv4 Neighbors for AS(2) H Address Interface Hold Uptime SRTT RTO Q Seq 4 192.168.1.1 Tu15 11 00:08:22 292 1752 0 55125 3 192.168.2.2 Tu93 10 4d02h 20 200 0 8264 2 192.168.3.3 Tu92 12 4d02h 45 270 0 18728 1 192.168.4.4 Tu16 12 4d02h 24 200 0 37045 0 192.168.5.5 Tu106 13 4d02h 20 200 0 554693´ # ./addhost.php THEROUTER mySNMPv2comm Trying community mySNMPv2comm ... Discovering THEROUTER (192) THEROUTER 192 ios (cisco) {snip} Discovered in 101.6 seconds # ./discovery -d h THEROUTER {no eigrp here} # ./poller.php -d -h THEROUTER {snip} including: includes/polling/cisco-eigrp.inc.php SQL[SELECT * FROM `eigrp_ports` WHERE `device_id` = '192'] Array ( ) DEBUG: SNMP Auth options = -v2c -c mySNMPv2comm /usr/bin/snmpbulkwalk -v2c -c mySNMPv2comm -OQUs -m CISCO-EIGRP-MIB -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp:/opt/observium/mibs/cisco udp:THEROUTER:161 CEigrpInterfaceEntry ! snipped, just one of each shown ! cEigrpPeerCount.0.2.25 = 0 cEigrpXmitReliableQ.0.2.25 = 0 cEigrpXmitUnreliableQ.0.2.25 = 0 cEigrpMeanSrtt.0.2.25 = 0 cEigrpPacingReliable.0.2.25 = 1 cEigrpPacingUnreliable.0.2.25 = 0 cEigrpMFlowTimer.0.2.25 = 0 cEigrpPendingRoutes.0.2.25 = 0 cEigrpHelloInterval.0.2.25 = 5 cEigrpXmitNextSerial.0.2.25 = 0 cEigrpUMcasts.0.2.25 = 0 cEigrpUUcasts.0.2.25 = 0 cEigrpMcastExcepts.0.2.25 = 0 cEigrpCRpkts.0.2.25 = 0 cEigrpAcksSuppressed.0.2.25 = 0 cEigrpRetransSent.0.2.25 = 0 cEigrpOOSrvcd.0.2.25 = 0 cEigrpAuthMode.0.2.25 = none cEigrpAuthKeyChain.0.2.25 = youWish 0 2 25 SQL[SELECT * FROM `ports` WHERE `device_id` = '192' AND `ifIndex` = '25'] SQL[SELECT * FROM `devices` WHERE `device_id` = '192'] SQL[INSERT INTO `eigrp_ports` (`eigrp_vpn`,`eigrp_as`,`eigrp_ifIndex`,`port_id`,`device_id`,`eigrp_peer_count`) VALUES ('0','2','25','9391','192','0')] +RRD[cmd[update /opt/observium/rrd/THEROUTER/eigrp_port-0-2-25.rrd N:0:0:0:0:0:0:0:0:0:0] stdout[OK u:0.01 s:0.01 r:1414.61] stderr[]]

1 0

LDAP authentication changes
by Tom Laermans 04 Jun '13

04 Jun '13

Hi everyone, I've made some modifications to the LDAP auth module; if you SVN up to r4114 or higher, please check if your LDAP auth is still working if you are using it. As far as I know (and on my test system), all defaults are sensible and should work just as before without changing anything, however some new features appeared, specifically to support active directory. An example config for AD is on the wiki. Nested groups don't work yet, I will hopefully get around to that soon. Thanks, Tom

1 0

(no subject)
by Pedro Serotto 04 Jun '13

04 Jun '13

Hi all, I would like to hide some unused ports on some device. I see the filters configuration "$config['bad_if'] = array" but if I correctly understand this filter is apply to all devices. Is there a way to do different filter for different device ? BR Pedro

2 5

ASR9k users?
by Robert Williams 03 Jun '13

03 Jun '13

Hi, Just curious if anyone is currently using Observium to monitor any Cisco ASR9k based devices? I can see that it has correctly identified all the various hardware modules, line cards and temperature sensors etc. but isn’t graphing things like temperatures or light levels like it does for our Cat6K workforce. Since this is the first ASR device I’ve tried polling on Observium I’m not sure if it’s just not fully supported or if we have some SNMP restrictions on the Cisco which are preventing it from getting what it needs. Any feedback or other experiences are welcome! Cheers all, Robert Williams Custodian Data Centre Email: Robert(a)CustodianDC.com http://www.CustodianDC.com

3 5

Memcached updates
by Britt Treece 03 Jun '13

03 Jun '13

Starting to use Memcached and found some of the current Memcached bits in Observium not working optimally. I wrote a new unix-agent script using python rather than PHP. The previous script relied on PECL memcached but didn't do anything for installations using PECL memcache. After a bunch of trial and error I determined that it was easiest to write the script in Python which requires the python-memcached library and is readily available on different OS types and versions. I also modified the applications/memcached.inc.php and polling/unix-agent.inc.php to accommodate the new script. They will also accurately separate different instances of memcached running on the same server as reported by the script. I'm still trying to work out a couple things. The main thing is when you enable Memcached as an application for a device Observium automatically creates an entry in the database but the instance is NULL. When the unix-agent reports on the Memcached instance(s) running on the server new entries get created in the database for each instance and the app with the NULL app_instance never gets used. I've been manually removing it from the database. If possible I would like to work it out so that it doesn't get created automatically. If that's not possible I was going to look into updating that entry with the first real instance reported by the agent. Also wanting to get the Apps tab for the Device to better identify each instance and maybe work on some additional metrics reporting for Memcached specific stats. Patch attached, let me know if it is acceptable or not... Britt Treece

3 4

Re: [Observium] (no subject)
by Adam Armstrong 02 Jun '13

02 Jun '13

Observium is designed for use in a service provider environment with a main design assumption being that one does not necessarily know what one wants to monitor until the data is needed. We purposefully make it difficult to delete or obscure port information, because that fits our requirements. Adam. Jason Lixfeld <jason(a)lixfeld.ca> wrote: >Why do you say that, Adam? I can sympathize with Pedro because I find myself faced with the same issues. > >In my specific use case, I don't have any interest in seeing interfaces in Device > Ports view that are down/down, dwdm (which I don't use and aren't active anyway, yet they still show up for some reason), loopback or vlan, personally. They tend to make it difficult to efficiently scroll through the ports page to find information that I am interested in (ie: which of my customers just started sending 4Gbps of traffic). > >Those types of ports I never care about seeing, so ignoring them in a global scope would be wonderful :) > >If I'm using Observium wrong too, I'd love to know what the intended right usage is :) > >On 2013-06-02, at 3:06 PM, Adam Armstrong <adama(a)memetic.org> wrote: > >> If you are hiding ports, you are using observium wrong. >> >> Adam. >> >> Pedro Serotto <pedro.serotto(a)yahoo.es> wrote: >> >> Hi all, >> I would like to hide some unused ports on some device. >> >> I see the filters configuration "$config['bad_if'] = array" but if I correctly understand this filter is apply to all devices. >> >> Is there a way to do different filter for different device ? >> >> BR >> >> Pedro >> >> _______________________________________________ >> observium mailing list >> observium(a)observium.org >> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium > >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

2 1

Re: [Observium] (no subject)
by Adam Armstrong 02 Jun '13

02 Jun '13

If you are hiding ports, you are using observium wrong. Adam. Pedro Serotto <pedro.serotto(a)yahoo.es> wrote: >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

3 2

Switching back to old location behavior
by Bryan Socha 01 Jun '13

01 Jun '13

I just updated and locations are all over the place. Is there a way to set it back to the old behavior where it groups by the snmp-location setup in each device? Thanks, Bryan Socha

2 1

Cisco 4500X VSS, no cpu values
by Bryan Socha 01 Jun '13

01 Jun '13

Hello, We added some Cisco 4500X switches to our network with a VSS setup and observium shows 0 for the cpu usage. Any ideas on how to get it to populate? Bryan Socha

1 0

Polling
by Wes Kennedy 01 Jun '13

01 Jun '13

I recently added in 25 new routers to be monitored and as of two hours later, they haven't been polled. They have all been Discovered properly, but not polled. All of my existing hosts are polling just fine. Has anyone seen this? -Wes

6 7

add power usuages
by Khapare Joshi 31 May '13

31 May '13

I added one of my HP enclouser, and ran the ./discovery.php -h it does not discover these : Temperatures: Humidity : Voltages: Frequencies: Current: Power: Fanspeeds : how do i get these into my observium ? also, is there a way to integrate total power in observium ? I get the total by snmpwalk : snmpwalk -c public -v2c test01.local enterprises.232.22.2.5.1.1.1.10 |awk '{ sum+=$4} END {print sum}' thank KV

3 2

switch errors meaning
by Mihai - Cristian Satmarean 31 May '13

31 May '13

Hi, I would like to ask what the switch errors mean, I have approx. 300 m Discards In on one several ports constantly and on other ports Discards Out. Please help me understand what is that. Thank you!

5 9

voltaire device
by Khapare Joshi 30 May '13

30 May '13

i have added my infiniband switch. It adds and produce the graph etc. but observium seem not recognize. as its icon comes with ? . maybe its matter of adding image for that device ? same thing with HP storage p2000 boxes Has anybody monitoring Infiniband switches with observium ? K

1 0

Help with ldap group access/permissions
by Tracy Phillips 30 May '13

30 May '13

Hi, I have configured Observium to authenticate against ldap and it works great. This is my configuration for authentication: $config['auth_mechanism'] = "ldap"; $config['auth_ldap_version'] = 3; $config['auth_ldap_server'] = "ldap.example.net"; $config['auth_ldap_port'] = 389; $config['auth_ldap_starttls'] = FALSE; $config['auth_ldap_prefix'] = "uid="; $config['auth_ldap_suffix'] = ",ou=People,dc=example,dc=net"; $config['auth_ldap_group'] = "cn=observium,ou=Groups,dc=example,dc=net"; Everything works as expected until I get to this point. When I go to add a new device or view the global configuration, I get the message: ERROR You have insufficient permissions to view this page. I have confirmed that I am in the observium-admin group. This is what I have for group permissions: $config['auth_ldap_groupbase'] = "ou=Groups,dc=example,dc=net"; $config['auth_ldap_groups']['observium-admin']['level'] = 10; There is nothing relevant in the observium.log. Is there something that I am missing or does anyone have thought that might help me resolve this issue? Thanks in advance for any hints you can throw my way! Tracy

1 1

I want to say THANK YOU
by Zach Underwood 30 May '13

30 May '13

I work for a small ISP. For the past few weeks things have been running slow on the network. After working on and off of the problem for some time will no fix, I had to take a break from the problem. So I setup a VM running centos, I installed your software (the rhel howto has some mistakes). I then added the core switches and routers, 9 devices later I had the core mapped. After the software was running for 10 mins I started to see a problem. The core router that links up two ISP with BGP was having interface errors. Comes to find out the interfaces was running at half duplex. After fixing the duplex error the network came back to life. I would like to say thank for the software and I am a kickstarter supporter. -- Zach Underwood (RHCE,RHCSA,RHCT) My website <http://zachunderwood.me> My photes <http://zunder1990.openphoto.me>

1 0

SNMP Printer polling
by Matthew Coene 30 May '13

30 May '13

Good morning everyone... Ok.. I have my Canon printers working polling when using jetdirect as the OS type in separate include files in /observium/includes/discovery/os/, I was hoping to make them their own icon, and subset in the definitions.inc.php but seeing as how jetdirect is already declared as its own (I assume string?) (forgive me I am not a programmer)... Can I still make a separate definitions entry code block and split the 'jetdirect' code to a separate entity so I can leave the HP printers stuff alone and do the Canon separately?? Cheers, M.

1 0

Re: [Observium] Canon printers
by Adam Armstrong 29 May '13

29 May '13

Tom has the printer fetish. It's probably not hard for him to add if proper MIBs exist. I think you're all weirdos for monitoring printers, though... Adam. Matthew Coene <mcoene(a)Bacardi.com> wrote: > >Good afternoon anyone... > >Was just curious as to whether anyone has had any luck integrating MIBS so >Canon iR series can be parsed as proper printers and have stats / toner >polled correctly?? > >I was a little spoiled at my last position where all printers were HP / >Konica and supported out of the box... Now not so much... > >Thanks in advance... > >Matt.C. > >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

4 5

Re: [Observium] SNMP Ubuntu
by Tom Laermans 29 May '13

29 May '13

Same way you monitor any other net-snmp box; see our documentation. Sent from my mobile, please excuse brevity and spelling errors. ----- Reply message ----- From: "Alex Vitola" <alex.vitola(a)gmail.com> To: "L - Observium" <observium(a)observium.org> Subject: [Observium] SNMP Ubuntu Date: Wed, May 29, 2013 19:36 how to monitor cpu and memory using snmp in ubuntu 12:04? Only network graphs are shown Alex Sandro de Lacerda Vitola http://br.linkedin.com/in/vitolahttp://www.vitola.net.br

2 1

Re: [Observium] Little more help with SNMP/MIB for Canon.
by Adam Armstrong 29 May '13

29 May '13

Try to detect it using sysobjectid. Doing it via sysdescr is discouraged in new code. You can probably catch them all with a single string match that way. Adam. Matthew Coene <mcoene(a)Bacardi.com> wrote: >Holy Hell... It works... > >Copied the jetdirect.inc.php to a new file called canon.inc.php > >snmpwalked... seems the sysDescr is unique to each model so I will just >replicate as necessary to reflect our fleet... > >I suppose the issue now is that because it is polling via "jetdirect" it is >assigning a HP logo... > >Small potatoes however... > >I am getting consumable and page counts for the beancounters... > >Cheers, > >Matt. > > > > >From: Adam Armstrong <adama(a)memetic.org> >To: Observium Network Observation System <observium(a)observium.org>, >Date: 05/29/2013 05:08 PM >Subject: Re: [Observium] Little more help with SNMP/MIB for Canon. >Sent by: "observium" <observium-bounces(a)observium.org> > > > >You'd need to create a new is type for canon printers. > >Then you can permit that is to find things using the jet direct MIBs. > >Adam. > >Matthew Coene <mcoene(a)Bacardi.com> wrote: > >> >>Good Afternoon All, >> >>Tom, Looks like you were right... Tried snmpwalking our Canon devices >with >>the HP-LASERJET-COMMON MIB... >> >>Looking at the rest of the code, specifically the "jetdirect" bits that >you >>mentioned... >> >><?php >> >>if (!$os) >>{ >> if (strstr($sysDescr, "JETDIRECT")) { $os = "jetdirect"; } >> else if (strstr($sysDescr, "HP ETHERNET MULTI-ENVIRONMENT")) { $os = >>"jetdirect"; } >> else if (strstr($sysObjectId, ".1.3.6.1.4.1.11.1")) { $os = >>"jetdirect"; } >>} >> >>I can say that the OID string that seems to be the most common to the ID >>is... (actually there are a few) >> >>iso.3.6.1.2.1.43.9.2.1.8.1.1 = STRING: "Canon" >> >> >>I am hoping there might be some way to just adapt the existing framework, >>to use the the new OID / MIB to report the Canon's specifically instead of >>generic? >> >>Am I even remotely on some sort of right track or way off on some far >>removed dead end tangent?? >> >>Cheers, >> >>Matt. >> >> >>_______________________________________________ >>observium mailing list >>observium(a)observium.org >>http://postman.memetic.org/cgi-bin/mailman/listinfo/observium >_________________________ >______________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium > > > >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

2 1

Re: [Observium] Little more help with SNMP/MIB for Canon.
by Adam Armstrong 29 May '13

29 May '13

You'd need to create a new is type for canon printers. Then you can permit that is to find things using the jet direct MIBs. Adam. Matthew Coene <mcoene(a)Bacardi.com> wrote: > >Good Afternoon All, > >Tom, Looks like you were right... Tried snmpwalking our Canon devices with >the HP-LASERJET-COMMON MIB... > >Looking at the rest of the code, specifically the "jetdirect" bits that you >mentioned... > ><?php > >if (!$os) >{ > if (strstr($sysDescr, "JETDIRECT")) { $os = "jetdirect"; } > else if (strstr($sysDescr, "HP ETHERNET MULTI-ENVIRONMENT")) { $os = >"jetdirect"; } > else if (strstr($sysObjectId, ".1.3.6.1.4.1.11.1")) { $os = >"jetdirect"; } >} > >I can say that the OID string that seems to be the most common to the ID >is... (actually there are a few) > >iso.3.6.1.2.1.43.9.2.1.8.1.1 = STRING: "Canon" > > >I am hoping there might be some way to just adapt the existing framework, >to use the the new OID / MIB to report the Canon's specifically instead of >generic? > >Am I even remotely on some sort of right track or way off on some far >removed dead end tangent?? > >Cheers, > >Matt. > > >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

2 1

Hardware scaling
by Adam Armstrong 29 May '13

29 May '13

Hi All, I put together a very basic set of scaling guidelines : http://www.observium.org/wiki/Hardware_Scaling My sample size is quite limited, do any of you have useful numbers from large (>10k ports) installations? How much RAM does your /ports/ page take? (mouse over the perf on the bottom bar). How much average CPU load does your system have, and what CPU is it? (assuming the host is Observium-only) What storage back-end do you use, and how well is it scaling? Thanks, adam.

3 4

map not showing
by Khapare Joshi 29 May '13

29 May '13

newbie here, I have 0.13.5.4047 version and world map is not showing in front page. Also, i could not configure alert, when i click on alert it does not gives me any option to configure. maybe this is simple somewhere config ? but earlier I installed 0.13.5.3992 and world map is displayed.

1 1

Re: [Observium] map not showing
by Adam Armstrong 29 May '13

29 May '13

There is no alerting. The tab is a ui placeholder. You need to explicitly enable the map by adding the freontpage order array. Its documented on the site. Adam. Khapare Joshi <khapare77(a)gmail.com> wrote: >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

1 0

Little more help with SNMP/MIB for Canon.
by Matthew Coene 29 May '13

29 May '13

Good Afternoon All, Tom, Looks like you were right... Tried snmpwalking our Canon devices with the HP-LASERJET-COMMON MIB... Looking at the rest of the code, specifically the "jetdirect" bits that you mentioned... <?php if (!$os) { if (strstr($sysDescr, "JETDIRECT")) { $os = "jetdirect"; } else if (strstr($sysDescr, "HP ETHERNET MULTI-ENVIRONMENT")) { $os = "jetdirect"; } else if (strstr($sysObjectId, ".1.3.6.1.4.1.11.1")) { $os = "jetdirect"; } } I can say that the OID string that seems to be the most common to the ID is... (actually there are a few) iso.3.6.1.2.1.43.9.2.1.8.1.1 = STRING: "Canon" I am hoping there might be some way to just adapt the existing framework, to use the the new OID / MIB to report the Canon's specifically instead of generic? Am I even remotely on some sort of right track or way off on some far removed dead end tangent?? Cheers, Matt.

1 0

Re: [Observium] observium Digest, Vol 34, Issue 77
by Wes Kennedy 29 May '13

29 May '13

Im positive I added them. What I ended up having to do was check the box to have them ignored, run the poller manually, uncheck that box then run the poller manually again. Now they're working just fine. It was just extremely odd. -Wes On 5/29/13 1:55 PM, "observium-request(a)observium.org" <observium-request(a)observium.org> wrote: >Send observium mailing list submissions to > observium(a)observium.org > >To subscribe or unsubscribe via the World Wide Web, visit > http://postman.memetic.org/cgi-bin/mailman/listinfo/observium >or, via email, send a message with subject or body 'help' to > observium-request(a)observium.org > >You can reach the person managing the list at > observium-owner(a)observium.org > >When replying, please edit your Subject line so it is more specific >than "Re: Contents of observium digest..." > > >Today's Topics: > > 1. Re: Graphics issue (David Torres) > 2. Polling (Wes Kennedy) > 3. Re: Polling (Nikolay Shopik) > 4. Re: Polling (Adam Armstrong) > 5. SNMP Ubuntu (Alex Vitola) > 6. Re: SNMP Ubuntu (Alex Vitola) > > >---------------------------------------------------------------------- > >Message: 1 >Date: Wed, 29 May 2013 08:16:54 -0600 (CST) >From: David Torres <d.torres(a)ct.co.cr> >To: Observium Network Observation System <observium(a)observium.org> >Subject: Re: [Observium] Graphics issue >Message-ID: <23878121.180651369837014091.JavaMail.root@poseidon> >Content-Type: text/plain; charset=utf-8 > >Thank you for all your soon answers, > >So apparently I will need to move all to a real stuff. Well if so, what >should be the best conditions I have keep for the application running >sufficiently well ?, I mean RAM requirements, disk space, BW and so on. > >Thank you > >Regards > >----- Mensaje original ----- >De: "Adam Armstrong" <adama(a)memetic.org> >Para: "Observium Network Observation System" <observium(a)observium.org> >Enviados: Mi?rcoles, 29 de Mayo 2013 1:21:55 GMT -06:00 Centroam?rica >Asunto: Re: [Observium] Graphics issue > >If you have devices being polled more than 7.5 minutes apart, this will >cause missing data as RRD will not generate an average and will just >insert a null value, leading to the white space. > >We recommend not running Observium in a VM or on shared storage as it >tends to cause these kinds of issues when the host/storage resources is >being used by other applications. > >adam. > >On 2013-05-29 08:06, Bj?rn Rhoads wrote: >> I had the same problem with Observium and ~100 devices with observium >> running in a KVM instance, for me it turned out that disk iops wasn't >> enough, so the pollers didn't finish before the next round started. I >> just got a small SSD and put the Observium virtual machine on that and >> everything worked ok since then. >> >> If your banding is of the same sort you should see that periodically >> you have too many pollers running at the same time. >> >> /Bj?rn R >> >> On 2013-05-29 08:57, Nikolay Shopik wrote: >> This faq section for different issue, has nothing to do with >> whitespace rrdtool. Is this device sits next to observium or in some >> remote location? If this on remote site this is mostly packet lost >> >> On 29.05.2013, at 6:58, David Torres <d.torres(a)ct.co.cr> wrote: >> >> Hi all, >> >> I've been using observium a while ago and since the beginning the tool >> showed me the graphics like I show you in the attached picture. >> >> There are several "white spaces" marked on the time period (X axis), >> where the tools is not showing information. I read the FAQs section and >> I found this advice: >> >> "I'm getting a blank browser window on some pages, but not others", so >> I incremented the PHP memory parameter from 128M to 384M whithin the >> php.ini file >> but this did not work either so I come to you beggin your advice in >> order to resolve this problem. Besides I have 90 network equipments >> monitored with the observium application on a virtual server (6 GB of >> ram and 3 CPUs 3.0 Ghz). >> >> >> >> thank you so much for any help >> >> >> Regards >> <graphs.png> >> _______________________________________________ >> observium mailing list >> observium(a)observium.org >> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium >> _______________________________________________ >> observium mailing list >> observium(a)observium.org >> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium >> _______________________________________________ >> observium mailing list >> observium(a)observium.org >> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium > > >------------------------------ > >Message: 2 >Date: Wed, 29 May 2013 16:35:17 +0000 >From: Wes Kennedy <wkennedy(a)lan.us> >To: "observium(a)observium.org" <observium(a)observium.org> >Subject: [Observium] Polling >Message-ID: <CDCBA684.31EC%wkennedy(a)lan.us> >Content-Type: text/plain; charset="us-ascii" > >I recently added in 25 new routers to be monitored and as of two hours >later, they haven't been polled. They have all been Discovered properly, >but not polled. All of my existing hosts are polling just fine. Has >anyone seen this? > >-Wes >

2 1