- observium - lists.observium.org

Ironware CPU Patch
by Matthias Cramer 30 Apr '12

30 Apr '12

Hi The scale of the discovered CPU's on Ironware based geer is off by factor 100. This patch will fix that. Tested on Brocade XMR4000 Regards Matthias Index: includes/discovery/processors/ironware.inc.php =================================================================== --- includes/discovery/processors/ironware.inc.php (revision 3098) +++ includes/discovery/processors/ironware.inc.php (working copy) @@ -15,11 +15,11 @@ { $usage_oid = ".1.3.6.1.4.1.1991.1.1.2.11.1.1.6." . $index; $usage = $entry['snAgentCpuUtil100thPercent']; - $precision = 100; + $precision = 1; } elseif ($entry['snAgentCpuUtilValue']) { $usage_oid = ".1.3.6.1.4.1.1991.1.1.2.11.1.1.4." . $index; $usage = $entry['snAgentCpuUtilValue']; - $precision = 100; + $precision = 1; } list($slot, $instance, $interval) = explode(".", $index); -- Matthias Cramer / mc322-ripe Senior Network & Security Engineer iway AG Phone +41 43 500 1111 Badenerstrasse 569 Fax +41 44 271 3535 CH-8048 Zürich http://www.iway.ch/ GnuPG 1024D/2D208250 = DBC6 65B6 7083 1029 781E 3959 B62F DF1C 2D20 8250

3 5

Web links
by Mark Murphy 28 Apr '12

28 Apr '12

I have installed the software and I get the login page bur every link I click on after that gives a 404 error. This is a Ubuntu server and the default web is only the monitoring software. I made the updates according to the install instructions.

2 1

No VLANs tab / Cisco 2960S / 0.12.4.3087
by Adam Armstrong 27 Apr '12

27 Apr '12

Aha. This is because we check for 1,2 and two, three for vtp version. This will be fixed later today, i'm just rewriting the discovery module to be more sane. adam. On 2012-04-27 07:15, Damien BASTIE wrote: > One result : > > SNMPv2-SMI::enterprises.9.9.46.1.1.1.0 = INTEGER: 3 > > ________________________________________ > De : Adam Armstrong [adama(a)memetic.org] > Date d'envoi : vendredi 27 avril 2012 00:25 > À : Damien BASTIE > Objet : Re: RE : [Observium] No VLANs tab / Cisco 2960S / 0.12.4.3087 > > and : > > > > .1.3.6.1.4.1.9.9.46.1.1.1.0 > > > > adam. >

1 0

No VLANs tab / Cisco 2960S / 0.12.4.3087
by Damien BASTIE 27 Apr '12

27 Apr '12

Hi, I am testing Observium for the first time I have no VLANs tab for a stack Cisco 2960S, it is normal ? Cisco 2960 S / cat29xxStack / C2960S-UNIVERSALK9-M / Cisco IOS 12.2(55)SE2 (UNIVERSALK9) If i can send any debug no problem ! Regards, Damien

5 10

data missing on BGP IPv4 prefixes graphs
by Dario Calamai 23 Apr '12

23 Apr '12

Hello! there are an "Error Drawing Graph" on BGP IPv4 Prefixes page graphs. (/routing/protocol=bgp/type=all/graph=prefixes_ipv4unicast/view=graphs/) I don't know if this error appears in last update... little question about mail notification: do you have disabled the notification of DOWN of a BGP Peer? it works when peering is going UP or FLAP only! many thanks! dario -- http://www.playnet.it phone +39 055 3226622 - fax +39 055 3226444 Brain Technology s.p.a. - AS 13284 http://www.linkedin.com/in/dariocalamai

2 2

problem with Sentry3-MIB, 3 phase CDU's
by Darrell Budic 21 Apr '12

21 Apr '12

Hi, Running Observium 0.11.12.2749 on SL5 for this issue. Updated to 3069 and still has the same issues. But anyway: Noticed my Sentry CDU's were not picking up current or voltage information, tracked it down to a issue with the MIB. Not sure if it's the MIB in general or my version of the snmp utilities, but the discovery module complains it can't find the MIB when run in debug mode. Turns out my snmpget doesn't like the whitespace before the "Sentry3-MIB DEFINITIONS ::= BEGIN" line. I removed the whitespace and discovered the sensors. Then I noticed I had 3 phase CDU's, and the discovery code for the Sentrys only deals with 1 phase units. Here's the diffs of my hack to enable detection of all 3 phases. I think this will work on single phase units as well, although it will probably take a little longer as it scans for the nonexistent phases. I have no single phase units to test it on, however, so YMMV. 16c16,20 < $infeed_oids = snmp_walk($device, "infeedID.$towers.1", "-Osqn", "Sentry3-MIB"); --- > # These PDUs may have > 1 infeed for 3 phase systems > $phase=1; > while ($phase <= 3) { > > $infeed_oids = snmp_walk($device, "infeedID.$towers.$phase", "-Osqn", "Sentry3-MIB"); 31c35 < $infeed_oid = "1.3.6.1.4.1.1718.3.2.2.1.7." . $towers . ".1"; --- > $infeed_oid = "1.3.6.1.4.1.1718.3.2.2.1.7." . $towers . "." . $phase; 33c37 < $descr_string = snmp_get($device,"infeedID.$towers.$infeed_index", "-Ovq", "Sentry3-MIB"); --- > $descr_string = snmp_get($device,"infeedID.$towers.$phase", "-Ovq", "Sentry3-MIB"); 37,38c41,42 < $high_warn_limit = snmp_get($device,"infeedLoadHighThresh.$towers.$infeed_index", "-Ovq", "Sentry3-MIB"); < $high_limit = snmp_get($device,"infeedCapacity.$towers.$infeed_index", "-Ovq", "Sentry3-MIB"); --- > $high_warn_limit = snmp_get($device,"infeedLoadHighThresh.$towers.$phase", "-Ovq", "Sentry3-MIB"); > $high_limit = snmp_get($device,"infeedCapacity.$towers.$phase", "-Ovq", "Sentry3-MIB"); 42c46 < discover_sensor($valid['sensor'], 'current', $device, $infeed_oid, $towers, 'sentry3', $descr, $divisor, $multiplier, $low_limit, $low_warn_limit, $high_warn_limit, $high_limit, $current); --- > discover_sensor($valid['sensor'], 'current', $device, $infeed_oid, "$towers.$phase" , 'sentry3', $descr, $divisor, $multiplier, $low_limit, $low_warn_limit, $high_warn_limit, $high_limit, $current); 48c52 < $outlet_oids = snmp_walk($device, "outletLoadValue.$towers.1", "-Osqn", "Sentry3-MIB"); --- > $outlet_oids = snmp_walk($device, "outletLoadValue.$towers.$phase", "-Osqn", "Sentry3-MIB"); 61c65 < $outletsuffix = "$towers.$infeed_index.$outlet_index"; --- > $outletsuffix = "$towers.$phase.$outlet_index"; 103a108,109 > $phase++; > } //if($phase) Hope it helps someone :) Thanks for the nice work! -Darrell

1 0

Data collection stopped
by Chris Stone 20 Apr '12

20 Apr '12

Data collection stopped today - running build 3060 - was working fine with that for awhile, so not sure what happened. Running poller manually I get the error: dot3StatsDuplexStatus/usr/bin/snmpbulkwalk -v2c -c public -OQUs -m EtherLike-MIB -M /opt/observium/mibs udp:axis-core-01.axint.net:161 dot3StatsDuplexStatus dot3StatsDuplexStatus = No Such Object available on this agent at this OID SQL[SELECT COUNT(*) FROM `ports` WHERE `device_id` = '1' AND `ifType` = 'adsl'] /usr/bin/snmpbulkwalk -v2c -c public -OQUs -m Q-BRIDGE-MIB -M /opt/observium/mibs udp:axis-core-01.axint.net:161 dot1qPortVlanTable dot1qPortVlanTable = No Such Object available on this agent at this OID /usr/bin/snmpbulkwalk -O QUs -v2c -c public -M /opt/observium/mibs -m Q-BRIDGE-MIB udp:axis-core-01.axint.net:161 dot1qVlanCurrentEgressPorts 2>/dev/null dot1qVlanCurrentEgressPorts = No Such Object available on this agent at this OID /usr/bin/snmpbulkwalk -v2c -c public -OQUs -m Q-BRIDGE-MIB -M /opt/observium/mibs udp:axis-core-01.axint.net:161 dot1dBasePortIfIndex dot1dBasePortIfIndex = No Such Object available on this agent at this OID SQL[SELECT * FROM `ports` WHERE `device_id` = '1'] Port lo(1) PHP Fatal error: Call to a member function get() on a non-object in /opt/observium/includes/polling/ports.inc.php on line 168 Fatal error: Call to a member function get() on a non-object in /opt/observium/includes/polling/ports.inc.php on line 168 Suggestions? Anyone else seeing this? Oh, this is PHP 5.3.10 and Observium build 3060. -- Chris Stone AxisInternet, Inc. www.axint.net

2 5

I/O wait
by Ciro Iriarte 19 Apr '12

19 Apr '12

Hi!, I noticed just today that observium doesn't graph i/o wait time from CPU usage. Only user/nice/system/idle, i/o wait is not provided by SNMP or it's just missing in Observium? Regards, -- Ciro Iriarte http://cyruspy.wordpress.com --

6 17

Remove ports from Web Overview
by Philip Möller 16 Apr '12

16 Apr '12

Hi, I have a performance issue with the Observium Web Interface. The Apache2 webserver needs about 4-5 minutes to load all graphs of some devices. The devices are CISCO NEXUS 5548 switches with CISCO NEXUS 2248 FEX (Fabric Extenders). One 5548 switch will have about 7488 Interfaces when fully populated (48 local interfaces + 48 local virtual interfaces + 24 * 52 + 52 interfaces for the extenders). I do not need to monitor or graph all the edge / access layer ports but only the backbone ports ( 144 ports ). As far as I understand the documentation all ports flagged as disabled are not polled any more and all ports that are deleted are not shown by the Apache2. Unfortunately the poller.php run updates disabled = 0 for all ports found on the device and they are back in the web interface overview. I cannot find the statement in the poller.php debug output but all Ports with deleted = 1 set manually are updated to deleted = 0 after a poller.php run. My question is if there is any way to delete available ports I do not want to have in the web overview or to drain those ports from the web interface. With bet regards, --- Philip Möller Senior Linux Systemadministrator International IT Operations Bigpoint GmbH Drehbahn 47-48 20354 Hamburg | Germany Tel: +49(0)40.88 1413-356 Cell: +49(0)151 27471095 E-Mail: p.moeller(a)bigpoint.net<mailto:p.moeller@bigpoint.net> --- CEO: Heiko Hubertz Business Registration: HRB-Hamburg 94148

3 8

Re: [Observium] Date Picker broken
by Timo Ament 12 Apr '12

12 Apr '12

Hi Folks, same issue as Bruce - setting the Date Picker reverts you back to the start-page. Normally, when opening a graph the url is something like: http://observium/graphs/type=customer_bits/to=1334238800/id=server/from=133… Now when changing the time-range and klicking "update" the URL is beeing transformed into: http://observium/index.php/graphs/type=customer_bits/to=1334238800/id=serve… When i remove the index.php from the url i am getting the desired result. Currently running 3026 Regards Timo Am 12.04.2012 13:00, schrieb observium-request(a)observium.org: > Re: Date Picker broken

4 3