- observium - lists.observium.org

24 Jan '16

Hi, i am a private user of the Observium CE, and I wanted to ask If the MySQL agent will already be fixed in the ne CE release, otherwise I would happily provide a patch for the agent I made just a moment ago when I updated my MariaDB Server to 10 and the innodb query stopped working[0], since it's removed with version 10 and version 5.6.1 of MySQL[1]. If you'd want me to do that, I also would rewrite the whole agent to use mysqli instead of mysql, since it's considered deprecated for some time now[2]. with kind regards Valentin [0]https://mariadb.com/kb/en/mariadb/xtradbinnodb-server-system-variables/#h… [1]https://bugs.mysql.com/bug.php?id=63383 [2]http://php.net/manual/de/migration55.deprecated.php

2 1

Arbor Network's MIBs
by Ryan, Spencer 23 Jan '16

23 Jan '16

Great to see Arbor MIB's getting updated in the SVN checkouts :) I don't monitor any of our own products, but I'm going to talk to the Peakflow guys and see if I can get access to some of the lab gear. We have at least one setup (not sure if it's TMS or SP) that gets about ~8Gb/s of anonymous traffic from a few ISPs. Should give me some useful data in Observium. Keep up the good work guys! *Spencer Ryan* | Senior Systems Administrator | sryan(a)arbor.net *Arbor Networks* +1.734.794.5033 (d) | +1.734.846.2053 (m) www.arbornetworks.com

5 5

Manually setting global temperature thresholds
by Joey Stanford 22 Jan '16

22 Jan '16

Hi, I have a bit of a puzzler and I’m hoping someone can point me in the right direction. I’m monitoring a number of Mikrotik devices. I’m unable to set the temperature thresholds on the device. They all seem to report different thresholds despite the fact that they are all spec’d at -30c to 70c. Instead of going into each device’s settings in Observium and making a custom threshold, is there any way for me to do it in the config file? or some other way using “os eq routeros”? Thanks, Joey

3 8

4 Byte BGP ASN's
by Serghei Golipad 22 Jan '16

22 Jan '16

Guys, Have you made any progress with 4 Byte ASNs? 23456 doesn't look very nice :( Cheers PLEASE CONSIDER THE ENVIRONMENT BEFORE PRINTING THIS EMAIL. Kelway (UK) Ltd is a Company registered in England and Wales. This email is sent for and on behalf of Kelway Holdings Limited. Kelway Holdings Limited is the UK holding company for Kelway (UK) Limited trading as Kelway. The registered office for Kelway Holdings Limited (Company Number 05872067) and its UK group undertakings is 10 Fleet Place, London, EC4M 7RB. VAT Registration 902 1949 39. Confidentiality: This e-mail contains proprietary information, some or all of which may be confidential and/or legally privileged. It is for the intended recipient only. If an addressing or transmission error has misdirected this e-mail, please notify the author by replying to this e-mail and then delete the original. If you are not the intended recipient you may not use, disclose, distribute, copy, print or rely on any information contained in this e-mail. You must not inform any other person other than Kelway or the sender of its existence. For more information about Kelway please visit www.kelway.co.uk

3 6

ASR9k fan failures not detected?
by Robert Williams 22 Jan '16

22 Jan '16

Hi, I've just noticed that fan failures on the ASR 9k are not detected by Observium. In fact, it seems have nothing at all under the 'status' tab? The 6500s have over 20 entries in there for "Fan - Fail" or "PSU - Fail" or "I'm not facing magnetic north - Fail" etc. but the ASR9000 have no status tab whatsoever. Reason for noticing this; is that we have a chassis which currently has a 'Major-Alarm' failure indicator LED on the RSP, due to a failed fan tray. Observium shows the device has being 'OK' and is not seeing either the Major alarm or the fan tray failure: canb-server[151]: %PLATFORM-CANB_SERVER-3-ALARM_INDICATION : Raise alarm from CBC in slot 0/FT0/SP, alarm code CBC_ALRM_FT_LED_RED_FAILED LED Information --------------------------------------------- R/S/I Modules LED Status 0/RSP0/* host Critical-Alarm Off host Major-Alarm On host Minor-Alarm Off host ACO Off host Fail Off FAN0 FAN1 FAN2 FAN3 FAN4 FAN5 FAN6 FAN7 FAN8 FAN9 FAN10 FAN11 FAN12 FAN13 0/FT0/* (Speed) 9900 9900 9870 9780 9870 9780 9780 10080 10080 9870 9990 9900 9810 9990 (fans are now OK again) Is this a case of Cisco not exporting any of this useful (and rather critical) information properly? Would be good to graph fan speeds if they are exported. Or are other people getting these OK and it's just us with the issue? Cheers! Robert Williams Custodian Data Centre Email: Robert(a)CustodianDC.com http://www.CustodianDC.com

6 25

Memory graph not showing as intended.
by Max Krabbenhöft 22 Jan '16

22 Jan '16

Hello, I just set up Observium on CentOS 7 and everything works great! Except for one thing. The Memory and Virtual memory graph on the dashboard for windows/linux servers won't draw. Switches work fine though. When doing a debug I get this error message: SQL[INSERT INTO `mempools-state` (`mempool_id`) VALUES ('5')] RUNTIME[0.00012898s] ERROR[Error in query: (1364) Field 'mempool_polled' doesn't have a default value] I've been looking all over and can't find a resolution for this. This is how it should look [http://i.imgur.com/NMkyhMB.png] This is how it looks: [http://i.imgur.com/VEFSP1K.png] Thank you for your help Best regards Max

5 10

STRICT mode on SQL.
by Dirk-Willem van Gulik 22 Jan '16

22 Jan '16

FWIIW - may be that the STRICT mode is not the only delta. Spotted below error[1] during the initial discovery/poll cycle; and [2] in the update-error log (we are trying to figure out why the port graphs stopped updating). All on a -latest community version as of a few hours ago. Against "MYSQL Server version: 5.6.27 Source distribution” and PHP 5.6.17 (cli). Dw 1: SQL[SELECT *, I.`port_id` AS `port_id` FROM `ports` AS I LEFT JOIN `ports-state` AS S ON I.`port_id` = S.`port_id` WHERE `device_id` = '1'] RUNTIME[0.00157404s] ….. SQL[INSERT INTO `ports` (`device_id`,`ifIndex`) VALUES ('1','29')] RUNTIME[0.00024199s] ERROR[Error in query: (1062) Duplicate entry '1-29' for key 'device_ifIndex'] SQL[SELECT * FROM `ports` WHERE `port_id` = ''] RUNTIME[0.00043392s] Adding: po1(29)() SQL[INSERT INTO `ports` (`device_id`,`ifIndex`) VALUES ('1','30')] RUNTIME[0.00025010s] ERROR[Error in query: (1062) Duplicate entry '1-30' for key 'device_ifIndex’] …. 2: [observium@observium /usr/local/www/observium/logs]$ more update-errors.log [2016/01/22 12:09:22 +0000] discovery.php(23667): ====== Schema update 137 -> 138 ============== [2016/01/22 12:09:22 +0000] discovery.php(23667): Query: ALTER TABLE `sensors` ADD `entPhysicalIndex_measured` varchar(16) COLLATE utf8_unicode_ci DEFAULT NULL AFTER `sensor_custom_limit`; [2016/01/22 12:09:22 +0000] discovery.php(23667): Error: (1060) Duplicate column name 'entPhysicalIndex_measured' [2016/01/22 12:09:22 +0000] discovery.php(23667): Query: ALTER TABLE `sensors` ADD `measured_class` VARCHAR( 32 ) NULL DEFAULT NULL AFTER `entPhysicalIndex_measured`; [2016/01/22 12:09:22 +0000] discovery.php(23667): Error: (1060) Duplicate column name 'measured_class' [2016/01/22 12:09:22 +0000] discovery.php(23667): Query: ALTER TABLE `sensors` ADD `measured_entity` VARCHAR( 32 ) NULL DEFAULT NULL AFTER `measured_class`; [2016/01/22 12:09:22 +0000] discovery.php(23667): Error: (1060) Duplicate column name ‘measured_entity'

2 1

ModbusTCP via SNMP Gateway. Possible to monitor in Observium?
by Henrik Cednert (Filmlance) 22 Jan '16

22 Jan '16

Hi I have a schneider tsxetg100 ethernet gateway that has a PowerLogic PM3250 attached to it. I can successfully monitor this in APC StruxureWare Data Center expert but trying to get away from that solution. Observium does find and see the tsxetg100 as a generic device but I don't see the PowerLogic attached to it. Is it possible to monitor a Modbus TCP device like this through the SNMP gateway? The tsxetg100: http://www.schneider-electric.com/download/ww/en/details/25779671-TSXETG100… The Powerlogic: http://www.schneider-electric.com/en/product-range/61274-powerlogic-pm3000-… Cheers and thanks -- Henrik Cednert cto | td | compositor Filmlance International Direct + 46 (0)704 71 89 54

4 15

Making older pre rfc 1759 printers happy
by Dirk-Willem van Gulik 22 Jan '16

22 Jan '16

Minor diff to make things work with an older type of printer. Dw --- toner.inc.php.orig 2016-01-21 19:49:49.528405666 +0100 +++ toner.inc.php 2016-01-21 20:06:36.382527604 +0100 @@ -39,10 +39,11 @@ $index = $split_oid[count($split_oid)-1]; if (is_numeric($role)) { - $type_oid = ".1.3.6.1.2.1.43.11.1.1.5.1.$index"; - $descr_oid = ".1.3.6.1.2.1.43.11.1.1.6.1.$index"; - $capacity_oid = ".1.3.6.1.2.1.43.11.1.1.8.1.$index"; - $toner_oid = ".1.3.6.1.2.1.43.11.1.1.9.1.$index"; + $class_oid = ".1.3.6.1.2.1.43.11.1.1.4.1.$index"; // prtMarkerSuppliesClass (supplyThatIsConsumed(3), receptacleThatIsFilled(4)) + $type_oid = ".1.3.6.1.2.1.43.11.1.1.5.1.$index"; // prtMarkerSuppliesType + $descr_oid = ".1.3.6.1.2.1.43.11.1.1.6.1.$index"; // prtMarkerSuppliesDescription + $capacity_oid = ".1.3.6.1.2.1.43.11.1.1.8.1.$index"; // prtMarkerSuppliesMaxCapacity + $toner_oid = ".1.3.6.1.2.1.43.11.1.1.9.1.$index"; // prtMarkerSuppliesLevel $classtype = snmp_get($device, $class_oid, "-Oqv"); $resourcetype = snmp_get($device, $type_oid, "-Oqv"); @@ -52,8 +53,10 @@ { switch ($resourcetype) { - case 3: - case 21: + case 3: // Toner + case 5: // Ink or particulate + case 6: // InkCardridge + case 21: // Toner Cardridge /* .1.3.6.1.2.1.43.11.1.1.8 prtMarkerSuppliesMaxCapacity * The value (-1) means other and specifically indicates that the sub-unit places * no restrictions on this parameter. The value (-2) means unknown. @@ -133,8 +136,11 @@ set_dev_attrib($device, "transferroller_cap_oid", $capacity_oid); } break; - case 4: - if (stristr($descr, 'waste') !== FALSE) + case 4: // Waste Toner + case 8: // Waste Ink or toner + case 24: // Waste liquid or Ink + if (stristr($descr, 'waste') !== FALSE || stristr($descr, 'absorber') == FALSE || $classtype == 3) { echo(" WasteTonerBox"); set_dev_attrib($device, "wastebox_oid", $toner_oid);

3 6

Proxmox
by Cody Cook 22 Jan '16

22 Jan '16

Hi everyone, Two things about Proxmox and my usage with Observium: Upgraded all of my Proxmox boxes except one to 4.0 from 3.4. I use zfs for all the file systems. Since switching to Proxmox 4.0, appears that snmp no longer includes the zfs volumes automatically in the SNMP reply. :( Does anyone have some suggestions on automatically showing their zfs vols in SNMP? I read a manpage for snmpd.conf but only recognized the ignoreDisk screen to manually add. Some other site said begemotSnmpdModulePath."hostres" = /usr/lib/snmp_hostres.so would be useful if I was on FreeBSD, but I am using Debian :P Another question I had is that I was reading that there is the ability to query Proxmox for VMs. Neither my 3.4 or my 4.0 box show VMs. My enable_libvirt is 1 and I use lxc+ssh and openvz+ssh in the libvirt protocols option in the config (lxc for 4.0, openvz for 3.4). I don't see anything VM. I do have tab for VMs for my vmware esxi server though. The public key from Observium is in the authorized_keys file for each of my root users on each proxmox box. What am I missing here? net-snmp version: 5.7.2.1+dfsg-1 Cody

2 2