- observium - lists.observium.org

graph NVIDIA GPU parameters
by Mihai - Cristian Satmarean 17 Apr '14

17 Apr '14

Dear all, sorry if this was already addressed, is it somehow possible to graph the Nvidia GPU parameters in observium? Load, memory, temp etc. Thank you! Mihai

3 6

1 rogue host on map in the mid Atlantic
by Ryan Prihoda 16 Apr '14

16 Apr '14

Hey all, I recently updated to the new community edition and once everything was up again, I have 1 rogue host showing in the mid Atlantic on the map. None of my hosts use the override "syslocation" function. The number of devices in "All Locations" is equal to devices my syslocation. Here's the kicker, when I click on the rogue host, it takes me to my devices page for my syslocation. Is there anything I can do to figure out what device is causing this issue ? Thanks, R. Prihoda

2 4

IP address weirdness
by Sander Steffann 16 Apr '14

16 Apr '14

Hi, This morning at 6:30 the IPv4 addresses of my servers suddenly started changing. I see a lot of messages like: IPv4 changed: 83.247.10.20/27 -> 83.247.10.20/3 IPv4 changed: 83.247.10.21/27 -> 83.247.10.21/3 IPv4 changed: 83.247.10.22/27 -> 83.247.10.22/3 IPv4 changed: 94.142.242.194/29 -> 94.142.242.194/5 IPv4 changed: 94.142.242.195/29 -> 94.142.242.195/5 IPv4 changed: 94.142.242.213/29 -> 94.142.242.213/5 IPv4 changed: 94.142.242.214/28 -> 94.142.242.214/4 IPv4 changed: 94.142.242.215/28 -> 94.142.242.215/4 IPv4 removed: 127.0.0.1/8 Besides showing me that I have some misconfigured subnet masks in the 94.142.242.208/28 subnet, it also seems to have some consistency: - A /27 turns into a /3 - A /28 turns into a /4 - A /29 turns into a /5 My VMware ESXi machines also suddenly show: IPv4 added: 83.247.10.11/32 IPv4 added: 83.247.10.12/32 IPv4 added: 83.247.10.13/32 I also noticed some broadcast addresses being added: IPv4 added: 83.247.10.31/27 (this is an Ubuntu 14.04 server with 83.247.10.29/27) IPv4 added: 83.247.10.31/27 (this is is a CentOS 5.10 server with 83.247.10.7/27) IPv4 added: 83.247.10.47/28 (this is a vcenter applicance) IPv4 added: 83.247.10.47/28 (this is a Dell DRAC card) IPv4 added: 127.255.255.255/8 (this is a vcenter applicance) So, subnet sizes went weird on some systems, 127.0.0.1 disappeared from some systems, some ethernet adapters now show both the real IP address and the broadcast address on some systems. But different mixes of results across different platforms. Might this be related to r5317? Cheers, Sander

2 2

dot1dbaseportifindex on Cisco not listing UP interfaces?
by Ulrik Ivers 16 Apr '14

16 Apr '14

Hi, I see a strange behavior on our Cisco-switches (different models) when listing the interfaces via SNMP using the mib BRIDGE-MIB. When walking dot1dbaseportifindex it excludes interfaces that are UP. Only interfaces that are DOWN or DISABLED are returned. Is this something that is well known? Seems to me that is a bug in ios... It breaks neighbor discovery via LLDP (lldp-mib.inc.php). Examples: !!!!!!!!!!!!!!!!!!!!!!!!!!! Cisco Switch - ports 1-7 are UP, the rest DOWN /usr/bin/snmpbulkwalk -v2c -c 'XXXX' -OQUs -m BRIDGE-MIB -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp 'udp':'test-host' dot1dBasePortIfIndex dot1dBasePortIfIndex.8 = 10108 dot1dBasePortIfIndex.9 = 10109 dot1dBasePortIfIndex.10 = 10110 dot1dBasePortIfIndex.11 = 10111 dot1dBasePortIfIndex.12 = 10112 dot1dBasePortIfIndex.13 = 10113 dot1dBasePortIfIndex.14 = 10114 dot1dBasePortIfIndex.15 = 10115 dot1dBasePortIfIndex.16 = 10116 dot1dBasePortIfIndex.17 = 10117 dot1dBasePortIfIndex.18 = 10118 dot1dBasePortIfIndex.19 = 10119 dot1dBasePortIfIndex.20 = 10120 dot1dBasePortIfIndex.21 = 10121 dot1dBasePortIfIndex.22 = 10122 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!! HP 10 port switch , all interfaces are always listed /usr/bin/snmpbulkwalk -v2c -c 'XXXX' -OQUs -m BRIDGE-MIB -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp 'udp':'test-host' dot1dBasePortIfIndex dot1dBasePortIfIndex.1 = 1 dot1dBasePortIfIndex.2 = 2 dot1dBasePortIfIndex.3 = 3 dot1dBasePortIfIndex.4 = 4 dot1dBasePortIfIndex.5 = 5 dot1dBasePortIfIndex.6 = 6 dot1dBasePortIfIndex.7 = 7 dot1dBasePortIfIndex.8 = 8 dot1dBasePortIfIndex.9 = 9 dot1dBasePortIfIndex.10 = 10 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Regards, /Ulrik Ivers Ulrik Ivers -- Excanto AB Sveavägen 63 | 113 59 Stockholm Mobil: 0709-90 95 06 Växel: 08-545 726 00 Web: www.excanto.se<http://www.excanto.se/> [cid:image001.jpg@01CF595E.6CC06620]

1 0

PPP links for Mikrotik
by David Peall 15 Apr '14

15 Apr '14

Hi The link names for ppp interfaces on the mikrotik routers are wrapped in <> i.e. <pptp-david> this seems to be causing some issues on the website. For example: <span class="entity-title"> <a href="device/device=28/tab=port/port=1462/" class="entity-popup " data-eid="1462" data-etype="port" data-hasqtip="1"><pptp-david></pptp-david></a> </span> Regards — David Peall

2 3

Juniper inventory
by Zalewski, Dominik 15 Apr '14

15 Apr '14

Hi, Are there any plans or development in progress of Juniper inventory module? Thanks Dominik

4 8

Alerting on high 'used' RAM, not counting 'cached'
by Robert Williams 15 Apr '14

15 Apr '14

Hi, Having my first experiment with the new alerting features - very nice! There is one thing I can’t seem to get quite right which is how to alert for Linux system when the ‘used’ RAM is greater than, say, 90% - which is best explained with an example of a device which looks like this: [cid:image001.png@01CF5826.67DDE2B0] Which shows that ‘used’ is 15.1GB (48%). However, for the purposes of alerting, if you use the mempool_perc gt 80 it alarms as being 98%. So I can’t specify ‘used’ as a pool because the only pools are “Swap Space”, “Virtual Memory”, “Physical Memory” etc. There isn’t one which is specifically the ‘used’ RAM element. So, is there a way around this or something I can tweak to get the desired result? Or what is everyone else doing to get alerting for linux boxes like this working for high RAM? Cheers guys! Robert Williams Custodian Data Centre Email: Robert(a)CustodianDC.com http://www.CustodianDC.com

3 2

Support for Huawei MA5600T GPON OLT
by Adam Wołk-Jankowski 14 Apr '14

14 Apr '14

Hello, I’m looking for NMS with support for full query and monitoring of MA5600T series Huawei GPON OLT, but with no luck. Huawei NMS – iManager U2000 costs huge amount of money and it not completely suit to my needs. We use Observium for few months with Cisco, Extreme Networks and D-Link devices and it seems to be ideal solution, but with MA5600T there works only basic things like traffic on Ethernet ports. I have full MIB library for this device but I’m not a programmer so cannot make support by myself. I have some devices in lab, so can share access and also would like to help if anyone with skills would like to work on this. -- Adam Jankowski a.jankowski(a)systel.pl

2 2

Device names on front page Google map?
by Ulrik Ivers 14 Apr '14

14 Apr '14

Hi, I'm new to Observium, so please be tolerant if I ask a stupid question. On the Google map on the front page - is it possible to have the device names displayed? Either "always" or when hovering on the green place marker. Now when I hover I get a label stating "1 hosts", "2 hosts" etc depending on the number of devices that have identical location. I've looked through the documentation and searched old discussion archives but haven't been able to find anything regarding this. Regards, /Ulrik Ivers

3 2

Re: [Observium] 6k Clock 2 status alarms
by Moerman, Maarten 14 Apr '14

14 Apr '14

I’m suffering from the same issue since last week or so. I’ve filed http://jira.observium.org/browse/OBSERVIUM-795 for this to get this solved/fixed. I didn’t exactly know what it was related to, but thanks for figuring that out! Thx, Maarten -- Maarten Moerman | Mgr, Network Engineering | eBay Classifieds | +31-655122247 | mmoerman(a)ebay.com From: Robert Williams <Robert(a)CustodianDC.com<mailto:Robert@CustodianDC.com>> Reply-To: Observium Network Observation System <observium(a)observium.org<mailto:observium@observium.org>> Date: Saturday, April 12, 2014 at 1:21 PM To: "Observium Network Observation System (observium(a)observium.org<mailto:observium@observium.org>)" <observium(a)observium.org<mailto:observium@observium.org>> Subject: [Observium] 6k Clock 2 status alarms Hi, Just a minor bug (or feature tweak, depending on your perspective!) when someone is next in the relevant area: [cid:image001.png@01CF5649.04D8EB80] The clock sensor number 2 on all 6k Catalyst chassis should probably not ‘alarm’ by default when it’s not in use. There will always be only one of the two clocks in use, so some kind of logic tweak is needed to treat this differently otherwise you always have a load of status warnings for the category of “State” under “Health”. Unless you go through them all and disable them, but then you won’t get an event when they do become active (which is usually a bad thing as something else has failed). I’ve seen it show up as these two names depending what IOS it’s running: OSR-7600 Clock FRU 2 - clock 2 clock-inuse Sensor<http://observium.cdc.local/graphs/id=579946/type=sensor_state> OSR-7600 Clock FRU 2 In Using Sensor - Clock 2 In Using Sensor<http://observium.cdc.local/graphs/id=580144/type=sensor_state> Actually, since there is also a value for the ‘ok-ness’ of the sensors it is probably worth totally skipping the ‘in-use’ element for the default alarming purposes, imho. If you need more info from me then let me know, cheers! Robert Williams Custodian Data Centre Email: Robert(a)CustodianDC.com<mailto:Robert@CustodianDC.com> http://www.CustodianDC.com

1 0

device status boxes on the front page show disabled devices
by Henrik Lund Kramshøj 14 Apr '14

14 Apr '14

Hi We have a problem with the front page, the red alert boxes are shown for devices that are disabled. We use the config.php $config['frontpage']['order'] = array('status_summary', 'device_status_boxes', 'device_status', 'overall_traffic' ); and I guess its the device_status_boxes is the one. Did something change, or do we need to change something? Devices shown (some of them) has check mark in: "Disable Disables polling and discovery. Device ignore Device ignore. " So they shouldnt show up, didnt sometime before - but unknown when they stopped working correctly. Version, subscription, did a svn up today. man apr 14 09:14:08 CEST 2014 Best regards Henrik -- Henrik Lund Kramshøj, Follower of the Great Way of Unix internet samurai cand.scient CISSP hlk(a)kramse.org hlk(a)solidonetworks.com +45 2026 6000 http://solidonetworks.com/ Network Security is a business enabler

3 3

6k Clock 2 status alarms
by Robert Williams 12 Apr '14

12 Apr '14

Hi, Just a minor bug (or feature tweak, depending on your perspective!) when someone is next in the relevant area: [cid:image001.png@01CF5649.04D8EB80] The clock sensor number 2 on all 6k Catalyst chassis should probably not ‘alarm’ by default when it’s not in use. There will always be only one of the two clocks in use, so some kind of logic tweak is needed to treat this differently otherwise you always have a load of status warnings for the category of “State” under “Health”. Unless you go through them all and disable them, but then you won’t get an event when they do become active (which is usually a bad thing as something else has failed). I’ve seen it show up as these two names depending what IOS it’s running: OSR-7600 Clock FRU 2 - clock 2 clock-inuse Sensor<http://observium.cdc.local/graphs/id=579946/type=sensor_state> OSR-7600 Clock FRU 2 In Using Sensor - Clock 2 In Using Sensor<http://observium.cdc.local/graphs/id=580144/type=sensor_state> Actually, since there is also a value for the ‘ok-ness’ of the sensors it is probably worth totally skipping the ‘in-use’ element for the default alarming purposes, imho. If you need more info from me then let me know, cheers! Robert Williams Custodian Data Centre Email: Robert(a)CustodianDC.com http://www.CustodianDC.com

1 0

How best to tell Observium that VMware is no longer unix [SEC=UNCLASSIFIED]
by Peter.Hine＠familycourt.gov.au 11 Apr '14

11 Apr '14

Back in the bad old days, esx was run on a version of RedHat Linux (RHEL). Now they run their own kernel. During discovery, i noticed that the unix agent was trying to be contacted and also has (unix). Reasonably sure VMware won't support it. Your native VM detection and polling is sufficient anyway. adlvm03 6 vmware (unix) Module [ junose-atm-vp ] excluded for device. Module [ cisco-ipsec-flow-monitor ] excluded for device. Module [ cisco-remote-access-monitor ] excluded for device. Module [ cisco-cef ] excluded for device. Module [ cisco-sla ] excluded for device. Module [ arista-software-ip-forwarding ] excluded for device. Module [ cipsec-tunnels ] excluded for device. Module [ cisco-ace-loadbalancer ] excluded for device. Module [ cisco-ace-serverfarms ] excluded for device. Module [ cisco-cbqos ] excluded for device. Module [ cisco-eigrp ] excluded for device. Module [ netscaler-vsvr ] excluded for device. Module [ aruba-controller ] excluded for device. Module [ cisco-vpdn ] disabled globally. Module [ wmi ] disabled globally. Observium UNIX Agent: Connection to UNIX agent on tcp://adlvm03:36602 failed. ERROR: 110 Connection timed out Connection to UNIX agent on tcp://adlvm03:6556 failed. ERROR: 110 Connection timed out Sensors: Module time: 20.0376s Uptime: 129 days, 2h 18m 13s Module time: 0.308s VMware VM: thanks Peter Hine Senior Technical Support Engineer (Servers) FCoA ITS peter.hine(a)familycourt.gov.au ********************************************************************** The information contained in this e-mail (including any attachments) is for the exclusive use of the addressee. If you are not the intended recipient please notify the sender immediately and delete this e-mail. It is noted that legal privilege is not waived because you have read this e-mail. **********************************************************************

2 1

Frontpage alert hostnames
by Spencer Gaw 11 Apr '14

11 Apr '14

Hey all, Has anyone noticed that some devices display the subdomain in their hostname on frontpage alerts and others do not? Example: Here are the hostnames configured on each device: /ring-rmcit.ftc> system show host-name// // //+------------------------------ HOST NAME -------------------------------+// //| Oper | ring-rmcit.ftc |// //| User | ring-rmcit.ftc |// //| DHCP | |// //+------+-----------------------------------------------------------------+// // //sw01.cos> system show host-name// // //+------------------------------ HOST NAME -------------------------------+// //| Oper | sw01.cos |// //| User | sw01.cos |// //| DHCP | |// //+------+-----------------------------------------------------------------+/ /redacted(a)access03.den> show version | match hostname// //Hostname: access03.den /Any ideas here? / /Regards, SG

4 4

Enhancement Request: Latest Reported Operating System
by techuser6＠gmail.com 11 Apr '14

11 Apr '14

It would be nice if there was another column next to "Operating System" for Network devices that had the latest reported IOS from Cisco/Dell. That way one could easily see which devices need to be upgraded to the latest IOS to ensure that security and bug fixes are addressed.

4 4

Device Ignore
by Marco Spicuglia 10 Apr '14

10 Apr '14

Hi there, running last stable version. If I disable/ignore a decommissioned device, then it still shows up as "down" in the frontpage device alerts module, while the alarm should disappear instead. Kind regards, Marco Met vriendelijke groet/ Kind regards, Marco Spicuglia Network engineer - Programmer ReasonNet B.V. Gyroscoopweg 134 - 140 1042AZ Amsterdam W. www.reasonnet.com E. Marco.Spicuglia(a)reasonnet.com T. F. ReasonNet zoekt datacenter support engineers. Ga voor deze en andere interessante vacatures binnen ReasonNet naar http://www.reasonnet.com/nieuws ReasonNet, shall not be liable for damages resulting from the use of electronic means of communication, including -but not limited to- damages resulting from failure or delay in delivery of electronic communications, interception or manipulation of electronic communications by third parties or by computer programs used for electronic communications and transmission of viruses and other malicious code.

1 0

Upgrades blow up my frontpage graphs
by Scott Brawner 10 Apr '14

10 Apr '14

It seems like every time there's a major upgrade with a sql script, my front page graphs go completely apeshit. Should look like this (in the first case) Best regards, Scott Brawner IS Manager - Infrastructure Kratos Defense and Security Solutions, Inc. o. 843-308-2364 m. 843-814-3912 Description: cid:image002.jpg@01CBD278.0CADB410 For every complex problem, there is a solution that is simple, neat, and wrong. -- H. L. Mencken

2 5

Trapeze/Juniper wireless controller
by Adam Armstrong 10 Apr '14

10 Apr '14

Hi All, Does anyone have a Trapeze/Juniper wireless controller they'd be willing to give us SNMP read access to so we can add support for it? (similarly for other wireless controllers, but initially we're looking for Trapeze/Juniper :) Thanks, adam.

1 0

Turning off some alerts
by Luca Salvatore 10 Apr '14

10 Apr '14

Hi Been running observium for a while but with alerting turned off. I turned it on recently and got spammed with so many alerts I had to turn it off again. I can't really find how I can tune the alert setting anywhere. Can this be done via the web or do I need to go to the config files? I'm on version 0.13.9.4493, basically i only want alerts for specific things, not every single event. If i go to the alerts section there is no option to configure any thing, there is no 'add alert' button as described in this link http://www.observium.org/wiki/Creating_an_Alert_Checker Is the alerting functions only available in the current version? thanks

2 1

Deleted status
by Teixeira, Mike 10 Apr '14

10 Apr '14

Hello, When we have ports in deleted status, is there an automagic process to purge those ports? Thanks, Mike

2 5

ifInOctets_perc / ifOutOctets_perc
by Ryan, Spencer J. 10 Apr '14

10 Apr '14

Hi All, Anyone know if these percentages are based on physical speed of the port or of they use the speed set via the "bandwidth" statement? ________________________________ Spencer Ryan | Systems Engineer Miller Canfield<http://www.millercanfield.com> 150 West Jefferson, Suite 2500 Detroit, Michigan 48226 (USA) T +1.313.496.7547 | F +1.313.496.7699 | Mobile +1.734.846.2053 ryan(a)millercanfield.com<mailto:ryan@millercanfield.com> ________________________________ _______________________________________________________________ NOTICE TO PERSONS SUBJECT TO UNITED STATES TAXATION (MCPS) DISCLOSURE UNDER TREASURY CIRCULAR 230: The United States Federal tax advice, if any, contained in this document and its attachments may not be used or referred to in the promoting, marketing or recommending of any entity, investment plan or arrangement, nor is such advice intended or written to be used, and may not be used, by a taxpayer for the purpose of avoiding Federal tax penalties. _______________________________________________________________

3 2

LDAPS Authentication
by Ellery Weber 10 Apr '14

10 Apr '14

Hey guys, I'm trying to get LDAP (through AD) authentication setup on the subscription version of Observium, and am running into some difficulties. My first question - does Observium only do a simple bind, or can it connect using SASL? We do not currently allow simple binds in our domain. Thanks for any assistance you can provide, [SELCO Community Credit Union] Ellery Weber, CCENT Network Administrator SELCO Community Credit Union Eugene - Downtown Corporate Headquarters Phone/Fax: 541-686-5349 | Cell: 541-357-9505 | eweber(a)selco.org<mailto:eweber@selco.org> selco.org<https://www.selco.org/> | facebook<http://www.facebook.com/SELCOCCU> | twitter<http://twitter.com/selcoccu> This message (including any attachments) is intended only for the use of a specific individual or entity or specific individuals or entities and may contain information that is non-public, proprietary, privileged, confidential, and exempt from disclosure under applicable law or may constitute as attorney work product. If you are not the intended recipient, you are hereby notified that any use, dissemination, distribution, or copying of this communication is strictly prohibited and may represent a tortious interference with a confidential business relationship(s). If you have received this communication in error, please notify us immediately by telephone and (i) permanently destroy this message if a facsimile; (ii) permanently delete this message immediately if this is an electronic communication; permanently destroy any electronic or printed copies of this electronic communication.

2 2

rancid multiple config/diff versions within observium
by Chip Pleasants 09 Apr '14

09 Apr '14

I have rancid working today within Observium, although I do not see multiple version ro config/diffs of the configurations like saw some time ago. I did migrate servers therefore its possible a configuration or permissions issue on the new server. On my previous server a couple months ago I was able to look at past diffs within observium. Any assistance is appreciated. Thanks, Chip

1 0

Enhancement Request: Export Inventory to CSV
by techuser6＠gmail.com 09 Apr '14

09 Apr '14

It would be really nice if there was a way to export the Inventory to a CSV. This would greatly help with renewing maintenance and SmartNet Contracts for Cisco Devices.

1 0

Multiple Pollers Guidance
by Chip Pleasants 09 Apr '14

09 Apr '14

When there are multiple pollers via NFS how do they pick the devices to poll? Basically how to they not step all over each other polling the same nodes? I'm seeing snmp and ping alerts like 15 or so every hour come in that didn't come in when it was a single server solution. It does seem to be the same 20 or so devices. These particular devices do typically take around 60 sec to poll. Looking at devices that generated alerts their cpu, snmp response time, and ping time was all over the place. Meaning cpu doubled (10% to 20%), ping times went up to 200ms from 10ms, and snmp response time average when from 50ms to 556ms. I'm wondering if I was polling these devices multiple times within 5 minutes? Would this be related to NFS IO issues? I reverted back to a single solution for now. Any assistance is greatly appreciated. -Chip Config Sniplet $config['alerts']['email']['enable'] = TRUE; $config['poller-wrapper']['alerter'] = TRUE; $config['snmp']['timeout'] = 6; $config['snmp']['retries'] = 3; $config['snmp']['max-rep'] = 10; $config['fping'] = "/usr/sbin/fping -t2000";

3 5