- observium - lists.observium.org

Geolocation through proxy
by Christian Hügel 15 Oct '13

15 Oct '13

Hi, is there a way to use the geolocation feature through a proxy with authentication? Our network policy doesn´t admint anonymous access through the proxy. Thx. Chris

2 2

Empty ADSL and SLA rrd
by David George 14 Oct '13

14 Oct '13

Morning all, I've just started trailing observium, and it's looking like a great system. Router health and interface counters seems to be logged and graphed correctly, so it's starting to look like a viable alternative to some of the larger closed source systems. I especially like the simplicity of the design which should allow for some additional sensor development down the that I'll submit back. So kudos on the system, I quite like it. Now - onto the issue, and I'm assuming given the size of the user-base this is likely to be configuration, but I've read and re-read the related doco and I'm unable to find anything wrong. I've added a Cisco 877 device, port counters and health are being logged and graphed correctly, but the ip-sla statistics and adsl line statistics are not. The header line includes dsl stats, and those stats appear to update according to the polling schedule, but the graphs do not. Also, the file timestamp is NOT updating on the adsl RRD file. it is however updating on the sla rrd file. rrdtool dump reveals that the rrd file contains no rows that have a value. config file contains $config['enable_ports_adsl'] = 1; # Enable ADSL-LINE-MIB $config['enable_sla'] = 1; Observium CE 0.13.10.4585 Apache 2.2.22 (Ubuntu) PHP 5.3.10-1ubuntu3.8 MySQL 5.5.29-0ubuntu0.12.04.1 RRDtool 1.4.7 [cid:image001.png@01CEC983.247047C0] [cid:image002.png@01CEC983.247047C0]

1 0

Did the config setting mibdir get removed?
by Jeremy Custenborder 14 Oct '13

14 Oct '13

Guys, We used mibdir heavily in our environment. After updating I see everything is going back to the default /opt/observium/mibs. What can we do to correct this? Has the setting changed? J

2 3

Error during polling: ERROR: unknown option 's'
by Jeremy Custenborder 14 Oct '13

14 Oct '13

Have any of you run into this before? I'm running centos 6.4. ERROR: unknown option 's' ERROR: unknown option 's' ERROR: unknown option 's' Module [ unix-agent ] disabled globally. Using SNMP Agent sysUpTime (4488767 seconds) ERROR: unknown option 's' Uptime: 51 days, 22h 52m 47s Module time: 5.0179s Hardware: cisco2811 Version: 15.1(4)M6 Features: ADVSECURITYK9 Serial: Asset: Module time: 1.0249sIPMI: Module time: 0.0087sSensors: Checking (snmp) temperature chassis ERROR: unknown option 's' 26 C Checking alerts Module time: 1.0116sProcessor Processor 1 ERROR: unknown option 's' 1% Checking alerts Module time: 1.025sMempool Processor: 25.95% ERROR: unknown option 's' Checking alerts Mempool I/O: 60.08% ERROR: unknown option 's' Checking alerts Mempool 2811 chassis - Processor: 25.95% ERROR: unknown option 's' Checking alerts Mempool 2811 chassis - I/O: 60.08% ERROR: unknown option 's' Checking alerts Module time: 8.8415sModule time: 0.0025sPolling Netstats: IPERROR: unknown option 's' TCPERROR: unknown option 's' UDPERROR: unknown option 's' ICMPERROR: unknown option 's' SNMPERROR: unknown option 's'

2 2

CE: Broken detailed Collectd charts
by Frank Ortmann 14 Oct '13

14 Oct '13

Hi I installed the community edition at the time of the split and have come across an issue with the detailed charts. When I click on the Collectd tab I see the various charts as expected. See the attached screenshot. However, when I then click on one of the charts to see more detail, I get the broken image icon. I would expect to see something like the following: I am doing something wrong, or is this a bug? Thanks, Frank

1 0

Test of List Posting
by Ron Wingfield 14 Oct '13

14 Oct '13

4 4

poller performance
by F.Reenders＠utwente.nl 14 Oct '13

14 Oct '13

Hi, I'm having problems with the performance of the poller. We have 200 switches and almost 16000 ports included and the poller is having trouble getting them checked in 5 minutes. With 20 threads it's working with a load of 15. When I add more threads it takes longer to get the data from switches. With less threads it also doesn't check all switches in 5 minutes. I'm using new fast hardware. I've implemented all the performance tuning tips on the observium site. I also disabled the checking of fbd-table, arp-table and mac-accounting. When I debug the poller the ports check takes the longest. About 90 % of the total time. 1 switch takes about 8 to 10 seconds now. Is there a way to speed it up? Maybe I can extend the check cycle time to 10 minutes? Distributed poller instances? Regards, Frederik Reenders | ICTS Operation Centre | University of Twente | ICT Service Centre | ICTS Server Operation | P.O.Box 217, 7500 AE Enschede The Netherlands | Drienerlolaan 5, 7522 NB Enschede | Campus building: Spiegel, room 416 | T: +31 53 489 2653/6723 | f.reenders(a)utwente.nl | www.utwente.nl/icts/en/

8 25

Brocade FastIron SX 1600, IronWare Version 07.0.01cT3e1
by Danilo Marques de Gouveia 13 Oct '13

13 Oct '13

Hello guys, I'm testing Observium in the company that I work, it's working great, however we realize that Observium is not getting the processors and memory utilization from ours FastIron SX family, which includes SX 800 and SX 1600. The Brocade MLX family that we use are working fine and we are able to get proc and memory, I guess that it's a MIB problem but I couldn't find it on google. Does anyone has this switches been monitored by observium? Below is the version info from observium: *Observium CE*0.13.10.4585 *Apache*2.2.16 (Debian) *PHP*5.3.3-7+squeeze17 * MySQL*5.1.66-0+squeeze1 *RRDtool*1.4.3 Thanks in advance, -- Danilo Marques de Gouveia

4 6

monitoring SFP interfaces on Juniper EX3300
by Rodrigo Augusto 13 Oct '13

13 Oct '13

Hi folks! Does anyone have any idea what I have to do to monitoring my 10GB interfaces on juniper ex3300?! On my MX observium monitoring normally ut on my ex3300 cant see DBM page. What I have to do to solve this?! Rodrigo Augusto Gestor de T.I. Grupo Connectoway http://www.connectoway.com.br <http://www.connectoway.com.br/> http://www.1telecom.com.br <http://www.1telecom.com.br/> * rodrigo(a)connectoway.com.br <mailto:rodrigo@connectoway.com.br> ( (81) 3366-7376 ( (81) 8184-3646 ( INOC-DBA 52965*100

1 0

Re: [Observium] [Observium-dev] r4545 - in observer/branches/devel/html/pages/device: . overview
by Paul Gear 12 Oct '13

12 Oct '13

On 10/10/2013 06:29 AM, Adam Armstrong wrote: > On 2013-09-28 03:12, Paul Gear wrote: >> On 09/26/2013 03:25 PM, observium-commits(a)observium.org wrote: >> Author: adama >> Date: 2013-09-26 07:25:19 +0200 (Thu, 26 Sep 2013) >> New Revision: 4545 >> >> Added: >> observer/branches/devel/html/pages/device/overview/processors-unix.inc.php >> >> Modified: >> observer/branches/devel/html/pages/device/overview.inc.php >> Log: >> more useful processor overview for unix hosts >> >> IMO, ucd_cpu (a.k.a. detailed processor) would be a better processor >> overview for Unix hosts; at least it includes a breakdown of the >> usage. Even better would be finding one that includes iowait time >> (I'm not sure if net-snmp exposes this). > > We don't read observium-dev, it's just svn spam. you need to > cross-post these to observium :> > > We could probably pull io-wait from the agent. P.S. It also should be called utilisation, not load. Load & utilisation are two separate figures and mixing their names is confusing. Paul

1 0

Devices not visible from drop down for user level less than 5
by Andrej K 12 Oct '13

12 Oct '13

Hi to all of you, I have found following problem,Normal user can`t see devices on map, also devices are not shown from upper drop down menu. Same with ports. Any ideas ? Thank you all. I use R4586 Andrej Kolakovic

1 0

Moving to the Subscription Edition on CentOS
by Stefano Cordio 12 Oct '13

12 Oct '13

Hello guys, I'd like to inform you that on my CentOS 6.4 I followed the steps for moving the installation (http://www.observium.org/wiki/Edition_Split#How_do_I_move_to_the_Subscripti on_Edition.3F), but I also had to move the logs folder: mv observium-old/logs observium otherwise the Apache startup gave that error on /var/log/httpd/error_log: (2)No such file or directory: httpd: could not open error log file /opt/observium/logs/error_log. Unable to open logs Kind Regards, Stefano

1 0

Power meter error graphs
by Ryan Milton 12 Oct '13

12 Oct '13

I have a number of power meters from tripplite and obviously they don't have network statistics for traffic. So, I get the attached errors for the ports graphs on the overview page and when I click on the ports, where it takes you to the graphs of all the traffic. I want to be able to disable the display of this, since everything else is correct. Is there an element in code somewhere that I can just comment that out or something? Regards, Ryan Milton MVS Network Manager o: 201-447-1505 x124 c: 862-249-5230 www.mvsusa.com<http://www.mvsusa.com/> [MVS final logo GOOD very small]

3 9

Wrong data in CPU Load graph
by Christian Hügel 11 Oct '13

11 Oct '13

Hi, the graphs for the CPU load of a linux server are not accurate. The CPU shows in vSphere client a load of 40-50% but in observium all 4 cores are permanently at 90%. Does anybody know why this happens? Thx Chris

4 10

URL Tracking
by Ryan Milton 11 Oct '13

11 Oct '13

I have been enjoying my Observium installation quite a lot, and I love the nice graphs of traffic and ports. Given that this level of detail can be pulled from the devices, is there anyone/anyway who have been able to use that info to create a URL tracker that could show to what URL's a source IP is going? I'd love to be able to integrate something like that into Observium. Regards, Ryan Milton MVS Network Manager o: 201-447-1505 x124 c: 862-249-5230 www.mvsusa.com<http://www.mvsusa.com/> [MVS final logo GOOD very small]

3 2

Ping Timeouts
by James Bensley 11 Oct '13

11 Oct '13

Howdy all, I'm getting lots of emails for a few devices that have basic ADSL connectivity. Pings from the Observium box have a high RTT and timeout (I guess). Said emails have the following subject; Device Down: device.network.net (ping) Device Up: device.network.net I am getting these emails 5 minutes apart. Each time the poller runs the host is either back up again or going down again. This is happening for a few devices with congested ADSL connections (which is a separate issue in itself!). I have tried to raise the the timeout options in the config, thus; $config['ping']['timeout'] = 3000; # timeout in ms $config['ping']['retries'] = 6; # how many times to retry the query When I am on the server pinging the devices that cause these alerts, the ping is never higher than 1500ms. Perhaps when the poller runs the delay is higher (although I haven't seen this behavious when running pings for 30+ mins). I did raise it to 6000ms but this didn't stop the alerts either. It seems to me that Observium isn't honoring those values. Has anyone had any problems with these settings before? The default timeout is 500ms which I wanted to raise anyway, but I'm not sure this is having any effect. Cheers, James.

3 3

Re: [Observium] Subs Activation?
by Adam Armstrong 11 Oct '13

11 Oct '13

There is a set of commands on the edition_split page linked from the blue message and from the main web site. You just move the old version, checkout the new one and move over your rrds and config.php. Adam. Terry Sweetser <terry+observium(a)skymesh.net.au> wrote: >How does one activate a subscription and remove the friendly blue >message from the SVN build? > >-- >http://about.me/terry.sweetser > >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

1 0

inventory api not work
by Keyman 11 Oct '13

11 Oct '13

hi, all i enable API like this: // Simple Observium API Settings $config['api']['enabled'] = 1; $config['api']['module']['inventory'] = 1; and then i use http://x.x.x.x/api.php?username=admin&password=admin&module=inventory&devic… to export data, but retured : {"login":{"success":{"code":"101","msg":"User authentification succeeded"}},"data":[]} the device id is right, i can find inventory data from url http://x.x.x.x/device/device=4/tab=entphysical/ what's wrong about this ? thanks for your answer. yuan

1 0

Re: [Observium] Wrong data in CPU Load graph
by Tom Laermans 11 Oct '13

11 Oct '13

Hi, Have you asked vmware too? Have you checked in the monitored machines what their usage is? Either way we report what the device's snmp stack reports. Sent from my mobile, please excuse brevity and spelling errors. ----- Reply message ----- From: "Christian Hügel" <christian.huegel(a)stonebyte.de> To: <observium(a)observium.org> Subject: [Observium] Wrong data in CPU Load graph Date: Fri, Oct 11, 2013 09:01 Hi, the graphs for the CPU load of a linux server are not accurate. The CPU shows in vSphere client a load of 40-50% but in observium all 4 cores are permanently at 90%. Does anybody know why this happens? Thx Chris _______________________________________________ observium mailing list observium(a)observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

1 0

Re: [Observium] Subs Activation?
by Tom Laermans 11 Oct '13

11 Oct '13

Hi. If you read the friendly blue message carefully, you'll find all you need to know. Should you choose not to, surely the project webpage will tell you the same... Sent from my mobile, please excuse brevity and spelling errors. ----- Reply message ----- From: "Terry Sweetser" <terry+observium(a)skymesh.net.au> To: "Observium Network Observation System" <observium(a)observium.org> Subject: [Observium] Subs Activation? Date: Fri, Oct 11, 2013 07:50 How does one activate a subscription and remove the friendly blue message from the SVN build? -- http://about.me/terry.sweetser _______________________________________________ observium mailing list observium(a)observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

1 0

Re: [Observium] Observium Installation Startup and Configuration -- Perhaps $PATH Problems?
by Tom Laermans 11 Oct '13

11 Oct '13

You're confusing a filesystem subdirectory with an http subdirectory, everything after that is moot. Sent from my mobile, please excuse brevity and spelling errors. ----- Reply message ----- From: "Ron Wingfield" <Ron.Wingfield(a)Archaxis.net> To: "Observium Network Observation System" <observium(a)observium.org> Subject: [Observium] Observium Installation Startup and Configuration -- Perhaps $PATH Problems? Date: Fri, Oct 11, 2013 05:32 Good evening Adam and all: Yes, I'm aware of the subdirectory caveat and I've installed in /opt/observium as advertised (/opt is in the root directory). I have been able to login, once I found that the password is MD5 encrypted and I changed the value in the users table element, password. (Have to use phpMyAdmin.) Regardless, my questions are with regard to the value for the base_url and the various *_dir values. See the following: Default directories (as per ./observium/config.php) // Location $config['install_dir'] = "/opt/observium/"; // moot considering the following: Default directories (as per ./observium/includes/default.inc.php) $config['install_dir'] = "/opt/observium"; #$config['html_dir'] = $config['install_dir'] . "/html"; #$config['rrd_dir'] = $config['install_dir'] . "/rrd"; #$config['log_file'] = $config['install_dir'] . "/observium.log"; #$config['temp_dir'] = "/tmp"; Un-commenting the (previous) last four instructions seems to make no difference . . .? // This should *only* be set if you want to *force* a particular hostname/port // It will prevent the web interface being usable form any other hostname // $config['base_url'] = "http://observium.company.com"; $config['base_url'] = ""; // Without this set to something (including a NULL value), // the index.php page will not render properly. (Noted by RTW) If the logout option is selected per the following associated code, <li><a href="logout/" title="Logout"><i class="oicon-door-open-out"></i> Logout</a></li> The system complains that: “The requested URL /index.php/logout// was not found on this server.” This seems logical because the href=”logout/” construct is apparently poorly formed. It suggests that there is a relative directory named logout, and nothing is specified as an object. Perhaps there is some concatenation problem. The same is true for other menu options, e.g.: <a href="logout/ . . .etc <a href="settings/ . . .etc <a href="preferences/ . . .etc <a href="about/ . . .etc . . .all fail. I suspect some path issues. Can someone tell me where/what object (PHP script?) executes the logout procedure? In other words, where is the <li><a href="logout/" title="Logout"><i class="oicon-door-open-out"></i> Logout</a></li> list element construct created? Finally, in order to get the "home page" (the one with the world map) to render following successful login, the following was necessary: RE: /opt/observium/html/pages/logon.inc.php Changed url to relative reference (removed “/” before “image”) to enable display of image:  <div class="well" style="padding: 50px; background-image: url('images/login-hamster-large.png'); background-position: ...etc. BTW, think of me as someone in "userland" that is PHP capable. Thanks for any suggestions; in the mean time, I'll revisit my Apache2.2 httpd configurations for possible path problems. I think that with a little "jump start" help, I can port this to FreeBSD. OTTF, Ron W. Adam Armstrong wrote: Observium doesn't work from a subdirectory. Also, FreeBSD is not really supported, as non[e] of the development team use it, and Observium relies a little more heavily on userland than most PHP applications. adam.

1 0

Subs Activation?
by Terry Sweetser 11 Oct '13

11 Oct '13

How does one activate a subscription and remove the friendly blue message from the SVN build? -- http://about.me/terry.sweetser

1 0

MIB requests
by Robbie Wright 10 Oct '13

10 Oct '13

Just throwing an idea out there, but for devices that are not currently supported, but have published MIB's, could we put some type of voting system out there, ala http://ideascale.com or something similar? Help guide Adam and Tom a bit in the development so they know where the interest is and possibly get some donations for the popular ideas/devices? Robbie Wright Siuslaw Broadband <http://siuslawbroadband.com> 541-902-5101 **For support issues, please email support(a)siuslawbroadband.com.**

3 4

Multiple year subscription
by Jeremy Custenborder 10 Oct '13

10 Oct '13

Hey All, Is there an easy way for me to pay upfront for a multiple year subscription? I have a client that would like to have a multiple year subscription but it's a bit of a pain for them to use PayPal. I would like to pay for 5 or so years ahead of time? J

2 2

Perf_time tables
by Jeremy Custenborder 10 Oct '13

10 Oct '13

Are the perf_times and devices_perftimes used for anything other than diagnostics at the moment? These tables are the significant majority of the data in our database. I truncated them once to free up some space. Is this something that I could do on a regular basis? perf_times => 2013-09-20 04:00:02 to 2013-10-10 17:23:21 generated 25,476,751 rows devices_perftimes => 2013-09-20 20:45:02 to 2013-10-10 17:40:21 generated 24,536,885 rows

2 1