- observium - lists.observium.org

Linux Agent redesign
by Adam Armstrong 07 Aug '13

07 Aug '13

Hi All, It has turned out that the agent is a little bit difficult to work with, trying to extract data using only shell scripting is a bit of a pain. It's difficult for us to do any form of authentication or encryption, and parsing of the returned data can be a little haphazard because we rely on correct formatting. I've been thinking of alternative designs, possibly using PHP (Perl or Python would probably be better, but they're not viable languages for us) and either an embedded web server or hanging off inetd as it currently does. This would allow us to do encryption, to require some form of authentication and to pass around easier to parse JSON or serialize()'d arrays Personally, I almost always install a PHP CLI interpreter on my servers because I do 'shell' scripting in PHP, but what do the rest of you think? Would you be willing to install php-cli on your servers if you got the ability to do encrypted/authenticated/compressed agent connections? What about if we allowed you to funnel SNMP queries via the agent? (only for UNIX hosts, obviously) adam.

2 2

OS/distro version parser
by Adam Armstrong 07 Aug '13

07 Aug '13

Hi All, We're testing a new version of the old 'distro' script used to detect distribution. Please run : ./scripts/os From inside your observium directory on updated installation (you may need to chmod 755 it first). I get: root@alpha:/home/observium/dev# scripts/os Linux|3.8.0-19-generic|amd64|Ubuntu|13.04 root@alpha:/home/observium/dev# The final part is the distro release. Please reply to this post if your distro release is returned as empty. The script might not return a distribution value for non-Linux OSes, this is normal. We're trying to make sure it can always ascertain the distro version and type. Thanks, adam.

4 7

Alerting, how to configure
by Rory Schellekens 06 Aug '13

06 Aug '13

Hi all, What is the status of Alerting for Observium? Also I don't see any documentation about it. The main goal is to add our own alarms to Observium. With kind regards, Rory Schellekens

2 5

Sensors
by simon schaller 06 Aug '13

06 Aug '13

hello all, i try to add sensors in my observium and i know that there is sensors in my servers. is there somenone who can explain me how to add it please? thanks to read this mail, schaller simon

3 2

Permissions Issue:
by Ravindra Gupta 06 Aug '13

06 Aug '13

Dear Team, Still we are facing below issue when we have click add device option in web panel .I have configured observing in Centos. I have been following many URL but did not get any solution. Error You have insufficient permissions to view this page. #tail -f /var/log/httpd/error_log File does not exist: /opt/observium/html/images/no-48.png, referer: http://localhost/addhost/ I am waiting for your help...

3 3

Re: [Observium] Pages Missing CSS
by Tom Laermans 06 Aug '13

06 Aug '13

I would, yes. (create a stable branch) And yes, I broke the thread with my mobile client especially for you :D Sent from my mobile, please excuse brevity and spelling errors. ----- Reply message ----- From: "Nikolay Shopik" <shopik(a)inblock.ru> To: "Observium Network Observation System" <observium(a)observium.org> Subject: [Observium] Pages Missing CSS Date: Mon, Aug 5, 2013 17:08 Would you introduce "stable" branch if someone start paying for it? :-D You see fixes are fixes, but most bugs comes when you start introducing new feature called "global domination" and this is where things get worse then just getting fixes. Its just selecting between I need this fix, but this brake that, because of new thing here, and I need choice what are more important for me. I know fixing bugs are boring, and introducing new features (yay \o/) are more fun. :) But its understandable why no stable exist for now, no man power to do qa and backport shit here and there, like in any pet projects, muhahaha. On 02/08/13 01:05, Adam Armstrong wrote: > On 2013-08-01 21:52, Job Snijders wrote: >> Isn't it your job to do QA? >> >> If you favor stability why would you do 'svn up' on a production >> element? :) > > Quite right. And for people who can't or don't want to do this > themselves, we can take their money in exchange for managing their > installation updates and maintenance :) > > adam. > _______________________________________________ > 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

1 0

Re: [Observium] Permissions Issue:
by Tom Laermans 06 Aug '13

06 Aug '13

Dear User, Please read the responses to your question instead of sending it again and again. It serves no purpose other than you getting ignored completely by mailing list members. The error in the logs is completely unrelated. Sent from my mobile, please excuse brevity and spelling errors. ----- Reply message ----- From: "Ravindra Gupta" <ravindragupta9(a)gmail.com> To: "Observium Network Observation System" <observium(a)observium.org> Subject: [Observium] Permissions Issue: Date: Tue, Aug 6, 2013 09:20 Dear Team, Still we are facing below issue when we have click add device option in web panel .I have configured observing in Centos. I have been following many URL but did not get any solution. Error You have insufficient permissions to view this page. #tail -f /var/log/httpd/error_log File does not exist: /opt/observium/html/images/no-48.png, referer: http://localhost/addhost/ I am waiting for your help...

1 0

Re: [Observium] patch for rrdtool syntax
by Adam Armstrong 06 Aug '13

06 Aug '13

My guess is that tobi won't release something which needlessly breaks a lot of existing stuff because of ocd parser code. :) Peter Childs <pchilds(a)staff.iinet.net.au> wrote: >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

1 0

decoupling UI instances from pollers/main instance
by Peter Childs 06 Aug '13

06 Aug '13

As a bit of thought food I have had some interesting success in decoupling a 'UI' instance of observium from the instance that is 'polling' etc (I will call that the 'main' instance). I installed another deployment of observium using the same SQL instance as the 'main' instance. The issues then are access to the RRDtool data. I complied a 'new' version of rrdtool on both main and 'ui' instance boxes from 'trunk' of rrdtool development. The current trunk has support for the 'FETCH' command from rrdcached (usually rrdcached is only used for caching writes to reduce IO load). I ran rrdcached instance on the 'main' box, and configured $config['rrdcached'] to point to the IP/port of the main box's rrdcached process. As rrdcached is looking for 'relative' names to the path it is running in I so I hacked rrdtool.inc.php (see below as example — do not apply this patch) There were various issues with graph syntax that cropped up due to changes in the syntax parsing in the 'trunk' version of RRDtool, but pretty easy to work around those with some minor formatting changes (I'm sure I haven't found them all). The other issue is that for various composite graphs 'is_file( $rrd_filename )' is called — I hacked around this by just dumping all the rrd files via rsync (nasty) on the local box. I assume this could be worked around with a rrd-file-exists() type function that could say 'yes' for 'remote' rrd files (I'm not sure yet how to determine that via the rrdcached protocol if at all possible… perhaps there are other possible solutions …) When rrdtool graph now draws a graph it hits rrdcached on the main box, gets all the metrics, and presto … So far I notice that Misc graph fails — see above .. Usually some syntax causing the 'new' parser to get sad Speed — there are noticeable delays in drawing some of the graphs as you whip around the UI. Its not 'glacial', and it is potentially not the end of the world, but I'd like to understand where these delays come from with decoupled rrdtool — I have a feeling that potentially the rrdcached is not dealing with a pile of concurrent requests well — there may be some ways to make this better, and/or work around this. Why oh why? Currently in my 'day job' we run lots of different systems for different networks doing metrics gathering and display. Most of them are 'topped' out despite large amounts of time over the years optimising them, putting in 'faster' stuff etc. I'd like a metric(s) backend that can eat whatever we need, and eats more based on adding more commodity hardware. If my metrics backend can 'talk' rrdcached protocol, and systems like Observium can use this, then I feel like this might be a 'good thing'. Need more polling, just split the pollers. Need more UI goodness, just add more UI nodes behind a http load balancer. Some of the new metrics backends have some interesting stuff — specifically I'm looking at the 'not-released-yet' OpenTSDB. I'm pretty sure it is not the 'right' thing for simple installations, but for larger platforms it looks appealing. Thoughts? Cheers, Peter Various hacks pjchilds@uitest:/opt/observium$ svn diff includes/rrdtool.inc.php Index: includes/rrdtool.inc.php =================================================================== --- includes/rrdtool.inc.php (revision 4254) +++ includes/rrdtool.inc.php (working copy) @@ -107,6 +107,7 @@ if ($config['rrdcached']) { + $options = str_replace( $config['rrd_dir']."/", '', $options ); fwrite($rrd_pipes[0], "graph --daemon " . $config['rrdcached'] . " $graph_file $options"); } else { fwrite($rrd_pipes[0], "graph $graph_file $options"); @@ -158,6 +159,7 @@ $cmd = "$command $filename $options"; if ($command != "create" && $config['rrdcached']) { + $options = str_replace( $config['rrd_dir']."/", '', $options ); $cmd .= " --daemon " . $config['rrdcached']; }

1 0

patch for rrdtool syntax
by Peter Childs 06 Aug '13

06 Aug '13

Gday folks. I have been looking at the 'trunk' version of rrdtool where some differences exist in the parsing of 'graph' commands. The attached patch prevents 'Unused Arguments' when rrdtool graph is called with 'empty' argument bits. I assume I have probably 'missed' some bits, and I understand this doesn't impact Observium users yet with currently released Ubuntu rrdtools packages, but I assume at some stage it will impact, and it might be good to 'miss' that impact ? Cheers, Peter

2 2

Pages Missing CSS
by Steve Nelson 05 Aug '13

05 Aug '13

Hi All I have now managed to at least get a build of Observium and also added a couple of test devices to poll. All seems well apart from one issue. All pages (apart from the initial index page) are appearing as if they had no CSS applied to them, also at the foor of each page I am seeing some mysql output which I am not sure should be there, items like : Time 0.005s MySQL Cell 0/0s Row 11/0.001s Rows 6/0s Column 0/0s Memory Cached 34.7kB Page 5.86MB Peak 6.14MB Any advice on getting the pages back on track would be great. Cheers Steve Steve Nelson 10Yr CCIE#10055, OSCP#3725 Senior Network Engineer steve.nelson(a)calibre-secured.net Telephone : 0845 5576355 Direct Line : 0191 6542060 Mobile : 07702247286

8 10

Help
by Eliezer Sousa 05 Aug '13

05 Aug '13

hello, I wonder how I can monitor equipment that do not have snmp. I have some data links that do not use SNMP, and must monitor them. from already thank you for your help. ass eliezer

4 3

Line colour selection
by Robert Williams 05 Aug '13

05 Aug '13

Hi, I’m currently setting up some of our own custom BMS hardware on Observium (it’s all loosely based on Carel components). It’s going fine and I’ve got most of the sensors on there and graphing ok so far. They are mostly temperature, humidity and pressures. Anyway, the reason for this post is that I can’t find how Observium selects the colours for the lines on the graphs when temperatures are involved? I’m sure there is a pattern (or setting for it) but I can’t see it. I know it’s only cosmetic, but can someone point me in the right direction please? Cheers :) Robert Williams Custodian Data Centre Email: Robert(a)CustodianDC.com http://www.CustodianDC.com

1 0

Re: [Observium] Insufficient Permission Issue:
by Adam Armstrong 05 Aug '13

05 Aug '13

Oh dear, and there was me thinking of looking at your login issues. Maybe not! You use Ubuntu/Debian because that is what we tell you to use. The vast majority of 'issues' reported by centos/rhel/etc users are related to 'features' and other braindeadness of their chosen distro. None of us use anything other than Debian/Ubuntu. Debian/Ubuntu is what we test on and Debian/Ubuntu are all we are prepared to discuss on this list. If you feel you know better, then you had better be able to make everything work properly yourself, because we *will* be rude to you for ignoring all of our warnings. After all, you get the service you pay for, right? Alternatively, I hear cacti runs fine on centos. Oh, and it hasn't helped our tolerance levels that most of the 'stupid' questions we get from people who want spoonfed step by step solutions come from centos users. It seems that stupidity and laziness often leads to choosing centos. Adam. Also notable is that thus far all of our major sponsors to date run Debian or Ubuntu. DaZZa <dazzagibbs(a)gmail.com> wrote: >On 3 August 2013 22:15, Adam Armstrong <adama(a)memetic.org> wrote: >> >> a) why centos > >Because Ubuntu is a pain in the backside for someone used to RHEL/SLES >type implementations? > >I use CentOS because I hate the way Ubuntu manages services, and >because I'm used to it. However, I don't belittle Ubuntu at every >opportunity, as seems to be the modus operandi here. I saw a quote by >someone who works in a mixed Debian/Ubuntu and RHEL shop the other day >which is particularly pertinent > >"I use XX because it works for me." > >Can we get away from the distro bashing and concentrate on fixing >issues? Yes, I know you use Ubuntu and love it, and I know you don't >want to support CentOS, but that doesn't mean the whole world needs to >follow you. > >You have an excellent product in Observium - it's a shame to see such >crap bringing it down for people who are only trying to make it work >(and, by definition, increase penetration of the product). > >DaZZa >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

3 2

Re: [Observium] Insufficient Permission Issue:
by Adam Armstrong 05 Aug '13

05 Aug '13

I can't say I've encountered this or heard of it reported before. Are you sure? Adam. DaZZa <dazzagibbs(a)gmail.com> wrote: >On 3 August 2013 16:34, Ravindra Gupta <ravindragupta9(a)gmail.com> wrote: >> We have configured observium monitor tool in centos 5.8, when we have >> requested url in the browser we got the below error. > >I get that one once in a while too - it generally means your login >session has timed out - if you try and access anything else, you'll >get dumped back to the main page and have to log back in. > >By "timed out", I mean "stopped working for reasons not necessarily >including inactivity", because I've had it happen between one click >and the next > >DaZZa >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

2 1

Insufficient Permission Issue:
by Ravindra Gupta 05 Aug '13

05 Aug '13

Hii, We have configured observium monitor tool in centos 5.8, when we have requested url in the browser we got the below error. [image: Inline image 1]

4 4

Re: [Observium] Pages Missing CSS
by Tom Laermans 02 Aug '13

02 Aug '13

If you want to tell us when to commit and not commit, you'd better be paying us during both of those timespans. Why don't you just no svn up before seeing what it does? Sent from my mobile, please excuse brevity and spelling errors. ----- Reply message ----- From: "Tristan Rhodes" <tristanrhodes(a)weber.edu> To: "Observium Network Observation System" <observium(a)observium.org> Subject: [Observium] Pages Missing CSS Date: Fri, Aug 2, 2013 18:34 Would it be possible to come up with a compromise solution that meets the requirements of stability and freshness? Here is one idea: What if SVN updates were only submitted on Sunday through Wednesday? Users can then update on the day of the week that meets their needs. For example, people who want to help test new (and potentially risky) code can update cron on Thursday or Friday. People who want a more stable environment choose Saturday. Of course, SVN updates would be made on any day to resolve an regression or bug that was found. What do you think of this idea? Cheers, Tristan On Thu, Aug 1, 2013 at 4:02 PM, Paul Gear <observium(a)gear.dyndns.org> wrote: I update automatically in cron.daily. Because you tell me to. If i have to choose between shiny/new and stable, i will always choose stable, all other things being equal. On 08/02/2013 07:23 AM, Adam Armstrong wrote: A majority of users actually seem to update pretty regularly. We assume this is because they like the new things (and the fixing of problems) rather than because we told them to, because the only people we tell to update regularly are the 10% of users present on this list :) Adam. Paul Gear <observium(a)gear.dyndns.org> wrote: On 08/02/2013 07:03 AM, Adam Armstrong wrote: On 2013-08-01 21:30, Stefan Milo wrote: A tool like observium is often used by people in Operations. People in Operations tend to favor stability over new features. The old if it's not broken mantra... Well then, you probably want to be using Cacti. Judging by the update habits of our existing userbase, they agree with us. "Agree" is much too strong a term; "tolerate" would be better; "it's better than using Cacti" would probably be more accurate than both. :-) _______________________________________________ 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 -- Tristan Rhodes Network Engineer Weber State University (801) 626-8549

1 0

Re: [Observium] Pages Missing CSS
by Adam Armstrong 02 Aug '13

02 Aug '13

I think anyone who wanted something like this would be better served by paying attention to the svn changelog and only updating after a day or two of inactivity. All of the information needed is available. We won't adopt any practices which increase our workload without obvious benefit. Delaying svn commits is the #1 cause of the "missing include" bug being committed. Adam. Tristan Rhodes <tristanrhodes(a)weber.edu> wrote: >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

1 0

Re: [Observium] Pages Missing CSS
by Adam Armstrong 02 Aug '13

02 Aug '13

A majority of users actually seem to update pretty regularly. We assume this is because they like the new things (and the fixing of problems) rather than because we told them to, because the only people we tell to update regularly are the 10% of users present on this list :) Adam. Paul Gear <observium(a)gear.dyndns.org> wrote: >On 08/02/2013 07:03 AM, Adam Armstrong wrote: >> On 2013-08-01 21:30, Stefan Milo wrote: >>> A tool like observium is often used by people in Operations. People in >>> Operations tend to favor stability over new features. The old if it's >>> not broken mantra... >> >> Well then, you probably want to be using Cacti. Judging by the update >> habits of our existing userbase, they agree with us. > >"Agree" is much too strong a term; "tolerate" would be better; "it's >better than using Cacti" would probably be more accurate than both. :-) > >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

3 2

Certain MySQL agent graphs not showing
by Shaun Bradley 02 Aug '13

02 Aug '13

Hi, Once again it's me. I have the mysql and apache agents running on a few machines and have observium picking the data up. This is working fine for apache and mostly for mysql, but I have missing graphs for 'process list' 'myisam indexes' (our dbs are mainly innodb based), table locks, and temporary objects. I have followed the agent instructions to a tee as far as I can see and have double checked my configuration, as well as googled for any info, but I am stuck on this one. Is there any way I can either remove these broken graph links or fill out the information? I can't see what I'd change to make them appear to be honest. Any pointers would be gratefully received. Thanks, Shaun

3 7

Digital optics monitoring sensors - cisco 2960S
by Petr Gregor 02 Aug '13

02 Aug '13

Hi guys, I am successfully using observium in our network. Monitoring about 180 cisco switches. Thanks! I am however missing graphs for optics SFP modules. I can see the sensors in snmpwalk: > ENTITY-MIB::entPhysicalName.1061 = STRING: TenGigabitEthernet1/0/1 > ENTITY-MIB::entPhysicalName.1062 = STRING: Te1/0/1 Module Temperature Sensor > ENTITY-MIB::entPhysicalName.1063 = STRING: Te1/0/1 Supply Voltage Sensor > ENTITY-MIB::entPhysicalName.1064 = STRING: Te1/0/1 Bias Current Sensor > ENTITY-MIB::entPhysicalName.1065 = STRING: Te1/0/1 Transmit Power Sensor > ENTITY-MIB::entPhysicalName.1066 = STRING: Te1/0/1 Receive Power Sensor And I can see the sensors listed in inventory page in observium. There are however no values or graphs on the health page. Thank you for any tips. Gregy

3 6

Locations menu
by Tom Henderson 02 Aug '13

02 Aug '13

Hi All Perviously my Devices > Locations submenu would show the custom location entered on the device properties page (under Override sysLocation). Is there a setting to revert to this behaviour? I've tried using the $config['location_map'] setting, but this seems to only apply to the actual sysLocation, not the overriden value. On a side note I'd love to be able to get closer on the map. The closest I could get it was the whole of New Zealand, on which I have a single green spot covering the whole of Auckland! Thanks for the great product. Cheers Tom Henderson IT Manager | Pack & Company | 027 461 4403 | 09 929 2750 | tom(a)pack.co.nz -- ---------------------------------------------------------------------------If you are not the intended recipient of this email, you must not disclose, copy, or distribute any information contained in or attached to it, or take any action/inaction based on it. Please contact us immediately, and delete this email from your system. Any views or opinions presented in this email are solely those of the author and not of Pack & Company (NZ) Limited (the "Company"), however any intellectual property contained in or attached to this email is the property of the Company. Thank you.

2 2

Web Interface display issue
by Colin Anderson 01 Aug '13

01 Aug '13

I've performed several installations, but have a problem with this one. The root/home webpage displays correctly, but all other pages lack the normal formatting. All of the menu items from across the top of the page are displayed as an unformatted bulleted list on everything except the root http://observium webpage. I compared my /etc/apache2/sites-available/default file to a working installation and it's exactly the same. I've gone back over the installation instructions twice. A2enmod rewrite is enabled, but I'm not sure it's working... If I click on the menu "Devices" -> "All Devices" it takes me to http://observium/devices/ with the trailing backslash. I can remove the trailing backslash and the page displays correctly. This doesn't seem to work with other pages further down the menu structure such as http://observium/devices/type=network/. The only difference between this installation and my other installations is that the others are on Debian 6 and this is Debian 7.1 (Wheezy). Sorry for the long explanation, but any help is MUCH appreciated. Thanks, Colin Anderson

3 3

Another frontpage patch
by Paul Gear 01 Aug '13

01 Aug '13

Hi all, Here's another patch i've made to the front page, this time to add a "micrographs" function similar to the minigraphs, but with the difference that multiple sections are allowed, each corresponding to a row in the $config['frontpage']['micrographs'] array. Here's an example config excerpt: ... $config['frontpage']['order'] = array('device_status_boxes', 'minigraphs', 'micrographs', 'status_summary', 'eventlog'); ... $config['frontpage']['micrographs']['network']['ids'] = '1,device_bits,router1;2,device_bits,router2;3,device_bits,switch1;4,device_bits,firewall1;5,device_bits,firewall2'; $config['frontpage']['micrographs']['network']['title'] = 'Network'; $config['frontpage']['micrographs']['servers']['ids'] = '6,device_bits,vmhost1;7,device_bits,observium;8,device_bits,fileserver;9,device_bits,vmhost2;10,device_bits,backup'; $config['frontpage']['micrographs']['servers']['title'] = 'Servers'; ... This produces the following on the front page: Both the row labels and the graph labels are optional - if you leave them both out you get just a bare row of clickable graphs. There's no label justification, which looks a bit funny when labels are different lengths, but i figure people can just add space or whatever to make their front page look how they want it. Patch is attached and also available at http://libertysys.com.au/patches/observium/observium-r4209-micrographs.patch Hope you like it. Regards, Paul

1 1

status
by Eliezer Sousa 31 Jul '13

31 Jul '13

hello guys, I would like to know how to change the refresh time of the equipment. because when charges that the equipment is down takes too long to update the status even when this is online. where can I change this time? thank you. ass eliezer

2 1