- observium - lists.observium.org

Re: [Observium] How to add linux server?
by Adam Armstrong 26 Sep '12

26 Sep '12

Services are deprecated. The code is still there because it would take time to remove. Do not use it. We have plans to implement better alerting in the future, but there is always a shortage of time. Adam. Andrey Lebedev <alg(a)condensa.ru> wrote: >On 2012-09-25 23:05:51 +0100, John Broome said: > >> I prefer my car, but that doesn't mean I'm going to use it to open a >> can of beans. I use the tool that's designed to do what I want, >> reliably. And if the creators of the software say it's not meant to >> be used for service monitoring I'm going to take their word for it. > >Hmm ... interesting that the author wanted to say a demo of their >product. Maybe this is a joke? >http://demo.observium.org/device/device=34/tab=services/ >http://demo.observium.org/device/device=7/tab=services/ >http://demo.observium.org/device/device=6/tab=services/ >http://demo.observium.org/ >kermit >Service Down >ssh > >It's no "service monitoring"? >Ok. :) > > > > >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

1 0

Re: [Observium] How to add linux server?
by Adam Armstrong 26 Sep '12

26 Sep '12

I think you misunderstand Occam's razor. We do not do up/down monitoring, no more than cacti or Mrtg do. Adam. Andrey Lebedev <alg(a)condensa.ru> wrote: >On 2012-09-25 21:45:12 +0100, Paul Cupis said: > >> "Observium is not intended to replace an up/down alerting system, but >> rather to complement it with an easy to manage, intuitive representation >> of historical and current performance statistics, configuration >> visualisation and syslog capture." > >Do not see any logic. >http://en.wikipedia.org/wiki/Occam's_razor >IMHO. > >But observium is preferd for me (over nagios or zabbix). > > >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

1 0

How to add linux server?
by Andrey Lebedev 25 Sep '12

25 Sep '12

Hi all, Sorry. 1) Offtopic. I unfortunately do not get your answers via mail. And look at them through http://postman.memetic.org/pipermail/observium/2012-September/thread.html How do I reply to this thread - but do not start a new topic every time? I have long lost touch (and forgotten) how to use mailing lists - last it was 10 years ago. 2) For monitoring unix servers use snmp? I do not understand a bit - but then why in wiki talk about the unix- agent. I saw that to unix-agent in any way no related to snmp. In my look this product have a lot of undocumented features that are unclear how they work. Regards, Andrey

4 13

How to add linux server?
by Andrey Lebedev 24 Sep '12

24 Sep '12

Hi all! Sorry for the newbie question. I just Installed Observium. Connected to many snmpv3 network devices - Cisco, Extreme, Juniper ... I'm fine. I'm happy with everything. Now I want to connect my first Linux server. But I find no documentation - how? I installed check_mk_agent on linux. root@monitoring:/opt/observium# telnet x.x.x.60 6556 | head Trying x.x.x.60... Connected to x.x.x.60. Escape character is '^]'. <<<check_mk>>> Version: 1.2.0p2 AgentOS: linux PluginsDirectory: /usr/lib/check_mk_agent/plugins LocalDirectory: /usr/lib/check_mk_agent/local AgentDirectory: /etc/check_mk OnlyFrom: 127.0.0.1 x.x.x.61 I do not know what to do next? Thank you for your help. Regards, Andrey

3 2

How to add linux server?
by Andrey Lebedev 24 Sep '12

24 Sep '12

+screenshot Only snmp devices: http://s11.postimage.org/6izs930cj/Picture_478.png

1 0

How to add linux server?
by Andrey Lebedev 24 Sep '12

24 Sep '12

Hello Paul , Thanks for the help, but I do not understand what you are saying (about web-interface). I do not have anything on the screen. I send the link to screenshot. http://s12.postimage.org/oddm5vdbx/Picture_477.png root@monitoring:/opt/observium# cat config.php | tail -f $config['poller_modules']['unix-agent'] = 1; ?> root@monitoring:/opt/observium# /etc/init.d/apache2 restart * Restarting web server apache2 ... waiting ...done.

1 0

Re: [Observium] SSD for better performance
by Adam Armstrong 19 Sep '12

19 Sep '12

The maths and specs suggest a modern SSD should survive observium workload for quite a while, but I'm still worried about the random failure rate and the likelyhood that individual cells get worn out. In general SSDs are such a strange tech, there's almost no way to really improve them except with fancy controller hacks, the flash itself is a dead end tech. Good luck :-) Adam Cushard Michael <mcc(a)us.axway.com> wrote: >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

3 2

Re: [Observium] SSD for better performance
by Adam Armstrong 19 Sep '12

19 Sep '12

I would recommend dumping the data to a real disk occasionally, as observium/rrd-style workloads are pretty hard on flash-based storage. I have SSDs in my workstations, but I'm not sure if I'd trust them in a server quite yet. Please report back if/when it breaks :-) Adam. Cushard Michael <mcc(a)us.axway.com> wrote: >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

3 2

Re: [Observium] port ignore partly fails on large number of ports
by Adam Armstrong 19 Sep '12

19 Sep '12

Haha. Glad it works :-) We should add that to the docs somewhere! Adam. Maarten De Wispelaere <maarten.dewispelaere(a)intec.ugent.be> wrote: >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

3 3

port ignore partly fails on large number of ports
by Maarten De Wispelaere 19 Sep '12

19 Sep '12

Hi, I recently discovered the following : When trying to set "Ignore" status on switches with large number of ports (eg. force10 e1200 or c300 chassis), the setting doesn't get saved for some of the ports. More specific, at the end of the list. This is what I did : Toggle all -> removed the toggle from the mgmt interface -> then save What happens: Most ports get "ignored" status, part of them don't After getting the status, they again have a 'not-ignored-status' (checkbox not checked). If I check them again & press save -> same issue. Seems like the checkboxes only get processed to a certain number of items ? Does this sound familiar ? Thanks in advance, Maarten De Wispelaere Department of Information Technology Internet Based Communication Networks and Services (IBCN) Ghent University - IBBT Gaston Crommenlaan 8 Bus 201, B-9050 Gent, Belgium E-mail: maarten.dewispelaere(a)intec.UGent.be Tel. +32 (0)9 33 14979 Fax +32 (0)9 33 14899

5 5

Accurate Graphs in Billing are probably not so accurate
by Matthias Cramer 18 Sep '12

18 Sep '12

Hi I've noticed two things in the "Accurate" graphs in the Billing part: 1. Are Interface counter overflows or equipment reboot) probably not catched? - I see very high spikes (200 Gbps and more) on a 1G interface sometimes in sometimes out, but not at the same time. the quick graphs are ok 2. If I monitore more tha one interface for billing (Unfotunately one of the is down) the accurate graps are at leaset 1000 times to high... Atached some graphs... I runn lates svn (0.12.9.3378). Any ideas ? Regards Matthias -- 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

2 1

Eaton Powerware UPS
by Aaron Maurer 17 Sep '12

17 Sep '12

Hi all, First time poster here. We use observium here at our office and think it is a great program. We were interested in having observium read SNMP information from our Powerware 9130 UPS's. Currently observium can't identify them. We were wondering if there was any interest in the community to write a snmp module to read from the Powerwares. The documentation for the SNMP values is posted in the PDF files which I can provide. If there is positive interest for anyone on helping take this on I can also provide SNMP walks of my Eatons. Here is a snippet of the first few lines of SNMP information: SNMPv2-MIB::sysDescr.0 = STRING: Powerware 9130 3000 SNMPv2-MIB::sysObjectID.0 = OID: SNMPv2-SMI::enterprises.705.1 DISMAN-EVENT-MIB::sysUpTimeInstance = Timeticks: (2144221738) 248 days, 4:10:17.38 Thank you, Aaron Maurer Diversified Business Systems Inc. Office: 440-942-6961 Fax: 440-942-5819 amaurer(a)dbsinet.com http://www.dbsinet.com

1 0

Re: [Observium] Checkpoint Firewall Support?
by Steve Costaras 04 Sep '12

04 Sep '12

Doing some more digging here, it appears UCD-SNMP-MIB values for ssCpuUser (system & idle) are being populated correctly from checkpoint utm/splat deployments. But other items like interrupts, etc are working. The cpu values appear to be stored in the checkpoint enterprise mib: -- CHECKPOINT-MIB::multiProcIndex.1.0 = INTEGER: 1 CHECKPOINT-MIB::multiProcIndex.2.0 = INTEGER: 2 CHECKPOINT-MIB::multiProcUserTime.1.0 = INTEGER: 0 CHECKPOINT-MIB::multiProcUserTime.2.0 = INTEGER: 1 CHECKPOINT-MIB::multiProcSystemTime.1.0 = INTEGER: 100 CHECKPOINT-MIB::multiProcSystemTime.2.0 = INTEGER: 2 CHECKPOINT-MIB::multiProcIdleTime.1.0 = INTEGER: 0 CHECKPOINT-MIB::multiProcIdleTime.2.0 = INTEGER: 97 CHECKPOINT-MIB::multiProcUsage.1.0 = INTEGER: 100 CHECKPOINT-MIB::multiProcUsage.2.0 = INTEGER: 3 -- Found the current version of the mib as well as a checkpoint 32x32 graphic that can be added to the project (attached) in case someone has more time to play with it than I do. I'm still trying to wrap my head around the coding here to update this for first, base OS (per-cpu) utilization, then expand on it to something similar to the ASA firewall tab. Also it appears that the vendor has an issue with the netsnmp package supplied with the default media 5.0.9) but they do have an update package (5.3.1). This at least provides for cpu enumeration but still no stats (per above).

2 1

Re: [Observium] Strange graph
by Adam Armstrong 04 Sep '12

04 Sep '12

This is probably because the device frequently returns invalid data. A common problem with tacked systems like the 3750 and VSS. Adam. Nicolas COULIN <nicolas.coulin(a)kleegroup.com> wrote: >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

2 1

Strange graph
by Nicolas COULIN 04 Sep '12

04 Sep '12

Hi, Graphs for my stack was not correct. I have 4 swtich in stack. And observium return values very strange. 500Go by interface.... :/ Cisco IOS Software, C3750 Software (C3750-IPBASEK9-M), Version 15.0(2)SE, I have the last svn update on centos 6.2 I test to delete the host, delete rrd directory, and add my host. But the problem come back. I don't understand why. Thanks [Logo Kleegroup] Klee Group is a great place to work - Il fait bon travailler ? Klee Group

1 0

No more graphs after upgrade
by Max Leon 03 Sep '12

03 Sep '12

Hello, I have a problem I had revision 3049, due to a small problem displaying the heath status of the drives I updated to HEAD and immediatly run ./discovery.php -h new. All functionality seems to be there but I have no new graphs, any ideas on how to troubleshoot/fix this problem? Thank you. Max Leon

2 2

Re: [Observium] [Patch] Collectd v5 Apache fix
by Adam Armstrong 03 Sep '12

03 Sep '12

Please to be not distributing broken patches. Adam. Chris Bennett <chris(a)ceegeebee.com> wrote: >Hi Tom, > >> I presume this breaks users who still use collectd <5? > >Yep, treat the email & bug submission as a report of the problem and >ignore the patch, to fix this properly. I'm just putting it out into >the public so others might find this when they are troubleshooting the >same problem :) > >Regards, > >Chris >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

2 1

[Patch] Collectd v5 Apache fix
by Chris Bennett 02 Sep '12

02 Sep '12

Hi there, I noticed collectd graphing of Apache graphs is broken after Collectd released version 5 of their software. The data source name has changed from 'value' to 'count'. See attached patch. I've also filed a bug report with the patch: http://jira.observium.org/browse/OBSERVIUM-122 Regards, Chris

2 2

Observirum Graphs ere empty
by Nicolas COULIN 27 Aug '12

27 Aug '12

Hi, I have install Observium since few days, and all my graphs are empty. Rrd files are present on folder. I don't understand why. I use centos 6.2. Thanks. [Logo Kleegroup] Klee Group is a great place to work - Il fait bon travailler ? Klee Group

5 7

Re: [Observium] Observirum Graphs ere empty
by Adam Armstrong 27 Aug '12

27 Aug '12

Tom Laermans <tom.laermans(a)powersource.cx> wrote: >Hi, > >You probably didn't do it right. > >Unfortunately, "my graphs are empty" doesn't gives us a lot of >information. > >Tom > >On Mon, 2012-08-27 at 09:57 +0000, Nicolas COULIN wrote: >> Hi, >> >> >> >> I have install Observium since few days, and all my graphs are empty. >> >> >> >> Rrd files are present on folder. >> >> >> >> I don’t understand why. >> >> I use centos 6.2. >> >> >> >> Thanks. >> >> >> >> Logo Kleegroup >> Klee Group is a great place to work – Il fait bon travailler à Klee >> Group >> _______________________________________________ >> 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

openBSD
by Miguel Alejandro González 26 Aug '12

26 Aug '12

Hello Can I monitor a machine with OpenBSD with Observium??? Regards!

2 1

Help with vmware
by Pavel Ondřej 22 Aug '12

22 Aug '12

Hello, Can you help me please? How can I connect observium to ESXI5 Thank you for greet program. Pavel

1 0

Re: [Observium] Checkpoint Firewall Support?
by Steve Costaras 16 Aug '12

16 Aug '12

I do have some older nokia systems (IP380's, IP530's, and I think an IP740) laying around somewhere, but most of the new systems are using Secure Platform (R65, R70, R71, R75) versions and/or UTM's (basically the same thing, they all RedHat Enterprise v5.4 9(2.6.18 kernel) slimmed down with an extra checkpoint mib tree for items outside the base OS). We also have some provider-1 deployments (management station) but that I would consider future if anything. If you know enough about handling the observium code itself I can try helping on the checkpoint side. I have some systems here to test against as well. Steve >-----Original Message----- >From: Rob J. Epping [mailto:rob.epping@gmail.com] >Sent: Thursday, August 16, 2012 04:17 AM >To: stevecs(a)chaven.com, 'Observium Network Observation System' >Subject: Re: [Observium] Checkpoint Firewall Support? > >Hi, > >On Mon, Aug 13, 2012 at 11:14 PM, Steve Costaras <stevecs(a)chaven.com> wrote: >> >> I'm just starting to look at observium for some monitoring. I see that it supports cisco based firewalls (ASA/PIX) but it does not currently support checkpoint. >> >> Checkpoint mainly runs on top of redhat linux. I'm able to get the general linux stats but not the firewall specific stats (connections/vpn/etc) similar to what's available with the cisco asa support. >> >> Basically, I'm asking if anyone has managed to get this to work. Or barring that, any information as to how to add support? > >I have some patches to submit. But these are for Nokia IP-platform with IPSO. > >Anything particular that interests you? >-- >GRTNX, >RobJE >

1 0

solaris 10 / net-snmp not getting cpu stats
by Steve Costaras 16 Aug '12

16 Aug '12

Still testing with various devices here, found an interesting issue. I have a M3000 sun box with solaris 10 on it, which gets added fine and most of the graphs appear to be eyeball correct, however am NOT stats for cpu utilization. The sun box is using net-snmp (included w/ solaris 10)(not the older sun snmpd daemon) which was the default with older systems. doing a manual walk I DO get stats: $ snmpwalk -v2c -c XXXX XXXXX 1.3.6.1.4.1.2021.11 iso.3.6.1.4.1.2021.11.1.0 = INTEGER: 1 iso.3.6.1.4.1.2021.11.2.0 = STRING: "systemStats" iso.3.6.1.4.1.2021.11.3.0 = INTEGER: 0 iso.3.6.1.4.1.2021.11.4.0 = INTEGER: 0 iso.3.6.1.4.1.2021.11.5.0 = INTEGER: 0 iso.3.6.1.4.1.2021.11.6.0 = INTEGER: 0 iso.3.6.1.4.1.2021.11.7.0 = INTEGER: 507 iso.3.6.1.4.1.2021.11.8.0 = INTEGER: 259 iso.3.6.1.4.1.2021.11.9.0 = INTEGER: 1 iso.3.6.1.4.1.2021.11.10.0 = INTEGER: 0 iso.3.6.1.4.1.2021.11.11.0 = INTEGER: 99 iso.3.6.1.4.1.2021.11.50.0 = Counter32: 1190528 iso.3.6.1.4.1.2021.11.52.0 = Counter32: 330481 iso.3.6.1.4.1.2021.11.53.0 = Counter32: 136097671 iso.3.6.1.4.1.2021.11.54.0 = Counter32: 0 iso.3.6.1.4.1.2021.11.55.0 = Counter32: 330481 iso.3.6.1.4.1.2021.11.57.0 = Counter32: 0 iso.3.6.1.4.1.2021.11.58.0 = Counter32: 0 iso.3.6.1.4.1.2021.11.59.0 = Counter32: 898868894 iso.3.6.1.4.1.2021.11.60.0 = Counter32: 736460424 so it is showing ticks. but I don't see that mib tree being referenced by doing a poller debug at all? is the solaris support only for perhaps the old snmp (solaris) version which has the entries in a different tree?

1 0

Re: [Observium] Switchmap integration
by Adam Armstrong 15 Aug '12

15 Aug '12

:-D Nikolay Shopik <shopik(a)inblock.ru> wrote: >Adam, just use two words - world domination > >On 15.08.2012 21:37, Adam Armstrong wrote: >> Almost everything that switchmap and netdisco does is on our long-term >> wish list, it's just a matter of finding time and resources to develop >> it all. >> >> We also would prefer to use our own code to generate the configs and >> "smokeping" stats too, as integration is a little one way and suboptimal >> to configure. >> >> adam. >> >> >> On 15/08/2012 17:59, Cushard Michael wrote: >>> Hi, >>> >>> Netdisco does have move features and is an all-around more complete >>> solution, but for me personally it was overkill. Switchmap is just a >>> few simple scripts you schedule and it pulls all the data into csv >>> files, then creates simple html files based on the csv data and has a >>> script for seaching through all the generated pages. It's pretty fast >>> as well. I can do roughly 10000 ports in under 30 minutes with it over >>> some pretty terrible WAN links. I currently have a Switchmap link in >>> our Observium installation as a custom menu item, but tighter >>> integration (of something) would obviously be better. >>> >>> Regards, >>> >>> Michael >>> -----Original Message----- >>> From: observium-bounces(a)observium.org >>> [mailto:observium-bounces@observium.org] On Behalf Of Michael Robbert >>> Sent: Wednesday, August 15, 2012 8:35 AM >>> To: observium(a)observium.org >>> Subject: Re: [Observium] Switchmap integration >>> >> Michael, >> How does Switchmap compare to Netdisco and/or SNMPInfo (The set of >> libraries that Netdisco uses)? Netdisco stores its data in a database >> that could be queried. We use Netdisco and I might be convinced to >> take a look at this kind of integration. I am not a core developer for >> either project so I can't say anything about its acceptance in the >> code base. I do think that it would have to be a plug-in of some type >> not a core feature. Both Swiitchmap and Netdisco are mostly Perl while >> Observium is mostly PHP. I see them as being separate projects, but >> Observium could have a configuration telling it to look at another >> database for more info to display. >> As a side note I'm surprised that I've never heard of Switchmap >> before. We work often with the Network engineers at NCAR/UCAR that >> wrote it. They manage the consortium that is our connection to the >> internet. >> >> Mike Robbert >> >> >> On 8/14/12 9:39 PM, Cushard Michael wrote: >>>>> Hi, >>>>> >>>>> Switchmap (http://sourceforge.net/projects/switchmap) is a most >>>>> excellent little utility written in perl. It goes out to all your >>>>> configured switches utilizing SNMP and pulls all the ARP and MAC >>>>> tables from them, and builds static HTML pages for each switch with >>>>> tables that list port number, speed, duplex, CDP neighbor >>>>> information, MAC address(es), IP address(es), and the fqdn(s) of >>>>> the device(s) seen on the ports at the time of polling. It also >>>>> provides a rudimentary search function that lets you search through >>>>> all the hosts to find which switch port across all your switches a >>>>> particular host was last seen on. This is very handy when you have >>>>> a large number of switches and ports and you need to go "hunting". >>>>> >>>>> >>>>> Now I may be having a UTS moment here, and if I am please correct >>>>> me, but while Observium seems to have something similar, it doesn't >>>>> appear to go to the depth that Switchmap does to actually tell you >>>>> what host(s) are actually connected to what switch ports by >>>>> utilizing the switches ARP and MAC tables. >>>>> >>>>> What would make Observium - and Switchmap - even better would be if >>>>> there was integration between the two, similar to the SmokePing or >>>>> RANCID integration. Switchmap stores all its polled data for each >>>>> switch in separate CSV files, there doesn't seem to be anything >>>>> mysterious going on as far as data storage is concerned. What would >>>>> it take to make a "Switchmap" tab in Observium under a device that >>>>> lists this data, similar to the way RANCID has a "config" tab and >>>>> SmokePing has a "Ping" tab that is integrated into Observium? >>>>> >>>>> Regards, >>>>> >>>>> Michael >>>>> >>>>> _______________________________________________ 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 >> >> _______________________________________________ >> 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