- observium - lists.observium.org

python-wrapper loop?
by John Klasa 30 Jun '17

30 Jun '17

Hi I've upgraded to this version: ./discovery.php -V Observium 17.5.8580 It seemed to be a great update but then MySQL died and I found this in the logs: [2017/06/29 15:05:20 +0200] poller-wrapper.py(29309): /opt/observium/poller-wrapper.py: URGENT: poller-wrapper not started because already running 34 processes, load average (5min) 37.17 And it kept increasing upton load over 70 where I killed all python, CROND and for some reason A LOT of ps processes. An example of processes below. 26133 ? D 0:00 ps -ww -o ppid,uid,args -p 23434 26331 ? R 0:00 ps -ww -o ppid,uid,args -p 2981 26354 ? R 0:00 ps -ww -o ppid,uid,args -p 10439 26355 ? R 0:00 ps -ww -o ppid,uid,args -p 11755 26356 ? R 0:00 ps -ww -o ppid,uid,args -p 5772 26357 ? R 0:00 ps -ww -o ppid,uid,args -p 2624 26360 ? R 0:00 ps -ww -o ppid,uid,args -p 19699 846 purify 20 0 342884 70760 1636 S 1.0 1.9 0:33.02 python 4424 purify 20 0 354536 11156 1632 S 1.0 0.3 1:27.40 python 4739 purify 20 0 356676 11832 1632 S 1.0 0.3 1:36.27 python 9718 purify 20 0 349728 10108 1636 S 1.0 0.3 1:03.00 python 20551 purify 20 0 361020 12280 1632 S 1.0 0.3 2:01.39 python 31260 purify 20 0 348144 76392 1636 S 1.0 2.0 0:56.62 python 402 purify 20 0 345244 33752 1636 S 0.7 0.9 0:43.67 python 1315 purify 20 0 343148 59904 1632 S 0.7 1.6 0:35.71 python 3219 purify 20 0 341844 47340 1632 S 0.7 1.3 0:28.92 python 3298 purify 20 0 353204 11376 1632 S 0.7 0.3 1:19.76 python Anyone know whats going on? I see this when I try to run the debug output Your PHP version is too old (5.4.16), | | functionality may be broken. Please update your PHP! | | Currently recommended version(s): 7.0.x or 5.6.x I guess I should upgrade PHP first. Regards -John

2 2

Process Level Alerts in Observium
by Kaustubh Parthasarthy, CLSA 30 Jun '17

30 Jun '17

Hi Team, First, I would like to subscribe to the Observium mailing list. Second, i am using the Community Edition of Observium. Following are the details: [cid:image001.png@01D2F0C3.971FFA70] I have setup some alert checks to capture the high CPU alerts. I was also successful to capture some alert. PFA the results. I would like to know, if we can also show, which are the top 10 processes, which are running and as a result the utilization is going high. If I can get the output of the top 10 or top 5 processes on the same mail which is attached, this would assist the support teams to troubleshoot. Could you please assist on this as to how can I setup ? Thank you and Regards Kaustubh The content of this communication is intended for the recipient and is subject to CLSA Legal and Regulatory Notices. These can be viewed at https://www.clsa.com/disclaimer.html or sent to you upon request. CLSA is ISO14001 certified and committed to reducing environmental impact.

2 1

getting multiple random down events Device status changed to Down (ping)
by Ron Culler 29 Jun '17

29 Jun '17

Started getting multiple events "Device status changed to Down (ping) " over the last few weeks with no major increase in number of devices. Issue started on Cento 6.9 install with PHP 5.4, MySQL 5.1 (260 devices) HW - dual - quad core xeon CPUs, 24 GB Ram, RRD, MySQL, OS - with Observium install all on separate SSD drives Observium SVN install 17.6.8672 (rolling) Upgraded PHP and MySQL to PHP70 with opcode caching enabled, MySQL 5.7.18 - no change Spun up a fresh install Centos 7.3, PHP70 with opcode caching enabled, MySQL 5.7.18 (install per docs) HW - dual - quad core xeon CPUs, 24 GB Ram, RRD, MySQL, OS - with Observium install all on separate SSD drives Observium SVN install 17.6.8672 (rolling) $config['snmp']['max-rep'] = TRUE; Currently only 139 devices added To eliminate DNS related issues I have added the devices to /etc/hosts and forced resolution to use it first. Cron has /opt/observium/observium-wrapper poller -w 5 running every 5 min. The devices are not down and if I manually poll the device with a "Device status changed to Down (ping) " notice it responds normally and comes right back up. Was wondering if anyone has seen any behavior like this with this build or before. Also is there a way to skip ping during the poller run and just do snmp, and if so is it something that I can enable globally? Thanks in advance Ron Culler

1 0

wget - Read Error in Headers (Connection Reset by Peer)
by Muhammed Murat Tutar 28 Jun '17

28 Jun '17

I want to install Observium (observium.org) on my virtual Ubuntu Server but when I try to download files, I get a read error. wget http://www.observium.org/observium-community-latest.tar.gz --2017-06-28 10:53:16-- http://www.observium.org/observium-community-latest.tar.gz Resolving www.observium.org (www.observium.org)... 37.187.79.83 Connecting to www.observium.org (www.observium.org)|37.187.79.83|:80... connected. HTTP request sent, awaiting response... Read error (Connection reset by peer) in headers. Retrying. I couldn't find any solution. I use static IPv4 address and DNS servers. Ubuntu 16.04 LTS works on VCENTER Virtual Machine Version: 11 and Network Adapter is VM Network. Any solutions? Thanks.

3 2

Re: [Observium] SNMP Device without sysObjectId.0
by Oberreiter Andreas 28 Jun '17

28 Jun '17

I've noticed a typing error Snmpget -v 1 -c public portal.nadlinger.at .1.3.6.1.4.1.14848.2.1.1.1.0 -----Ursprüngliche Nachricht----- Von: observium [mailto:observium-bounces@observium.org] Im Auftrag von Oberreiter Andreas Gesendet: Donnerstag, 4. Mai 2017 10:53 An: Observium Public Support <observium(a)observium.org> Betreff: Re: [Observium] SNMP Device without sysObjectId.0 Snmpget -v 1 -c public portail.nadlinger.at .1.3.6.1.4.1.14848.2.1.1.1.0 Should answer SNMPv2-SMI::enterprises.14848.2.1.1.1.0 = STRING: "Version 1.62" -----Ursprüngliche Nachricht----- Von: observium [mailto:observium-bounces@observium.org] Im Auftrag von Mike Stupalov Gesendet: Donnerstag, 4. Mai 2017 10:00 An: Observium Public Support <observium(a)observium.org> Betreff: Re: [Observium] SNMP Device without sysObjectId.0 Hi, can you privide temporary snmp access to this device? I will do this faster :) Oberreiter Andreas wrote: > I've added print_message lines to see if my code is called. > > When I try another device e.g. a zyxel device I can see my code is executed. > > But when I try the device in question the discovery code isn't executed. > > > -----Ursprüngliche Nachricht----- > Von: observium [mailto:observium-bounces@observium.org] Im Auftrag von > Mike Stupalov > Gesendet: Mittwoch, 3. Mai 2017 17:21 > An: Observium Public Support <observium(a)observium.org> > Betreff: Re: [Observium] SNMP Device without sysObjectId.0 > > This device also not have sysDescr.0 too? > > You also should add os detecting file (from > $config['os'][$os]['discovery_os']): > includes/discovery/os/messpc.inc.php > > with content like this: > > <?php > > if (!$os) > { > // SNMPv2-SMI::enterprises.14848.2.1.1.1.0 = STRING: "Version 1.62" > $data = snmp_get($device, '.1.3.6.1.4.1.14848.2.1.1.1.0', '-Osqnv'); > if (strlen($data)) > { > $os = 'messpc'; > } > } > > // EOF > > And better os name: messpc-env > > Oberreiter Andreas wrote: >> Hi, >> >> The Device is from http://www.messpc.de >> >> I've tried to add the following lines to >> includes/definitions/os.inc.php // MessPC $os = "messpc"; >> $config['os'][$os]['text'] = "MessPC Etherbox"; >> $config['os'][$os]['vendor'] = "MessPC"; >> $config['os'][$os]['type'] = "environment"; >> $config['os'][$os]['discovery_os'] = "messpc"; >> $config['os'][$os]['snmpable'][] = ".1.3.6.1.4.1.14848.2.1.1.1.0"; >> $config['os'][$os]['mibs'][] = "ETHERBOX-MIB"; >> >> I've put the ETHERBIX-MIB file in mibs/messpc/ETHERBOX-MIB >> >> and restarted the Webserver. Then I tried to add the device again no change. >> >> I can do a snmpwalk -v 1 -c public IP .1.3.6.1.4.1.14848 >> >> The result of >> snmpget -v 1 -c public 172.30.7.2 .1.3.6.1.4.1.14848.2.1.1.1.0 is >> SNMPv2-SMI::enterprises.14848.2.1.1.1.0 = STRING: "Version 1.62" >> >> What did I wrong? >> I'm on the latest commercial Version 17.4.8476 >> >> Kind Regards >> Andreas >> >> -----Ursprüngliche Nachricht----- >> Von: observium [mailto:observium-bounces@observium.org] Im Auftrag >> von Mike Stupalov >> Gesendet: Freitag, 28. April 2017 15:02 >> An: Observium Public Support <observium(a)observium.org> >> Betreff: Re: [Observium] SNMP Device without sysObjectId.0 >> >> Hi, >> >> seems as another china/indian device? ;) >> >> For valid device adding required one of this OIDs to be exist on device: >> sysObjectID.0 or sysUpTime.0 >> >> If both OIDs not exist for device, use 'snmpable' os param. >> See example for os 'hikvision-cam': >> >> $os = "hikvision-cam"; >> $config['os'][$os]['text'] = "Hikvision Network Camera"; >> $config['os'][$os]['type'] = "video"; >> $config['os'][$os]['vendor'] = "Hikvision"; >> $config['os'][$os]['discovery_os'] = "hikvision"; >> $config['os'][$os]['snmpable'][] = >> ".1.3.6.1.4.1.39165.1.1.0"; // This device complete not have >> sysObjectID.0 or sysUpTime.0 >> $config['os'][$os]['mibs'][] = "HIK-DEVICE-MIB"; >> >> >> Oberreiter Andreas wrote: >>> Hi All, >>> >>> I have a device which doesn't provide the SysObjectId.0. >>> It start right away with the Enterprise entry's: >>> SNMPv2-SMI::enterprises.14848.2.1.1.1.0 = STRING: "Version 1.62" >>> >>> Is there a way to add such an device? The values I'm interested in are in the enterprise part of the mib and could be easily retreated by a custom OID. >>> >>> But without an SysObjectId it seems impossible to add a device. >>> >>> Any Suggestions? >>> >>> Kind regards >>> Andras >>> >>> >>> _______________________________________________ >>> observium mailing list >>> observium(a)observium.org >>> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium > > -- > Mike Stupalov > Observium Limited, http://observium.org > _______________________________________________ > 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 -- Mike Stupalov Observium Limited, http://observium.org _______________________________________________ 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

4 3

rrdtoolx
by austin wonderly 28 Jun '17

28 Jun '17

Hello, What would be the correct way to extract rrd data for a certain period of time (last 7 days for example). I'm trying to extract data for the following graph http://imgur.com/a/Gw9y2, although when I open up the data in excel, I'm not able to get anything of use. I used this command to produce an RRD export first: /opt/observium/scripts/rrdtoolx.py addrra netstats-udp.rrd /tmp/testrrd after which I used: rrdtool dump /tmp/testrrd > /tmp/rrd.xml to get the file into an xml format, and finally I imported the xml into excel. I'm trying to put together a list of time stamps and their appropriate data points, so for example: time1 : 22 errors time2: 10 errors time3: 30 errors .... Would it even be possible to extract this kind of data from the RRD files? Thanks for your help, Austin

2 3

MSSQL database for Professional edition
by Jan Horacek 27 Jun '17

27 Jun '17

Good evening gentlemen, I have an question about possibility run Observium PE on MSSQL database instead of MySQL/Maria DB. we are going to implement observium PE to live production and our developers are wondering if any official MSSQL plugin exists or not. Thanks for reply best Regards Jan

2 2

WAN link monitoring
by Luke Johnson 27 Jun '17

27 Jun '17

Hi, I have looked for previous questions regarding this, but haven't been able to find one, so I apologise if it has been asked/answered before. We have several paths to a remote router that we monitored via it's loopback IP address. Observium does a great job at alerting us when that router is offline, but, I'm unable to find a way to be alerted when we have a single WAN link failure. During a WAN link failure we can still access the remote router via a redundant path, however, we are often not aware that the primary is offline since Observium can still reach the remote router. Apart from checking if utilisation on the link drops below a certain level, is there any other way that I can perform a check and alert on the above scenario? Thanks, Luke

3 2

Question on geolocation dropdown menu
by Ed Iwanski 23 Jun '17

23 Jun '17

We have a largish installation and have locations / devices in every state in the US and every province/territory in Canada. We have been adding all of these to Observium and allowing them to be geolocated based on SNMP location strings (works perfectly, thank you) but found that we are running into an issue where the scrollbar is too tall for many average browser screens. There does not seem to be a way to scroll here so there are some locations that will just not be accessible or seen on this dropdown menu. Is there a way to tune or change this? Also any possibility that the states/provinces/territories can be alphabetically listed on the dropdown menu instead of in seemingly order they were added or random? Screenshot attached, there are more locations at the bottom but that was all I could fit on the screen for the capture. Thanks for any input.

2 2

snmpv3 context support
by Tomasz.Kwiecien＠nordea.com 21 Jun '17

21 Jun '17

Hi, Are you planning to add snmpv3 context support to observium? I can see that in file includes/snmp.inc.php: switch ($device['snmp_version']) { case 'v3': $cmd = ' -v3 -l ' . escapeshellarg($device['snmp_authlevel']); /* NOTE. * For proper work of 'vlan-' context on cisco, it is necessary to add 'match prefix' in snmp-server config --mike * example: snmp-server group MONITOR v3 auth match prefix access SNMP-MONITOR */ $cmd .= ($vlan) ? ' -n "vlan-' . $vlan . '"' : ' -n ""'; // Some devices, like HP, always require option '-n' Observium adds context vlan-$vlan for cisco switches if $vlan variable is defined. Would be great to have option to add device with context during new device definition (like option to choose transport protocol, port timout etc). Best regards, Tomasz Kwiecień

3 2