- observium - lists.observium.org

VMware ESXI 6 - False Errors
by Daniel Bray 04 Jan '16

04 Jan '16

We have recently upgraded our test environment from VMware ESXi 5.5 to 6.0. After the upgrade, we are seeing an issue where errors are getting reported on the vmk1 - vmk5 (VMkernel) ports that are connected to distributed switches for SAN access. However, there are no errors (none reported in vSphere nor the SAN). It seems to be a SNMP issue, possibly. We have a near identical production environment that is still running VMware ESXI 5.5, and the similar ESXi hosts are not showing these errors in Observium. This is a completely separate network, with a different Observium server, however both are running similar installs, and patched to the latest Professional version. We have verified using normal snmp commands that everything seems to be working as expected. We have deleted and re-added the ESXi hosts. Issue remains, but only on the vmk ports that are connected to distributed switches. The normal vmnic (vSwitch for management VLAN) show no errors. I've checked the mailing list and Googled this issue for a bit, but not finding anything similar to our issue. Anybody else running Observium with ESXi 6.0 and notice an issue with false Error reports on vmk ports? Any suggestions on what else to try? We have compared everything in the working production environment to the broken test environment, and the only thing we can come up with is that it is an issue with ESXi 6.0 and how it is responding to the Observium SNMP requests. Any guidance or direction would be appreciated. Thanks, -DB

2 1

Frontpage questions
by Jason Lixfeld 04 Jan '16

04 Jan '16

Hi, The default frontpage with an empty array in the config.php ($config['frontpage']['order'] = array(‘’);) still shows some stuff depending on the width of the browser window. I assume a blanked out array would show nothing at all on the front page, so maybe the presence of these artifacts are a bug (see attached images)? Secondly, is it possible to have the front page only be a front page, removing the title bar as well and only showing the contents of the frontage order array? In that case, I would presume the front page would be some sort of a special URL http://$obsevium/frontpage so you don’t get locked out? My intention is to use the frontpage as a super clean NOC screen mounted on a wall. Anyone who saw an alert would login to Observium through their own workstation if they needed a deeper dive. Thanks!

2 1

How to aggregate multiple graphs
by ChanTy 02 Jan '16

02 Jan '16

Hi, Is it possible to aggregate multiple graphs into a single graph ? If yes, please help to provide me the steps. Regards, chanty

1 0

How long until removed ports are deleted?
by Ron Marosko 01 Jan '16

01 Jan '16

Happy New Year, all ya’ll Observium folks. So while everyone else is on holiday, I’m getting to do work. And this brought up an interesting question. I’m mostly removing old/unutilized 10/100/1000 line cards from a Catalyst 6509 today, and in a couple cases installing 10Gb linecards. Right now, I’ve already pulled three linecards from one chassis, and decided to force some discovery/polling of the chassis to see if the old ports will go away… and they haven’t. Doing a “show modules” on the switch confirms the line cards have been removed, but in the device port properties page, Observium still considers them “disabled / down”. Manually running the housekeeping script with the –p option says there are no ports to delete, even when I have manually gone into config.php and set the time duration to be really short (like 1*1 instead of 30*86400). Which brings me to the question… how long does it take for Observium to consider a port or ports “deleted” after it has been removed from the chassis? Thanks, Ron -- Ron Marosko, Jr. . . . . . . . . . . . . . . . . . . . . . . . . . . CCIE No. 4526 (R/S), NN5DX Consulting Network Architect, ATS/Wireless Global Technology Resources,Inc. 1108 West Dickinson Blvd, Suite A Fort Stockton, TX 79735 USA o: +1 432 336 5600 x110 c: +1 432 290 6344 e: <mailto:rmarosko@gtri.com> rmarosko(a)gtri.com pgp pubkey: <http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x778B3B0258AB8B5C> 0x58AB8B5C "To know me is to fly with me." This message contains confidential information and is intended only for the individual named. Please notify the sender immediately by e-mail if you have received this e-mail by mistake and delete this e-mail from your system. Any opinions presented in this email are solely those of the author and do not necessarily represent those of the company. E-mail transmission cannot be guaranteed to be secure or error–free; the sender therefore does not accept liability for any errors or omissions in the contents of this message, which arise as a result of e–mail transmission. Diese Nachricht enthält vertrauliche Informationen und ist nur für den einzelnen, namentlich genannten gedacht. Bitte benachrichtigen Sie den Absender umgehend per E-Mail, wenn Sie diese E-Mail irrtümlich erhalten haben und löschen Sie diese E-Mail aus Ihrem System. Alle Meinungen in dieser E-Mail die vorgestellt werden sind ausschließlich die des Autors und nicht unbedingt die des Unternehmens. E-Mail-Übertragung kann nicht garantiert werden, sicher oder fehlerfrei sind; der Absender übernimmt daher keine Haftung für irgendwelche Fehler oder Auslassungen in den Inhalten dieser Nachricht, die als Ergebnis der E-Mail-Übertragung ergeben. 此消息包含机密信息，并仅用于指定的个人。通过e-mail，请立即通知发件人如果您误收到此电子邮件，并从您的系统中删除此电子邮件。这封电子邮件中提出的任何意见仅仅是那些作者，并不一定代表公司。发送E-mail不能保证是安全或无差错，因此，发件人在该消息的内容，电子邮件传输的结果产生任何错误或遗漏不承担责任。 -- This message has been scanned by E.F.A. Project and is believed to be clean.

1 1

Discovery script crash on build 7383.
by Ron Marosko 31 Dec '15

31 Dec '15

All, Not sure if this is just isolated to me or what… just upgraded to 7383 via svn, and happened to force a discovery on a node trying to troubleshoot why neighbors were no longer being displayed on the ports page on a couple Catalyst 6509s I’m polling, and got the following error: ##### Module Start: bgp-peers ##### No BGP on hostPHP Fatal error: Uncaught TypeError: Argument 1 passed to print_cli_table() must be of the type array, null given, called in /opt/observium/includes/discovery/bgp-peers.inc.php on line 458 and defined in /opt/observium/includes/functions.inc.php:2226 Stack trace: #0 /opt/observium/includes/discovery/bgp-peers.inc.php(458): print_cli_table(NULL, Array) #1 /opt/observium/includes/discovery/functions.inc.php(242): include('/opt/observium/...') #2 /opt/observium/discovery.php(182): discover_device(Array, Array) #3 {main} thrown in /opt/observium/includes/functions.inc.php on line 2226 Anyone else seeing this problem? …Ron -- This message has been scanned by E.F.A. Project and is believed to be clean.

2 4

Re: [Observium] 6880-X Fan Tray Failure
by Sam Hernandez-gill 30 Dec '15

30 Dec '15

Going back through some old mailing lists, didn’t see a reply to this one: http://postman.memetic.org/pipermail/observium/2015-April/010117.html I have the same issue here on Cisco 6880-X’s. Sh Env shows no problem with Fan Tray 1 50V Power Supply status, but observium alerts on the snmp status of the C6880-X Fan Tray 1 – Fan Tray 1 50V Power Supply. [cid:image001.png@01D1419B.9C7B7D10] From any C6880-X itself: fan-tray 1: fan-tray 1 type: C6880-X-FAN fan-tray 1 mode: Mode1 fan-tray 1 fan-fail: OK fan-tray 1 50V Power Supply: OK fan-tray 1 firmware version: 7 I believe this is OID http://tools.cisco.com/Support/SNMP/do/BrowseOID.do?local=en&translate=Tran… Object cevSensorFanTrayStatus OID 1.3.6.1.4.1.9.12.3.1.8.37 Currently with a status alert checker observium will generate an alert for this. So how can I help determine if the switch is really saying the Fan Tray power status Is bad via snmp or if it’s just a problem in how observium is interpreting the response? Thanks, ---------------------------- This message (including any attachments) contains confidential information intended for a specific individual and purpose, and is protected by law. If you are not the intended recipient of this e-mail (even if the e-mail address above is yours), (i) you may not use, copy or retransmit it, (ii) please delete this message and (iii) please notify the sender immediately. Any disclosure, copying, or distribution of this message or the taking of any action based on it, is strictly prohibited. ----------------------------

3 3

TenG port shows up as 1g
by Jeffrey d'Ambly 30 Dec '15

30 Dec '15

[cid:1F6CEEEE-9AE3-4666-8C8B-B90AE68BB8F7] I’m using Observium Enterprise and it seems to think this port is 1g (port speed is displayed as 1Gbps) , clicking on the data button shows the following for int speed, and int hight speed. Are these the right values for a 10g port? fSpeed=>1000000000 ifHighSpeed=>1000

2 1

rabbitmq and ELK stack
by Stefaans Mostert (GBL) 30 Dec '15

30 Dec '15

Morning all. Pretty new to Observium so I need to ask some questions, as I am under a little time pressure. A bump to the correct doc's will also be appreciated. First server I would like to do is an ELK collector. Ubuntu 14.04 Logstash Rabbitmq Observium unix_agent Now how do I integrate this into my Observium monitoring? unix_agent plugin? I want to monitor things like exchanges, queues and such like. Found this(for zabbix) that I can use to get my rabbitmq stats. Thanx a lot, after the first one I will be able to handle it myself :D Steff Stefaans Mostert Contractor: Linux System Administrator Tel: +27 11 459 1804 Fax: +27 11 459 1872 Website:http://www.grindrodbank.co.za Fourth Floor, Grindrod Tower, 8A Protea Place, Sandton PO Box 78011, Sandton, 2146

2 3

Collectd ping plugin graph issue
by Senses0 30 Dec '15

30 Dec '15

Hi, I have collectd configured in Observium and pretty much all the plugins I tried work. There is an issue rendering one graph generated by ping plugin though. Say, I'm measuring latency to "netflix.com". After measurements, the plugin sends following 3 RRD files to Observium - ping-netflix.com.rrd - ping_stddev-netflix.com.rrd - ping_droprate-netflix.com.rrd While observium renders stddev and droprate graphs, it doesn't render "ping-netflix.com.rrd". RRD command produces below error. "RRDTool Output: ERROR: No DS called 'ping' in '/opt/collectd-rrd/server-f04/ping/ping-netflix.com.rrd'" RRD dump has "value" as DS name in *ALL* 3 rrd files (as opposed to "ping"). For some reason, Observium doesn't use the correct DS name for the first file.

1 0

storage wrong on ubuntu 14.04
by Simon Smith 29 Dec '15

29 Dec '15

Hi All, I’m just wondering if any one has the same issue I’m having im monitoring a ubuntu 14.04 container running vpn software for some reason the storage in observium is showing /run, /run/lock, /run/shm, /run/user, /sys/fs/cgroup and completely missing / i use the unix-agent to get extra info out and i can see in there the following <<<df>>> /var/lib/vz/private/105 simfs 10485760 1307760 9178000 13% / none devtmpfs 1048576 4 1048572 1% /dev <<<mounts>>> /dev/simfs / simfs rw,relatime 0 0 so i don’t understand why its not showing the / mount ? i currently also monitor a ubuntu 12.04 container and that shows the storage correctly ? is it the SNMP thats the issue? or any other ideas? regards Simon

2 6