- observium - lists.observium.org

Historical Data Graph
by Tasnim Tamanna 10 Dec '20

10 Dec '20

Everyday at a certain time, all historical graph data gets deleted. This is happening for all devices except one. Can anyone tell what could be wrong?

2 5

interface description parsing
by Tasnim Tamanna 09 Dec '20

09 Dec '20

Hi All, For interface description parsing, I have copied the whole *port-descr-parser.inc.php *file to *custom-port-descr-parser.inc.php *file, and added below lines in the new file: cust: *cust_name* [*speed*] (*a_note*) {*circuit_id*} Finally, added below lines in *config.php* file: $config['int_transit'] = 1; // Enable Transit Types $config['int_peering'] = 1; // Enable Peering Types $config['int_core'] = 1; // Enable Core Port Types $config['int_customers'] = 1; // Enable Customer Port Parsing $config['port_descr_parser'] = "includes/custom-port-descr-parser.inc.php"; But it's not working. Need help :-(

3 3

Aggregating two Graphs in observium | 22-Nov-2020
by Tasnim Tamanna 08 Dec '20

08 Dec '20

Hi, Can anyone please help me with aggregating two graphs in observium? I have not found any documentation on this. Regards Tasnim Tamanna Dhaka, Bangladesh 8801316814417

4 12

AUTO: Marios Palechoros is out of the office (returning 14/12/2020)
by Marios.P＠crowe.com.cy 07 Dec '20

07 Dec '20

I am out of the office until 14/12/2020. I will have access to my emails, however for urgent matters please contact: George Paroutis: georgep(a)siviriventures.com / 22473186 Note: This is an automated response to your message "[Observium] Same Subnet devices" sent on 7/12/2020 6:30:34 μμ. This is the only notification you will receive while this person is away.

1 0

Upgrade Fail
by Nash, Bob 07 Dec '20

07 Dec '20

Upgraded our subscription version of Observium to the latest release that was displayed on the webpage and after upgrading, NO devices are being monitored/discovered. This was working perfectly fine prior to the upgrade. Need assistance as soon as possible. Sincerely, Bob

2 1

Observium not working after upgrade
by Nash, Bob 07 Dec '20

07 Dec '20

Just upgraded to current subscription version and Observium is no longer working. Need help!

2 2

cisco ip local pool usage monitoring
by Matteo Sgalaberni 07 Dec '20

07 Dec '20

Hello, I want to monitor the ip local pool usage of a cisco device in Observium. I found the respective mibs already present on Observium source code tree: ./mibs/cisco/CISCO-IP-LOCAL-POOL-MIB ./mibs/cisco/CISCO-IP-ADDRESS-POOL-MIB Is there a way to add these graphs in order to monitor ip local pool usage of the device monitored? Thanks Matteo

1 0

Bug in alert email for BGP AFI/SAFI?
by Johann Mallet 30 Nov '20

30 Nov '20

Hello, I have tried the BGP AFI/SAFI alert and seems to have a bug in email/alert/entity when alert/recovery was sent. The alert name in subject and alert/entity in tab seems to be missing. Also, the device line seems in double (in all of my alerts). https://i.imgur.com/91xVpLI.png https://i.imgur.com/CZJANWz.png Johann

2 1

TLS 1.3 and Opsgenie
by Adam Armstrong 28 Nov '20

28 Nov '20

Hi Guys, A number of services are moving their APIs to require TLS 1.3 because of a vulnerability in TLS 1.2, which means you need to be running pretty recent versions of PHP and OpenSSL to access them. PHP needs to be 7.4. Opsgenie is already requiring TLS 1.3, so if you're using them for your alerting, please make sure you're running a recent enough version of PHP for the API calls to function. You can test this on your system with the test_alert.php CLI script in the Observium root directory. In general, Observium is a fairly demanding PHP application so it's good practice to try to keep PHP as up to date as possible. We've seen some very nice performance improvements over the past few years, and 8.0 is going to bring even more. For this reason we don't recommend running Observium on distributions like RHEL/CentOS which hold old versions for as long as possible. Thanks, Adam.

1 0

Dashboard based on Port Description ?
by Pieter De Wit 25 Nov '20

25 Nov '20

Hey Guys, Is there a way to create a dashboard that will include all graphs of ports that match a certain description ? Dynamically that is :) Cheers, Pieter

3 5

CE No RRD Updates Since Upgrade to 20.9
by Scott Driemeier-Showers 24 Nov '20

24 Nov '20

I run two instances of Observium. Professional is installed at the office and CE is installed in my home lab. I followed the steps to upgrade CE (I know, everyone says that) in my lab from 19.8 to 20.9 on November 8 and none of the graphs have updated since. I have rebooted, checked permissions, verified all expected services are running, etc. Reviewing the event log in the Web UI and the logs on the server all show activity. The timestamps of the individual RRD files are updating every time poller runs - just nothing happening in the UI. Can someone help me think of anything else to check? CentOS 7.9 MariaDB 10.3.27-1 PHP 7.2.34 (OPcache Enabled) Thanks, Scott

1 0

OSPF Neighbour ID isn't always the IP address
by Pieter De Wit 23 Nov '20

23 Nov '20

Hey Guys, I noted this morning when I added a device that Observium tries to auto discover the OSPF Neighbours - this is awesome, but won’t work in our env. (Firewalls and IDs) ##### Module Start: neighbours ##### o LLDP-MIB o CISCO-CDP-MIB o OSPF-MIB o Trying to discover host 10.x.x.255 (10.x.x.255) through OSPF (ospf) o Trying to discover host 10.y.y.255 (10.y.y.255) through OSPF (ospf) o Trying to discover host 10.y.y.0 (10.y.y.0) through OSPF (ospf) We force our ID’s to be higher or lower depending on what we need. I tried to see if the real IP address is in the SNMP dump, but my SNMP magic is failing :( Cheers, Pieter

2 1

Limit Radius user to specific devices/ports
by Joelly, Alexander 22 Nov '20

22 Nov '20

I’m authenticating against RSA , set the groupmemberattr Filter-ID and some auth_radius_groups with different privilege levels. What i could not figure out is how i can limit one Filter-ID to just a specific device or port. In the GUI it’s possible to configure some Roles, once added they cannot be removed anymore via the GUI? Thanks, Alex

2 1

STP in Observium
by Torstein Eide 22 Nov '20

22 Nov '20

Hi Observium On Jira multiple people ask for status on adding Old/RSTP/MSTP Spanning tree protocol. So it sounds like it is a much sought after feature. https://jira.observium.org/browse/OBS-3509 https://jira.observium.org/browse/OBS-1853 In OBS-3509 I have added proposals on how the database and a layout for the frontend. Based on that Observium has excellent handling of port status and layer 3 features. What is stopping us from implementing it? -- Torstein Eide Torsteine(a)gmail.com

3 2

Re: [Observium] Graph Data Missing after Upgrade RAM
by Aulia Rahman 18 Nov '20

18 Nov '20

Dear Adam, Sorry to resend because I didn't get the answer, thanks before. > It’s likely that you need more CPU, rather than more RAM. Ending up in these out of memory scenarios are usually because the CPU was too slow and too many processes piled up. The processor full if there's a poll is running after that it will normal again. I've follow the tutorial to using persistent ramdisk https://docs.observium.org/persistent_ramdisk/ If ramdisk, RAM and swap are full, does the RRD still written ? Because in that condition, the graph not written and there's an error when access the gui: *"DB Error 2002: Can't connect to local MySQL server through socket '/var/lib/mysql/mysql.sock' (111)*" after restart the DB service it will normal again. please advice. >If you can’t see historic data, it means the RRD containing the historical data has been somehow overwritten. Observium won’t do this by itself, so somehow you managed to delete or move the original RRDs. >You can restore the RRDs from backup, and it should write to these. I can manual restore from backup of the backup ramdisk, thanks. Because when I try to manual extract the ramdisk data from LZOP first backup file it end with error. > Though, if the RRDs have somehow disappeared, I’d check to make sure you don’t have some form of filesystem/storage corruption, because the two might be linked. After investigate, when the shutdown process, the ramdisk didn't write the latest data before the server turn off. Because I didn't see any log about Ramdisk write data to hardisk. How to achieve it ? because in the document didn't explain detail about it. When startup process it succesfull load the data from ramdsk via rc.local script but the data loaded is corrupt, I can confrm this because when I try to manual extract from LZOP compression end with error, only some graph can be restored. Please advice. > BTW, please don’t send images to the mailing list. Sorry I've resend the mail but forgot to cancel the first one. Demikian yang dapat kami sampaikan, terimakasih. Regards, Aulia Rahman On Thu, Nov 12, 2020 at 3:36 PM Adam Armstrong <adama(a)observium.org> wrote: > It’s likely that you need more CPU, rather than more RAM. Ending up in > these out of memory scenarios are usually because the CPU was too slow and > too many processes piled up. > > > > If you can’t see historic data, it means the RRD containing the historical > data has been somehow overwritten. Observium won’t do this by itself, so > somehow you managed to delete or move the original RRDs. > > > > You can restore the RRDs from backup, and it should write to these. > > > > Though, if the RRDs have somehow disappeared, I’d check to make sure you > don’t have some form of filesystem/storage corruption, because the two > might be linked. > > > > BTW, please don’t send images to the mailing list. > > > > Thanks, > > Adam. > > > > > > Demikian yang dapat kami sampaikan, terimakasih. > > > > Regards, > > Aulia Rahman > > > > >

2 1

New OS support
by Craig Smith 18 Nov '20

18 Nov '20

Hello, I'm looking for someone that can add support for a couple of Telco Systems (www.telco.com) models (T-Marc 3308 & T-Marc 3348S) to Observium. I can provide MIBs and access remotely to the devices for development. My company is willing to pay a resonable price for a set of agreed features being supported, and for the work to be contributed to Observium.

2 1

Ignore device through cli
by Svensson Fredrik A 17 Nov '20

17 Nov '20

Hello! I´d like another system to set status on a device to ignore in observium, is there a cli-command for that? Regards Fredrik

2 1

No pages after CE upgrade
by Arrigo Savio 17 Nov '20

17 Nov '20

Hi all. I update Community Edition to Version 20.9. I followed steps reported in page https://docs.observium.org/updating/ <https://docs.observium.org/updating/> cd /opt mv observium observium_old}} wget -Oobservium-community-latest.tar.gz https://www.observium.org/observium-community-latest.tar.gz <https://www.observium.org/observium-community-latest.tar.gz> tar zxvf observium-community-latest.tar.gz}} mv /opt/observium_old/rrd observium/ mv /opt/observium_old/logs observium/}} mv /opt/observium_old/config.php observium/ I update DB schema and PHP scripts (discovery and poller) work correcty. I can see homepage, but if I try to enter devices section https://observium2.qcom.it/device, <https://observium2.qcom.it/device,> I receive a blank page with "File not found" message. Any ideas? Thanks. Arrigo

2 1

WG: question and a problem in Check alert
by Tibi, Saleh 13 Nov '20

13 Nov '20

Hi guys, our instance: observium 20.10.10799(rolling) i have two issues (a question and a problem ) in the subject Alert Checkers. The question: Is it Possible to configure my alert Checkers that test two different sensors (different entities ) I have two sensor - flowrate Sensor - Valve Sensor For example like this: Require all conditions: (flowrate Sensor)sensor_value eq 0 (Valve Sensor)sensor_value eq 0 What I saw in the documentation that we can check only one entity at a time. The Problem: Steps: 1- Creating new alert Checker 2- filtering devices or sensors in Entity Association Ruleset (for example: Sensor Group in WaterSensor) 3- save changes (all sensor will be listed ) 4- rebuild 5- all sensor appearing in status unknown 6- wait a 3 or 4 minutes all sensors will disappear any advices or hint ? thx in advanced an have a nice Day Saleh Tibi Mit freundlichen Grüßen Saleh Tibi ---------------------------------------------- Ing: Saleh Tibi (Herr) IT-Spezialist für Maschinensäle und Überwachung Zentrum für Informations- und Medientechnologie (ZIM)|Raum 25.41.02.27 Heinrich-Heine-Universitaet Duesseldorf Universitaetsstrasse 1 40225 Duesseldorf Email: Tibi.saleh(a)uni-duesseldorf.de<mailto:Tibi.saleh@uni-duesseldorf.de> Phone: +49-(0)211-81-14051

2 6

Syslog view sort order by date
by Stempien, Dave 12 Nov '20

12 Nov '20

Hello, Is it possible to sort the syslog messages in both the syslog view and in the dashboard widget by date? The event log shows newest entries on top but syslog shows oldest (which is 30 days – the number of days in our case that we retain syslog data). Thanks, -- David Stempien, University of Rochester University IT - Core Technology Services Tel: 585.276.6957 Email: david.stempien(a)rochester.edu<mailto:david.stempien@rochester.edu>

1 0

Polling partitioning / distributed polling
by Brad Bendy 12 Nov '20

12 Nov '20

Hi, I've set this up per what I can find in the docs. Master server has RRDcached running, MySQL and web server. I have another server acting as poller server. I go to that server and add hosts, the host add find. All the graphs are failing and on the local server i t shows nothing in the RRD dir (which is right I would think), on the remote hosts it shows RRD files but when I do a debug I see "file already exists" and it does on the remote server. Looking at the rrd.journal file I see updates being issued and see no errors. On the polling server I have added: $config['rrdcached'] = "x.x.x.x:42217"; $config['rrd']['no_local'] = TRUE; And updated cron to: */5 * * * * observium /opt/observium/observium-wrapper -i 1 -n 0 discovery --host new >> /dev/null 2>&1 # Run multithreaded poller wrapper every 5 minutes */5 * * * * observium /opt/observium/observium-wrapper -i 1 -n 0 poller >> /dev/null 2>&1 As I have one poller server and this is the first one. I see no other errors in other logs but all graphs show "error drawing graph". Is there something else in my config I am missing to make this work? Thanks

4 9

support for PacketLight - in-progress case OBS-3176
by Henrik Aarfeldt 12 Nov '20

12 Nov '20

Hi, There is an in-progress case with support for PacketLight DWDM boxes. It seems to be from December 2019 and with Mike Stupalov as assignee. https://jira.observium.org/browse/OBS-3176?jql=text%20~%20%22packetlight%22 Any news regarding when this will be working ? We have some new PacletLight boxes we would like to monitor, and there dosen't seem to be any sensors available. TIA. -- Henrik Aarfeldt Region Hovedstaden Center for IT, Medico & Telefoni (CIMT) Borgervænget 7, 3. sal 2100 København Ø ________________________________ Region Hovedstaden anvender de personoplysninger, du giver os i forbindelse med din henvendelse. Du kan læse mere om formålet med anvendelsen samt dine rettigheder på vores hjemmeside: www.regionh.dk/persondatapolitik

3 2

Graph Data Missing after Upgrade RAM
by Aulia Rahman 11 Nov '20

11 Nov '20

Dear All, I'm using Observium 20.1.10233 (stable) Enterprise version. Observium sometime error on database when access from web Image1.jpeg At that time the graph stop add new data. there's a notification in console like picture below, so I need to restart the database service. image2.png Access server and try to see HTOP image3.png Then check the log image4.png After restart database service, the graph starts to add new data again. Then, I upgraded the server RAM from 12 GB to 24 GB on 9 November 2020 at night. This server uses ramdisk configuration on SSD. After the server booted up the graph didn't add new data until today 11 November 2020 at 00:00 the graph started to add new data again. But, all previous data graphs on all devices before 11 November 2020 are missing. Here's the screenshot for the missing data for example on one device after upgrade. image5.png Here's the HTOP from the observium server after upgrade image6.png I've restore to date 8 and 7 November but the all graph show "Draw Graph Error" image7.jpeg & image8.jpeg When check the rrd folder from the server restore, It just only show a few device folder. I assume that folder still hold on ramdisk, but if I see log for ramdisk it sync every 2-5 minutes to hardisk. Please tell me step-by-step how to troubleshoot & recover it. All image you can see on this link: https://drive.google.com/drive/folders/1u8sx3_a_qJN9Kpr_gmgEND3MsV4Vd0Vj?us… Demikian yang dapat kami sampaikan, terimakasih. Regards, Aulia Rahman

1 0

Trial version for a period of 30 days, “3data”company.
by Alexander Agoshkov 10 Nov '20

10 Nov '20

Dear colleagues, Being the representatives of one of the largest data centers in Russia – “3data”company, we are considering the use of Observium system to monitor our network. At the moment, the testing of the monitoring of our network with the help of the Community version has given us completely positive results. We would like to transfer to the paid Professional version, but before it we need to have some additional testing on the Professional version. Therefore, we would like to ask you for a trial version for a period of 30 days. We also would like to express our gratitude for your creating and developing of the monitoring system and we are looking forward to our further cooperation. Yours faithfully, Agoshkov Alexander | IT support engineer | 3data Premium Data Centers +7 (495) 800-18-00 | + 7 (963) 762-48-04 | aag(a)3data.ru<mailto:aag@3data.ru> | 3data.ru<http://3data.ru/>

2 1

Rancid git with remote pollers
by Brad Bendy 09 Nov '20

09 Nov '20

Hi, I know this is slightly off topic but was wondering if anyone has ran into this. We have rancid on our remote pollers working and grabbing configs. Now on a single Observium instance you just enable it in the config and the integration works spot on as the SVN/GIT is local. Has anyone got this to work where they store the configs locally on the remote poller and then push a copy to the master observium box so they show up in the GUI? Or are you using a central GIT server on the master and just the proper settings in the rancid.conf file? Thanks

2 1