- observium - lists.observium.org

sensors and devices up down false alarms
by Simone Montagnani 10 Jun '25

10 Jun '25

In the last months I'm starting to get some issues I cannot debug well . I have some alert of devices up&down that are not really going down even at ping check. I'm sure that there are no network issues as we are talking about near switch devices, I'm starting to think that I need to debug if our observium is starting to "saturate" and the performance get poor. is there something I can check to find if we need to tune observium ? Or best practices tI can use ? CPU occupation graphs are kind of weird, processors occupation is different between Processors and Detailed Processors graphs : any idea is welcome Best Regards Simone

2 1

Install Script on Debian 12 error with mysql fix
by Bryan Fields 08 Jun '25

08 Jun '25

I just ran the install script on debian 12 as I'm rebuilding our server. I let it install mysql via the script and it died out as mysql was unable to start. I had to manually install the following packages: mariadb-plugin-provider-bzip2 mariadb-plugin-provider-lz4 \ mariadb-plugin-provider-lzma mariadb-plugin-provider-lzo \ mariadb-plugin-provider-snappy The errors showing these were missing: > mariadbd[1605]: 2025-06-08 5:48:20 0 [ERROR] /usr/sbin/mariadbd: unknown variable 'provider_bzip2=force_plus_permanent' > mariadbd[1605]: 2025-06-08 5:48:20 0 [ERROR] /usr/sbin/mariadbd: unknown variable 'provider_lz4=force_plus_permanent' > mariadbd[1605]: 2025-06-08 5:48:20 0 [ERROR] /usr/sbin/mariadbd: unknown variable 'provider_lzma=force_plus_permanent' > mariadbd[1605]: 2025-06-08 5:48:20 0 [ERROR] /usr/sbin/mariadbd: unknown variable 'provider_lzo=force_plus_permanent' > mariadbd[1605]: 2025-06-08 5:48:20 0 [ERROR] /usr/sbin/mariadbd: unknown variable 'provider_snappy=force_plus_permanent' > mariadbd[1605]: 2025-06-08 5:48:20 0 [ERROR] Aborting > systemd[1]: mariadb.service: Main process exited, code=exited, status=7/NOTRUNNING Once I did this, it came up and works. -- Bryan Fields 727-409-1194 - Voice http://bryanfields.net

1 0

phantom sensor and status alerts
by Tony Guadagno 30 May '25

30 May '25

Hi, I am running pro version [cid:image001.png@01DBD0A7.1BDDACF0] today, I saw there were a couple of status alerts and a sensor alert...however, when I click on these links, those alerts are not listed, only the warning is listed.... I think this was working in the past....can someone explain why it indicates sensor and status alerts but does not list them anywhere? Thanks in advance [cid:image002.png@01DBD0A7.1BDDACF0]

2 2

RAM needed per number of ports
by Arnoldo Vidal Bravo 27 May '25

27 May '25

To update a server for Observium, we want to estimate the amount of RAM needed based on the number of ports. We have 30,000 ports on 650 devices. 28,000 ports are ignored (user only). We're interested in monitoring 2,000 ports. Can you confirm if 8GB of RAM will be sufficient? Or should we consider more RAM? -- Saluda cordialmente

1 0

New Implementation
by rudym88＠hotmail.com 21 May '25

21 May '25

Good day, We are considering purchasing Observium, either the Professional or Enterprise, and would like to know if it will meet our specific requirements. 1 - Will we be able to monitor the following vendors: Sonicwall, APC, Granstream (switches, PBX), Ubiquity, Netgear switches, Dell, servers, and workstations? 2 - Due to some required pro-activity monitoring, we use syslog. Can Observium handle syslog. (e.g. we need to be notified as soon as a port goes down instead of waiting for the polling cycle) 3 - We need to have the system send an alert to our ticketing system. Does Odservium send one alert per email, or does it group all the alerts into a single email? Looking forward to a reply to move forward. Thanks Rudy

2 1

Removing questions in Observium forum/personal info
by scott.brucher＠nelsongp.com 17 May '25

17 May '25

I asked a question with my work email and of course forgot to remove my signature. How do I remove all my information? Getting a lot of BOT and sales calls.

4 3

Missing include /opt/observium/includes/alerting/syscontact.inc.php
by Tony Guadagno 12 May '25

12 May '25

Hi, today I noticed that my syscontact email is not getting sent on alerts. Other contacts get the emails but not the syscontact. I used the test_alert script and got this output: ./test_alert.php -c 1 Observium 25.5.14009 Test Alert Notification o Missing include /opt/observium/includes/alerting/syscontact.inc.php when I look, this file is not present * I am using Observium 25.5.14009 (stable)<https://www.observium.org/> Anyone know whats going on here? Tony Guadagno O +1 585 577 1003 C +1 585 703 6700 E tonyg(a)guadagnoconsulting.com<mailto:tonyg@guadagnoconsulting.com> [cid:image001.jpg@01D84DD6.FC9912E0]

1 0

Migration went wrong / Windows SNMP Service detection not working
by IT 09 May '25

09 May '25

Hello, i am quite new to the administration of Observium. The old admin is out of reach and gone. I tried the migration process like described. Old Server: Ubuntu 22.04.5 LTS with Observium 25.2.13882 (stable) New Server: Ubuntu 24.04.2 LTS with Observium 25.4.13987 (stable) The Migration seems to be not 100%. Problems: 1. Only 10 of 43 Probes have been migrated 2. After re-adding some check_ping probes I stumbled across the folder /usr/lib/nagios/plugins The folder was empty Checked the old Server; full folder Copied the hole folder to the new Server. Check_ping now possible and working 1. Check_snmp_win is still not possible. Started the old server; still able to check the Windows services How can I check, what's wrong with my new Server and the Windows Service Check? GVB

2 1

Proxmox support
by Marco Bungalski 29 Apr '25

29 Apr '25

Is there a plan for Observium to support Proxmox? This also includes the display of individual VMs, as is done with VMware. Best Regards, Marco

5 4

Wrapper discovery misbehaving
by Milton Ngan 18 Apr '25

18 Apr '25

I am running multiple pollers, and we just noticed that the discovery of new hosts is not doing what we would expect when trying to partition the workload. Instead of doing a third of the new hosts, it does a third of all the hosts, essentially doing a full discovery, which is causing some our PDU’s to get overloaded, since they are being hit by 3 parallel discoveries and also possibly a poll at the same time. /opt/observium/observium-wrapper discovery --host new -i 3 -n 0 If I remove the -i and -n options, then it does what I would expect, although it is not distributed, which isn’t a problem because the number of new devices should be pretty small. So the --host and the -i and -n options appear to be mutually exclusive, which isn’t stated in the help at all.

2 1