Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!


Observium or a custom solution
New on LowEndTalk? Please Register and read our Community Rules.

All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.

Observium or a custom solution

jhjh Member
edited June 2014 in General

I am looking for a software product I can host myself that is either free or very reasonably priced, to help detect server problems before they cause serious issues, and to alert both me and the end user. The user should be able to log in and see their servers in a read-only manner, and I should be able to configure servers, users and alerts.

The system should play nice with CentOS and cPanel, and should be very easy to install on the agent servers. It should be bug-free and resource-efficient.

To those who have used Observium, does it tick all of these boxes, and do you think it would be suitable? The alternative I'm considering is to write my own, similar to NodeQuery.

Comments

  • InfinityInfinity Member, Host Rep
    edited June 2014

    @jhadley said:
    I am looking for a software product I can host myself that is either free or very reasonably priced, to help detect server problems before they cause serious issues, and to alert both me and the end user. The user should be able to log in and see their servers in a read-only manner, and I should be able to configure servers, users and alerts.

    The system should play nice with CentOS and cPanel, and should be very easy to install on the agent servers. It should be bug-free and resource-efficient.

    To those who have used Observium, does it tick all of these boxes, and do you think it would be suitable? The alternative I'm considering is to write my own, similar to NodeQuery.

    It depends what sort of problems you are referring to, if you can report those problems over SNMP then Observium would be ideal - it's not terribly user friendly for clients but it works.

  • jhjh Member
    edited June 2014

    The most common/important problems are simply running out of memory or disk space. I don't want to have to faff with SNMP every time a customer joins or leaves, so I was hoping that the monitoring system could do it all internally (similar to NQ).

  • Observium would tick all your boxes, if you don't want to run SNMP then there is some sort of agent available (haven't really looked into it). The paid version includes alerts etc.

  • wojonswojons Member

    @jhadley said:
    I am looking for a software product I can host myself that is either free or very reasonably priced, to help detect server problems before they cause serious issues, and to alert both me and the end user. The user should be able to log in and see their servers in a read-only manner, and I should be able to configure servers, users and alerts.

    The system should play nice with CentOS and cPanel, and should be very easy to install on the agent servers. It should be bug-free and resource-efficient.

    To those who have used Observium, does it tick all of these boxes, and do you think it would be suitable? The alternative I'm considering is to write my own, similar to NodeQuery.

    It really depends on how much stuff you really want to intrergrate after spending some time intergrating my own SAAS based monitoring system. Just make sure to scope out the project before you start.

  • jhjh Member

    Thanks @INIZ. Maybe someone else could comment on the agent.

  • jhjh Member

    wojons said: It really depends on how much stuff you really want to intrergrate after spending some time intergrating my own SAAS based monitoring system. Just make sure to scope out the project before you start.

    If I were to implement something of my own, it would start by looking very similar to NQ and build up from there. The most common issues are very simple and easy to detect (e.g. nearly out of disk space) so that would be iteration 1.

  • I'm running observium alongside with the UNIX agent, it's just an 'addon' to the statistics SNMPd gets, such as sensors, apps (nginx, mysql stats); etc.

    I don't think you can run the UNIX agent without SNMPd on clients

  • gbshousegbshouse Member, Host Rep
    edited June 2014

    .

  • mikhomikho Member, Host Rep

    The free version of PRTG (by Paessler) can do most of these things, you would need a Windows machine to run it. Can monitor both Win & Nix machines.
    I think the free version allows you to have 10 sensors if that is enough?
    It's a very nice software but costly if you go above the free amount of sensors.

  • You could use Nagios for stuff like memory and disk space, while working on yournown solution?

Sign In or Register to comment.