Thanks Adam,
I appreciate you getting back to me so quickly and providing those answers. We'll buy a subscription.
If you ever decide to add dependency support, it would be great for a device to be able to select another device, as it's parent (this could nest well), so we're not inundated by alerts/red devices in the weather map for unreachable devices. Would you consider development payment to add this or is it impractical/too niche?
Kind regards,
*Polaris Networks Support*
*T: *+44 203 397 3535 *W: *www.polarisnetworks.co.uk
*The e-mail and any attachments to it are confidential. Unless you are the intended recipient, you may not use, copy or disclose either the message or any information contained in the message. If you are not the intended recipient, you should delete this e-mail and notify the sender immediately. * *Any views or opinions expressed in this e-mail are those of the sender only, unless otherwise stated. All copyright in any Polaris Networks Limited material in this e-mail is reserved.* *All e-mails, incoming and outgoing, may be recorded by Polaris Networks Limited and monitored for legitimate business purposes.* *Polaris Networks Limited exclude all liability for any loss or damage arising or resulting from the receipt, use or transmission of this e-mail to the fullest extent permitted by law.*
On 4 August 2017 at 18:41, Adam Armstrong adama@memetic.org wrote:
Hi,
- It's not possible to set dependencies in Observium because we're not
sure their behaviour would be as expected for our platform. We poll and monitor a huge number of metrics, and the only way to do dependencies is to either generate unexpected alerts for "dependent" entities or to wait for a subsequent polling cycle to ensure both entities are down. We think it's better to maintain understandable behaviour than to introduce the possibility of confusion and/or delayed notifications.
- Devices are only marked as down if they fail to respond to $retries
number of attempts in $timeout period. These can both be configured, but are set for sane defaults. If you're trying to work around things going down because VPN tunnels are unreliable, don't bother. For the purposes of the poller you don't want the poller running if the tunnel is down. Alerting is based on rules. You can set the rules to require the entity to be "failed" for one or more poller periods before it's notified. This is commonly used for device up/down and especially things like port saturation and cpu usage alerts.
- The subscription edition allows you to supply a custom logo which will
appear at the top of the page in place of the Observium logo, withthe Observium logo moving to the bottom (where the version normally is).
Thanks, adam.
On 04/08/2017 18:16:16, Polaris Networks Support < support@polarisnetworks.co.uk> wrote: Hi,
We're currently using Observium CE 0.16.9.8118 (30th September 2016) and mostly love the system. Not being a native Linux user has kept me on my toes!
Our deployment is monitoring customer premises equipment over hub (us and Observium server) spoke (each client network) VPN tunnels. We're ready to start spending some money with you guys but had a few questions please?
- Is it possible to set device dependencies, so for instance, if a
WAN router fails, all of the devices behind it don't show as down?
- Is it possible for a device to not show as down unless it fails to
respond to multiple polling requests (such as three attempts?) Alerting to be based on these rules?
- Can we brand the login page an UI with our company details, as we
offer the whole system in a MSP role?
Thank you.
Kind regards,
Jon Bennell *Polaris Networks*
*T: *+44 203 397 3535 <020%203397%203535> *W: *www.polarisnetworks.co.uk _______________________________________________ observium mailing list observium@observium.org http://postman.memetic.org/ cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium