UIM OnPrem - SaaS

Hi, when you say cloud servers are these AWS or Azure instances? There is a dedicated probe for API integration in these instances, that may allow you to accomplish your goal without a relay hub. To your question, you certainly can use an additional relay hub in the cloud. Setup and configuration is not challenging, and would follow the same processes that many users follow for monitoring multiple DC's or offices. I'd pay close attention to the tunnel and queue configuration between the two, likewise be aware of how many metrics you are collecting and how often, as that can impact the resources required by the hub.

Disclosure: I work for CA as a solution engineer for UIM

Currently it is not possible to mix UIM on premise with DXI IM (aka UIM SaaS). So in order to cover the cloud servers I’d recommend setting up a secondary hub in the DMZ and then configure the corresponding cloud probes (AWS, Azure…) to point to the cloud based servers which is a simple operation: install the hub in the DMZ, use the tunnel setup wizard to connect it to the primary hub, set up the set of message queues on both sides (DMZ hub and primary one) and you should be good to go. If you encounter any difficulties please refer yourself to docops.ca.com in the UIM installation section and if the information found there is not sufficient please create a service request on support.ca.com
Kind regads,
Gilles

I would suggest also looking at GroundWork Monitor. Our product easily scales and allows you to monitor your entire infrastructure (physical/virtual/cloud/applications). You can visit www.gwos.com to learn more

I would be happy to help. Where are you located?

I need more detail to comment intelligently. If you have access and manage the servers in the cloud no problem. Are you Windows or Unix/Linux?

Transfer it to Agile Operation Architects / SMEs.

Thanks to all of you guys. @Skyline these servers can be anywhere AWS, Azure Public, WAP, VCenter, Openstack, the only thing you need it´s an internet connection to reach the "SaaS Hub". Can´t use the azure and openstack probes cause i need to go deeper on the server metrics collected and the scope of cloud probes it´s on performance, availability and disc space. Can´t monitor logs, services, process, etc.

Yes, totally agree with you @aysha, but with Nagios we can also monitor application service like services running in windows/linux.

I agree that both are good, but I would actually go with a third option: Solarwind's Network Monitoring Software - Network Monitor.
Here are a few reasons:

Simple installation with wizards
Administrators can install by themselves, no professional help needed.
Installation & configuration takes less than 2 hours and you're ready to start monitoring applications.
SAM alerts are based on correlation based events
SAM gives better insight on where the errors have occurred and what triggered the alert.
Users can set up dependencies to suppress alerts and only respond to alerts causing the problem.

On the Nagios side, installation and configuration can be quite complex and requires Linux know how. Also, administrators may need professional services assistance to get Nagios monitoring applications. One thing to keep in mind with Nagios is that alerts are only triggered based on dependencies and the occurrence of the alert or errors is not explained.

As a quick summary I would only say that while Nagios is quite in depth, Solarwinds holds their on in this regard and is more user friendly.

how ever as my friend mentioned below it is fairly easy to upgrade in the cloud for secondary hub space..

Nagios or zabbix

We use SevOne, it not cheap but it seems to do everything we need ATM.

Hi,

Installing a secondary hub is quite straight forward and easy, depending on the network monitoring supplier you are using,

Most deployments have at least one additional hub. For load balancing, enterprise deployments can have several to hundreds. Secondary hubs are typically dedicated servers. They are used:

* To provide failover capability if the primary hub is unavailable.
* In enterprise deployments, to host services and management consoles, such as Unified Management Portal (UMP) and the Alarm Server (nas).
* For data collection and dispersion. In an enterprise deployment, tunnels and queues connect secondary hubs to form a hierarchy that securely transports monitoring data to the primary hub.
Contact your provider and ask them if they can create a secondary or subnet server. Feel free to message me if you need any further help..

Regards

Aysha

Hi

We do around 35000 devices using Saas but not with CA UIM.

We have our own developed software called Sintelligent.”
https://www.youtube.com/playlist?list=PLQCMXdWyI4daPZSHqK8XVC-PZR05MDWCj “

The trick is to make sure that your throughput can be handled by the
hardware your solution runs on and also provides a scalability factor for
growth.