Best Practices for Monitoring 2U Rack Servers with SolarWinds

luciaonmatteo

Hi everyone,

I'm currently working on optimizing our server monitoring for a 2U rack setup, and I'm looking for some advice on the best practices when using SolarWinds to monitor 2U Rack Servers. We've been using it for a while, but I want to make sure we're making the most of it.

Here are some specific things I'm looking for help with:

1. Temperature Monitoring: How do you set up SolarWinds to monitor the temperature and ensure optimal airflow? Any tips on setting thresholds for alerts?
2. Resource Utilization: Best ways to track CPU, memory, and disk usage for 2U servers without overwhelming the system with too many alerts?
3. Network Performance: We’re monitoring network traffic and bandwidth, but are there any specific metrics I should be paying extra attention to for 2U rack servers?
4. Custom Dashboards: Any advice on building dashboards that provide clear insights briefly for our sysadmins?
5. Alerting: How do you manage alert thresholds for a system like this? We want to avoid false alarms but also catch critical issues early.

I’d love to hear your experiences and tips. Thanks in advance!

marcrobinson

First, I assume you have SAM or SWOSH/hco licensing. The legacy SAM module is what you want. I would start by looking into the SolarWinds Academy. There are both instructor led and video instruction. I would recommend the Office Hours too. The instructors have open hours that you can join and directly asked questions. Some of the videos are short bits that focus on single topics too. SolarWinds Academy | SolarWinds Success Center

Second, Each of your questions is a topic in itself. See above about the Academy. It is very helpful. 

Finally, some responses to point you in the right direction. 

1. Temperature -  This is part of hardware health monitoring. Thresholds ought to be something that will be tuned to your environment. You can look at setting thresholds based on metrics after a week for example. Monitor hardware health

2. Resource - Make sure that CPU and memory are checked on the node details page in the node management UI. Same for Volumes that you want to track. 

3. Dashboards - Check the Academy. Here is another link and a good start: SolarWinds Observability | Create and manage custom standard dashboards

4. Alerting. Alerts are based on polled data that is in the database. Look into the anomaly alerting and setting up thresholds. There is no single master threshold beyond use what works for your environment. Beyond that - persistent high utilization is usually the default recommendation. SolarWinds Observability | Alerts and  Anomaly-Based Alerting in SolarWinds Observability Self-Hosted and General threshold types in the SolarWinds Platform. Also - CHECK THE ACADEMY - it is full of good stuff.