Blank droplet - CPU spike every hour, Disk I/O spike

Hi there @Kay86,

This is quite interesting, I’ve created a new fresh CentOS 8 Droplet and install the DigitalOcean monitoring agent to see if I can replicate that behavior. I will let it run for a few hours and update you once I have some more data in the graphs.

What is the region that you’ve deployed the Droplet at?

Also do you have any Cron jobs at the moment? You can check that with:

crontab -l

Did you also use any of the images from the DigitalOcean Marketplace, or was it just the default CentOS 8 Droplet?

Regards, Bobby

Observing periodic spikes in CPU, load, and Disk I/O on a fresh DigitalOcean droplet with CentOS 8, especially when no significant activities have been performed, can be intriguing. There are a few common reasons why this might be happening:

1. System Cron Jobs

• Description: CentOS, like most Linux distributions, runs a series of scheduled tasks (cron jobs) that execute at regular intervals. These tasks can include log rotation, system updates checks, and other maintenance tasks.
• Diagnosis: To see if this is the case, you can check the cron schedules. Look into directories like /etc/cron.hourly, /etc/cron.daily, etc., or check the crontab files with crontab -l and sudo crontab -l.
• Resolution: If a cron job is causing the spike and it’s a necessary system function, it’s usually best to let it continue. However, if you identify an unnecessary task, you can modify the schedule or disable it.

2. DigitalOcean Monitoring Tools

• Description: If you have DigitalOcean’s monitoring tools installed, they might perform regular checks and data collection. This is less likely if you haven’t set anything up, but it’s worth considering.
• Diagnosis: Check if any DigitalOcean specific monitoring services are running.
• Resolution: These monitoring services are generally lightweight and shouldn’t be causing significant resource usage, but if they are, you can configure or disable them.

3. System Updates

• Description: Automatic system updates might be running in the background, especially on a new droplet.
• Diagnosis: Check the logs for any automatic update services like yum-cron for CentOS.
• Resolution: If automatic updates are causing the spikes, you can configure them to run at a less impactful time or update the system manually.

4. Pre-installed Software

• Description: Even on a new droplet, some pre-installed software might be running background tasks.
• Diagnosis: Review the list of running services and processes to identify any potential culprits.
• Resolution: If you find unnecessary services, you can disable or uninstall them.

5. Resource Monitoring

• Description: Utilize tools to monitor your resources in real-time to catch what process is spiking the usage.
• Diagnosis: Tools like top, htop, or iotop can be used to monitor system resources and identify what processes are using them.
• Resolution: Once you identify the process, you can research it further to understand its purpose and how to manage it.

Investigating Further

• Checking Logs: System logs (/var/log/messages, /var/log/syslog, or journal via journalctl) can provide insights into what’s happening on your system.