Can't connect to my droplet(ipv4, console)!?. Need to turn-off the droplet and turn-on it again to fix this.

Hey @sittiphansittisak,

It seems like your Apache Workers were full(if you are using Apache). Assuming you are using Apache, what you are describing is not high load or websites going down but your Apache Workers filling up.

So first, what are Apache Workers and what are they used for:

Imagine you are in a cinema, and there are 300 chairs inside. Those are the Apache Workers and the movie is your Webiste. If someone wants to watch the movie(your website) they need to sit in the chair (Apache Workers). If all 300 chars are taken, the 301 person needs to wait for someone to leave in order for them to sit and watch the movie. That’s how ApacheWorkers work. Once all ApacheWorkers are taken, a new person that wants to open your website waits for a spot to be open and only after that they can see your website.

Usually, these limits are set by 256 by default but they can be increased.

You can open your Apache configuration file. In there you should see something similar to:

ServerLimit 256
MaxClients 256
MaxRequestsPerChild 4000

Increase both the ServerLimit and MaxClients options, restart Apache, and see if this would solve your issue.

It’s possible they do not exist (haven’t been added) and the default values are the ones above. Adding them to the prefork content of apache mods should be fine as well.

Don’t forget to restart Apache afterward.

Hi there,

I would recommend starting with the following:

• Check your server logs to identify what might have caused the crash. For example, you could use this script here which will help you summarize your access logs and see if there was some suspicious activity :

BASH Script to Summarize Your NGINX and Apache Access Logs

• Check your system logs in the /var/log directory and look for any errors that might have caused the problem like an out of memory errors and etc.

• Check your resource utilization graphs in your DigitalOcean Control panel to verify if your server was running low on resources like RAM or CPU. Additionally, you can monitor server resources via SSH directly and keep an eye on your server’s resource usage, such as CPU, memory, and disk space. High resource usage can cause a server to become unresponsive. Use tools like htop, vmstat, and iostat to monitor your server’s performance:

How to find the processes that are consuming the most server resources?

• If you are running low on resources, make sure to add more RAM and CPU to your server or add a swap file as well so that you could have some extra buffer.

• Use the DigitalOcean Uptime service so that you could get notified if your website is not reachable:

https://www.digitalocean.com/pricing/uptime-monitoring

Hope that this helps!

Best,

Bobby

Hello @sittiphansittisak

Can you share more details about the droplet configuration? Do you run a busy website that gets a lot of traffic?

One thing you can do i to create an uptime alert so you can be aware when your application/website is down. Check the docs here:

https://docs.digitalocean.com/products/uptime/how-to/create-alerts/

Another step is to monitor the droplet resource usage so you can act before the droplet becomes unresponsive:

https://docs.digitalocean.com/products/monitoring/how-to/set-up-alerts/

Hope that this helps!