For VPS services, VEDOS is responsible only for the operation of the virtual machine itself and its connectivity. Management of the operating system and other software is entirely the customer’s responsibility. We do not perform any installations, modifications or other interventions.
Need help with your website? Find your specialist on VEDOS Profi ⧉.
In this article, you will learn:
- What problems you will most often encounter with the VPS service
- How to check that the VPS service is working
- How to restart the VPS service
- How to check server resources in Linux
- How to deal with a server compromise
- Frequently Asked Questions
Common VPS problems
Common VPS problems include the unavailability of the server as a whole or of certain applications.
If the server does not respond at all, not even to ping or an SSH connection attempt, check that the VPS is working and, if necessary, restart it.
The unavailability of a key application can still cause problems with displaying your website or with access via SSH. Check the application’s status, its logs and the server resources. If SSH is not working, connect via KVM.
Checking VPS functionality
If the server is not working for an unknown reason, first check that the service itself is working. First, visit the page Status ⧉page, where you will find information about current outages. Information about planned maintenance can be found in the Customer Administration ⧉ and in your email.
If there is no outage or maintenance, open the administration of the service itself using these steps:
- Log in to the Customer Administration ⧉.
- In the top menu, select Server services VPS SSD, or VPS ON.
- In the overview, select the service you want to check.
First, in the Basic information table, check that the service Status service active. If it is not active, the service has expired and is therefore switched off.
If the service is active, also check whether it is running and connected to the network.
- For VPS HDD/SSD perform the check directly in the administration. In the left menu, click VPS status and information. Verify that Status has the value running and Network connected.
- For VPS ON log in to the OpenNebula Sunstone ⧉interface. In the virtual machine management, check that the state is RUNNING and that the graphs on the Network tab show activity.
If the Network tab of the VPS ON shows removed IP addresses, the VPS has been disconnected from the network on our side.
The network may also be unavailable due to an error in the server configuration. Connect via KVM and check the network configuration with the command ip a (Debian, Ubuntu), or ipconfig.
Restarting the VPS
A restart can solve a large number of problems, including increasing system resources and a so-called kernel panic, a crash of the operating system in a non-standard situation. Follow these guides:
If the server cannot be started at all and you have a backup available (your own or ours), restore the server from the backup, or reinstall it and upload your data.
Checking server resources in Linux
A lack of system resources can show up as a slowdown of the server, or as freezing or crashing of individual services or of the entire operating system.
Connect to the server via SSH or VNC (KVM).
VNC (KVM) works even when SSH is unavailable and when the server is disconnected from the network.
- Connect to VPS HDD/SSD by following the guide VPS – Accessing the Server via VNC.
- Connect to VPS ON in the ON Sunstone ⧉interface. After selecting the virtual machine, click the display icon in the control panel .
Run the commands:
ps,topandhtopto check RAM and CPU. If you are hitting the limits, consider restricting some processes or upgrading your VPS HDD/SSD or VPS ONplan. Note that in the administration a VPS plan can only be upgraded, not downgraded.dfanddf -hto check the disk and the number of files. Large volumes and large numbers of files can cause problems. Clean up disk space regularly.
Avoid solving memory problems by enabling swap. Access to data on disk is significantly slower and the server will slow down even more.
If checking server resources and resolving their exhaustion did not help, go through the system logs and the logs of individual applications (usually in the directory /var/log).
Server compromise
Servers are most often compromised in order to send spam and to carry out attacks as part of a botnet.
Outdated applications (for example old, non-updated versions of PHP) and various content management systems are commonly attacked. You can detect the problem by carefully checking the running processes with the command ps aux.
On our side, we usually handle a compromise by informing the server’s subscriber by email, or by SMS in the case of a larger attack. In cases of a very serious compromise, or if the subscriber does not respond, we disconnect the server from the network in an effort to minimize damage until the service administrator removes the problem.
A server disconnected from the network is not accessible via SSH, but you can connect via VNC (KVM).
If you want to report that the compromise of the server has been resolved, reply to the technician’s original message or write via the form ⧉. Provide the name or number of the service and a description of the solution.
Frequently Asked Questions
I cannot connect via SSH. What should I do?
Connect using VNC (KVM) following the guide for VPS SSD/HDD, or in the ON Sunstone ⧉.
I don’t remember the root password for logging in to the operating system. How can I log in?
If you use a preinstalled operating system, you can use the form ⧉ to request that the service setup email be sent again (VPS HDD/SSD), or change the root password in ON Sunstone ⧉ following this guide (VPS ON). If you have your own OS, or you changed the root password of the VPS HDD/SSD service after setup, unfortunately we cannot provide it to you.