Hello everybody,

I am battling a strange and piquant situation. Since Feb. 06 ( when I signed up for a VPS with EV1 ) my VPS has crashed atleast 10 times.

We are using less than 10% of the disk space but our disk inode usage is very very high. It keeps hitting the limit they have set.

Coldfusion, Tomcat etc. is all off. Still this problem.

Never had this problem with earlier VPS service providers. Nothing has changed in the configuration or no. of virtually hosted sites.

I will be grateful for any sugesstions, pointers, solutions?

Hi, did you get a solution to this issue? I seem to have the same problem with my VPS and the disk free ("df") command varies between 4gb space available (i.e. total VPS space) and 20gb space available.

The disk inode issue is caused a lot of times by simply too many files that are usually small in size to be able to fit into the allocated space.

Are you saying your vps literally changes the disk usage in huge amounts from time to time? If that is the case something else is going on - first check from your side it is not the logs or something weird.

This is from the response to my ticket at Ev1. To their credit I must say that the systems support specialist (once he got in at 1pm) resolved the issue quickly and even called us twice to update us on the situation (once to confirm that he was working on the issue and once again after the issue was resolved).

But that doesn't explain what gave EV1 the right to experiment with disk backup tools on a production server? I am sure its not much trouble to make the script check if there's enough disk space before doing any auto-backups. I'm told by the tech that they are tweaking the script now.

==================
The problem is that we are currently building a system to backup all of our VPS servers on a nightly basis. This backup had filled the hardware node for your VPS and thus you were getting invalid information reported on disk space usage.

I logged into the hardware node and deleted these old backups and configured a script to run and clear these out after they are sent offsite.