It has been a very laggy Rust Wipe on the Voodoo Server.
We have been doing tests to find out what the cause of the issues are.
It was not what we thought it would be.
First we had to check the server hardware to make sure the server can run the Rust server application. We confirmed that the hardware can run the server application.
The next step was to check if there was any data centre issues and the data centre reported that there are no issues and everything is running 100%. No network lag reported.
The last step was to check the server configuration and settings and here we found a lot of information pointing to the following possible causes.
- Server Upkeep time and Entities
- Server Seed and Map Size.
Server Upkeep time and Entities
Rust servers have a limit of how many Entities it can handle and how many new entities it can handle at the same time.
The longer the server upkeep time is on a server the bigger the bases are and bigger bases mean more entities. On busy servers this can cause server lag when players in big bases build fast or all players build at the same time.
Default Decay.upkeep_period_minutes “1440” (How many minutes does the upkeep cost last? 1440 = 24 Hours)
Current Decay.upkeep_period_minutes “9000” 6.25 Times longer than the default
All the suggestions point to returning the server Decay.upkeep_period_minutes back to the default setting.
Server Map Size
The Current server Map Size is bigger than the suggested maximum map size of 3000.
Next wipe pick a map with a maximum map size of 3000.
Buy Maximum number of slots from the hosting Company.
Our Plan going forward
- Increase the server slots to the maximum amount of slots available form the hosting Company. We will implement this immediately.We will be monitoring the server performance for the rest of the wipe. If this do not resolve the server performance issue we will move the the next option.
- Return the Decay.upkeep_period_minutes back to the default setting of “1440” If this do not work we will move to the final step.
- Map size not bigger than 3000.
Our hope is that the first step will resolve the server performance.