Author Archives: maboughey

AoE server outage 15/12/17

What happened?

The server crashed due to an erroring chunk in the world. The server failed to restart after that crash.

Why didn’t the server restart itself?

I forgot to restart the monitoring script after the last server issue

How were things fixed?

The erroring chunks were rolled back from backup to a version that was working (about a day) which only affected one small area of the world.  The monitoring scripts were also restarted.

Updates on the server

After yesterday’s issue with the server, there have been a couple of changes:

1. Backups

Server backups are now back in place. Once every 3 hours there is a backup. They have been checked and they are working.

2. Forestry blocks

There is no way to change the existing errored blocks. The worldborder has however, been changed to 22,00 blocks wide (up from 16,000). You will be able to find new blocks outside of the old word border.

 

Debrief: Server Crashes 5th December

You may have noticed that the server was starting and stopping constantly earlier today.

So what happened?

There was a tile entity in the mining world that was causing the server to crash when it was started. Our monitoring services picked this up and kept trying to restart the server.

Fixing the error

The erroring tile entity was removed from the server once it was found.

Forestry blocks, what happened to them?

In the process of trying to find, and fix, the cause of the server crash a configuration option was enabled to allow the server to replace erroring tile entities automatically, this did not work however.
I am of the opinion that this option replaced the forestry blocks

Please generate new terrain to find new forestry blocks

But wheres your backups?

Due to the recent server move, there has been some issues with the backups due to using a different backup storage location. We are currently in the process of sorting these issues out.

What have we learnt from this?

  1. Our monitoring service works to restart the server when it is down
  2. We need to have a way to notify the relevant people when the monitors fail to restart the server
  3. The config option will never be used again
  4. We need to focus on getting backups working properly again.

A big thank you

I would like to take a minute to thank you all for making MCNSA the place that it is. An extra thank you goes to those who have supported the server and keep the lights on, without you, MCNSA would not still be running.

Unfortunately, due to life and studies, I am unable to visit and play on the server, but I have heard great things about the people that play on the server and make it what it is now.

I never thought that MCNSA¬†would still be running. It’s been over 5 years since the server was started, not many servers last as long as we have. We have had many different heads of the server, each one bringing something new but, at the same time, keeping the core principles of the server alive. Life always happens, but MCNSA lives on thanks to those who step up to the mark.

You, the player, are the most important thing to the server, you, after all, are the community. Thanks to every one of you for making the community what it is.

It’s been nearly 2 years since I took over ownership of MCNSA. Even though I am not around, my role devolved to just paying the bills, I always like to keep track on what’s going on with the server.

I’d like to personally thank the admin staff for being there when I cannot. Without you guys, the server wouldn’t be updated and problems wouldn’t be solved.

To everyone, keep the server great and let’s see if we can get to 10 years.

maboughey