Update 1/12/14 – Crash #2

Today we had a second crash that caused some damage to the server that was unrelated to the first problem listed below.

What happened?

The live map creates things called “Tiles” that are generated and stored on the server, this are generally really large files,  Some more of the map was explored, tiles generated and our server HDD filled up! Now this created a catastrophe of events that spiral out and create damage which includes:

• World borders deleted (Fixed)
• Ranks deleted (Fixed & added borders for End & Nether, as well as new /warp nether)
• Autorank data deleted (Fixed)
• Stats configuration deleted (Fixed)
• Corrupted all crates, Sadly there is no way to fix this as I was not keeping backups. This ones on me folks, sorry.

Theres a second part to this Crash no 2, which I will call Crash 2.1, this was only for people that joined the server while everything was broken (Note: I didnt start the server while it was broken, I can only assume the host or scheduler did) This may cause some of these problems for you if you:

• Homes/Bed deleted

We can remedy this by having staff tppos you to the right locations (once the maps re-rendered)

Now, we have fixed this by moving the entire map to a new hard drive on the server by itself, and system linking it, this should stop any future issues of the map obliterating hdd space, I have ran a full render of the main world, so please be patient while that re-generates, If anyone spams “Y TEH MAP BLACK??” please direct them to this thread.

Again, I will be creating a “Care package” for these 2 crashes for some lost items, please discuss what should be part of this one time kit below.

Care package created for everyone! Limited supplies so act fast, its open for 2 weeks, there is 100 packages use /donate in game then click on the cake then the cookie, or visit the direct store page: http://donate.piratemc.com/category/381565

Crash # 1 29/11/14

I think there must be someone that just sits and waits until im nowhere near a computer, then when im out of reach destroy my server.

I have been away this weekend, finally happy with getting the server stable, I get a message at about 8.30pm on my phone saturday night (While im out eating dinner) that the servers gone down, great. typical; I message SuperGL to boot it back up for me, I think the problems solved, my phone battery dies.

Get home about 1-2am, Servers down and wont start, quick google of the error message turns up we have a corrupt chunk in the world stopping the server from starting.

So whatever happened saturday night something caused the world to corrupt a chunk(s) I have a local backup of the world at 3GB, I am currently trying to get one of our backups to run, but im getting more errors at present.

This is the soonest I can get to a PC to type this up, but yet again, I am very busy today as it is my mothers 50th birthday. Extreamly poor timing.

If someone knows the reason for this mess, please step up and tell me.

Progress Update

1. I backed up the current broken chunk world locally (3gb)
2. I restored a backup from the 29th, but there is a problem with the hosts backup system, its set the wrong file permissions for the copy over, this results in session.lock, then a 30k error log of permission errors, The server is whitelisted, I jumped in the world just to see whats going on visually, and it was completly fresh, Im hopeing this was due to the permission errors not being able to read/write correctly.
3. I am waiting on a reply from the host to see if they cna sort out the absolut catastrophe; While I wait, I am going to repair the local version of the worlds I have to see if that works, but I will not be able to upload them back to the server, due to their sheer size and my poor internet connection.
4. The local scan of the main world & the end turned up no errors, thankfully there was 1 broken chunk in the Nether, I managed to delete this chunk and the world loaded! so problem found and fixed now comes the second problem; uploading this insane amount of data to the main server on a uselss internet connection, this took 8 hours before and failed, so im not looking forward to this.
5. There was an issue with the way the host copied over backups from their backup server to the minecraft server, it set the wrong owner permissions, which causes chaos, I asked them to do it again manaully, and this seems to have worked, we have a world backup from 24 hours previous working now.
6. Me, Tim, Wog and Ani scanned the nether for anything that could cause trouble, I regenerated 2 structured out of the world, I downloaded the nether, scanned it in the region fixer and it showed 2 “shared” errors that I fixed and im uploading the nether now, I hope this solves the corrupt chunk and that it wasnt anything else, im going purely on assumptions.
7. Server whitelist removed 20:00 – 30/11/2014, be warned were not sure if the problem is fixed.
8. I added a new /rule to not allow building outside of worlds, for example its against the rules to build on /top of the nether now.

Sidenote:

I wish I never pressed that restore button, now I have to re-upload everything, unless my host can sort out the permission errors they are causing from their backup server, hopefully nothing at all will be lost, and the culprit was someone messing around in the nether.

Update after putting the server live:

I will be creating a 24hours /kit once I get some time to make up for the downtime as a once off free kit for everyone to use, what do you think should be in it? It will be building blocks only & food, any stupid requests will be deleted.

Im not going to answer the stupid “OMG, LOOK, ITS GONE, TPHERE GODSDEAD NOW” people on the server, can I ask that as a community these people are delt with by the community as I have run out of patience for these kind of people.

Thank you for your patience.