Hardest stress on a hard drive comes from power-up/power-down also defraging I heard. I notice this server requires alot of resetting to make sure that everyone doesn't get bored plus DM's adding things. Would you say this is the biggest reason you think this caused the melt down?

I heard also with the large amount of people on a server it helps that everyone has a direct connection to the server. Also when a DM resets a computer it's best to wait a bit to log in while the server is trying to reset itself. This would be a way to avoid lag which some people have complained about in the past ;).

Shortcut:

1. Open up your Neverwinter porgram file and find NWNMain.exe.

2. Make a shortcut to it to your Desk Top area.

3. Right click on the shortcut and click on "Propertys"

4. under "Target" you will see "C: \ NeverwinterNights\NWN\nwmain.exe"

5. add the server address so it looks like "C: \ NeverwinterNights\NWN\nwmain.exe +connect 69.143.245.52:5121"

6. click on "Apply" and "OK" to close it out.

Trinistad Stonemender Hardest stress on a hard drive comes from power-up/power-down also defraging I heard. I notice this server requires alot of resetting to make sure that everyone doesn't get bored plus DM's adding things. Would you say this is the biggest reason you think this caused the melt down?

Uh...that ip address is not efu @_@ Use: efu.us.to

I also find it very difficult to believe that connecting 'directly' like that has any thing to do with server preformance. I would rate this as optional ^_^

Came from a Head DM running a server that had a large amount of people logged in that Direct Connect helped his server so I would put that under "helpful" :D

Arkov

Biggest reason is probably power failures/surges plus 24/7 operation for a year. :)

Having my profession running large scale server farms for a couple of Fortune 500 companies, I can say with certainty that Arkov is probably closer to the truth as to the cause of the failure than any other reasons.

Believe it or not, hard drives have defined lifespans listed in their, often hard to find and cryptic, specifications. These lifespans are based on the number of rotations the platters make. The number is quite large, but when a machine runs 24/7 a hard drive reaches that number much faster than it would under normal computer use (basic internet, office type work, etc).

Not only does it chew through it's lifecycle faster, but it also creates and maintains a higher temperature in the PC. Even with a top notch cooling system, computer components are still affected by their own internal temperatures. Then add in dust, dirt, power surges, the fragility of the controller boards on the drives themselves, Raid controllers, and the overall complexity of a computer and you get a million and one reasons why a computer could fail.

And while Howland might have taken all the precautions in the world to keep his machine running effecient, cooled and clean, even servers at billion dollar companies like Bank of Scottland, Microsoft and GE, have more dust collecting on thier fans than most people have under their couches at home. It would truly repulse you if you've seen what I've seen on servers that serve critical applications like Email and serving IP addresses to broadband customers.

So in short, lets assume that there is little we can do to prevent a hardware failure on a server that none of us probably even know the location of.

Except to throw some money to Howland to get an UPS with surge protector if he hasn't so far. And if he has, there is the cost of new HDD, maybe upgrading the cooling system , etc. So it will surely help.

or you could retrofit the refridgeration unit from a small freezer into it and have it runing off an alternate powerpoint to prevent power dips like my pc sufferes. while your at retrofitting stage why not retrofit a car turbo to periodically pound 10psi worth of air through the system to keep the dust moving so it dosnt settle on it. however the down side to both is that their relativly noisy so you would have to build in "sound stealth" which limits the airflow. can be good can be bad but it all depends on "to what extremes" your willing to go. even Pay to play servers go down from time to time so i would think of the little drop outs to be only a minor pain compared to what value you get for your $$... oh ps i ya want that retrofitted cooling tower im more than happy to build one although it wont look to pretty the only cost would be the postage lol the parts for it i can find local lol.

Nagypapi Except to throw some money to Howland to get an UPS with surge protector if he hasn't so far. And if he has, there is the cost of new HDD, maybe upgrading the cooling system , etc. So it will surely help.

I whole-heartedly agree. Though we can't stop hardware from failing we can aid in lessening the financial blow to those who maintain a module that plays more like a pay-for-play MMORPG than some casual module by the average fan.