|
|
| Author |
Thread Statistics | Show CCP posts - 10 post(s) |
|
Kaemonn
Forum Moderator
Interstellar Services Department
   |
Posted - 2006.08.08 04:04:00 -
[1]
Yes the server crashed. The GMs are aware of the issue and are trying to get it resolved. Updates when I get them.
UPDATE: GM Spiral made a news annoucement here
UPDATE 2: Server Hamsters Make a press announcement here
forum rules | [email protected]
You mean to tell me, theres a game that goes with the forums?
|
|
|
GM Spiral
|
Posted - 2006.08.08 04:15:00 -
[2]
Servers are on the way up again. More news once we sort through the fallout. Stay tuned.
|
|
|
GM Spiral
|
Posted - 2006.08.08 04:22:00 -
[3]
 Originally by: Commodore Andrews
http://img481.imageshack.us/img481/1641/hamstersrevengema7.jpg
By
AgentApple
This is going into our report to the developers. ;)
And make it into a link, images are baaad, mmkay? :)
|
|
|
GM Spiral
|
Posted - 2006.08.08 04:26:00 -
[4]
DON'T PANIC!
Due to the fix the server needs a tiny short reboot again.
Our apologies.
|
|
|
GM Spiral
|
Posted - 2006.08.08 04:30:00 -
[5]
Originally by: Xaeon
This is starting to get annoying. I wish it would just make its mind up.
No play until after a dinner and a movie, sorry. 
|
|
|
Valar
|
Posted - 2006.08.08 04:31:00 -
[6]
We've located and fixed the problem. It was caused by a moiconfiguration (by me) of the autogrow settings on the datafiles.
As every DBA knows, using autogrow is bad, but we have it turned on so the cluster doesnt crash if the datafile runs out of space. I had been having issues with autogrowing during extended downtimes, so before I left on vacation, I changed the settings to what I though would grow the files 1gb or 500mb depending on what file it is, instead of 10%.
For some reason, when I looked at the config today it had reverted to using % growth and was growing the files 200-1280000%... or well... trying to do that and in turn fill up our disk arrays.
I've fixed this setting, increased the size of the transaction log file and I hope that this is the end of the problems of the last few days.
I also want to announce that I will at most extended my weekends by one day... bad stuff happens if I take any more vacation then that 
------
Valar
Database admin - Server operations team
CCP Games
How to write a good bugreport |
|
|
GM Spiral
|
Posted - 2006.08.08 04:35:00 -
[7]
Originally by: Valar
I also want to announce that I will at most extended my weekends by one day... bad stuff happens if I take any more vacation then that
CCP needs more Valar. Even jovians need a vacation now and then. 
|
|
|
Valar
|
Posted - 2006.08.08 04:52:00 -
[8]
I was just looking at the market and system loading queues. The loading process is slower then usually because I did a failover on the database server during the crisis, so I could backup the transaction log.
Initiating a failover makes the SQL Server release all of its RAM, and thus all of its cached data, so the first startup after a failover is always very long and laggy.
This will sort itself in ca.30-45 minutes.
------
Valar
Database admin - Server operations team
CCP Games
How to write a good bugreport |
|
|
Valar
|
Posted - 2006.08.08 04:57:00 -
[9]
All market regions are up, and the load queue for solarsystems is being processed faster now, due to increased cache hits on the db server 
Everything seems to be in order, so I'm going back to bed. The *on call* guy is also heading to bed, but I doubt he's getting much sleep anyways since he recently had twins
I'm certain I'll be woken up if there are any more issues tonight.
Good night everyone.
------
Valar
Database admin - Server operations team
CCP Games
How to write a good bugreport |
|
|
GM Spiral
|
Posted - 2006.08.08 05:02:00 -
[10]
Originally by: Valar
Everything seems to be in order, so I'm going back to bed. The *on call* guy is also heading to bed, but I doubt he's getting much sleep anyways since he recently had twins
I was wondering why the poor man sounded so sleepy on the phone. We were worried he had fallen asleep on the way to his computer. 
|
|
|
|
|
|
| |
|