Give us your feedback on the coming hardware upgrades

[Ibn]

Use this thread to give us your feedback on the hardware upgrades.

[Mogosh]

I think I can speak for many when I say I appreciate the move. I'd much rather have some downtown in the short term, with a week of notice, than another rollback at an unexpected moment.

[zuizide]

Thank You.

How much was the final bill? :p

[zathros]

New database hardware is nice, but what about new server hardware? Serverside lag seems to be the cause of agroing monster on the edge of radar for example. More pack space can't be added because it would be too much of a load on the system. These are just a few symptoms of older servers.

New servers are expensive, but is there an ETA? Maybe if new advertising attracts enough new players? Maybe the servers are relatively new and I don't know what I'm talking about? =X

[Silverwuf]

This is wonderful news! Thanks for all your hard work :)

Silver

[Mat Icefalcon]

IBN im a little confused or something. Could ya please point me in the right direction of where the writeup/info of the coming hardware upgrades could be found.

Thanks in advance.

Opps found it.

[boneyard]

so what about the 4 other worlds, we gotten new hardware some time ago but you now say its not really working as well as you want. can we expect that to get upgraded also or are we just out of luck?

btw: on which cluster is vt running? or does it have a cluster of its own currently?

[Ibn]

I'm not sure what you're referring to Boneyard -- the database hardware that Darktide, Morningthaw, Harvestgain and Wintersebb are on is working fine. Verdantine also runs off that database cluster.

[Rhinehold_LC]

sounds good to me, thanks for the fair warning, and your dedication to this game.

-R

[pacesetter]

Thanks for the notice and once again showing ya care for us and the game.

[quote] Once the worlds were up and running, we were able to take a step back and look at the long-term consequences of the incident. While the alternate hardware is capable of running the four worlds, the performance and reliability of the cluster isn’t where we want it to be.[close quote]

as you know, we expect perfection. :p so knowing you are willing to go this far in an attempt to give it to us because " the performance and reliability of the cluster isn’t where we want it to be", shows involvement for our benefit.

Thanks Turbine!

[boneyard]

While the alternate hardware is capable of running the four worlds, the performance and reliability of the cluster isn’t where we want it to be. i thought this was about the other worlds that got new hardware some months ago, but i see i read it wrong.

sorry.

[Ibn]

Originally posted by zuizide
Thank You.

How much was the final bill? :p

I can't say exactly, but I'll tell you this: I REALLY wish Jessica had let me put it on my Discover card. CASH-BACK BABY!

:)

[Rage_Lao]

Originally posted by Ibn
I can't say exactly, but I'll tell you this: I REALLY wish Jessica had let me put it on my Discover card. CASH-BACK BABY!

:)

LOL

[Silifi Of Death]

hmm... is Ibn secretly working for Discover Card? It's a conspiracy... of some sort...


:)

[Kajabor]

While you are at it, could you put a few ds3 lines directly on the Darktide server, please?

[Luminescent]

hardware upgrade?:confused:

[Joe]

Could we get the exact specs of the old hardware and the spec of the new hardware?

[boneyard]

very doubtfull, partly because of company secret and partly because whatever they post people gonna whine because they feel it could have been better.

[Bob Down]

We were all under the impression that Solclaim migrated to new servers some time back. Wasn't that the gist of the whole Shadowclaim thing?

[boneyard]

solclaim was given new hardware for its gameserver part and not for its database part which it shares with 3 other worlds.

This server cluster maintained the databases for the worlds of Frostfell, Leafcull, Solclaim and Thistledown.

[vagrant]

as long as we see better performance I'm happy.

[Ma Yuan]

So is TD still on old hardware or has it been upgraded?

[boneyard]

the servers for all worlds except solclaim are running on "older" hardware, meaning they havent been recently upgraded.

[Sortileger]

I am always happy to see hardware upgrades as in most cases they increase performance and reliability of both hardware and software.

Besides new hardware, are there any other plans to insure data integrity and disaster recovery? In light of all the recent rollbacks across all the worlds, and understanding that sometime failures just happen, are there any plans to review any of these procedures/platforms to insure that data integrity issues are identified more quickly and action taken to resolve those issues happens faster? Maybe more aggressive monitoring of transaction logs. Regular sampling of data streams. Hot standby clusters where switchover can occur at the first sign of an issue?

I have no idea what type of infrastructure you are dealing with. I have no idea what limitations you have due to operating systems or code. I actually know I don't understand a lot about your environment. But I do understand that every time a rollback occurs, everybody loses something. I'm willing to bet most players won't agree with me, but I would rather endure downtime than a rollback. Downtime just stops me from playing, but a rollback erases all progress I have made since that point. Its very disheartening to feel the accomplishment of finally acquiring that certain item or reaching that certain landmark only to find that it was all for not. I have no doubt that you understand this issue as you always must take the heat that follows their occurrence. And I would assume that most of you are players yourselves. It would be very hard for me to work there and not sample the product.

I don't mean to sound ungrateful, but I must ask the hard question of what are you going to do to prevent rollbacks in the future? What can be done? Is this in your infinite wisdom the solution to the problem? Is there a research team that analyzes your Quality of Service and your monitoring procedures to insure that they are in fact the best they can be? In this last rollback that occured, you had prior knowledge that there was an issue but had no idea of the depth of that issue, yet you chose to let the worlds continue running. As a result of that decision, you have now impacted the gaming experience of thousands of players. You will always receive my sympathy and understanding for downtime as it is sometimes a necessary evil, but you will get no consolation for data loss of any type. In this profession, data loss is the ultimate sin that is not forgiven in the corporate or government environments. To accept it because the client interface happens to be entertainment based in nature is an insult to our profession, imho.

I apologize for the harshness and aggressiveness of these opinions, and I realize they may not be shared by my brethren.

For those of you whose initial response will be to "get a life" or to "get over it" I can only add this. I have a very good life due to the fact that I don't get over it.

As always, these are just the opinions of a player.

Sortileger

[cliser]

Posted on 2nd August


I'm not sure what you're referring to Boneyard -- the database hardware that Darktide, Morningthaw, Harvestgain and Wintersebb are on is working fine. Verdantine also runs off that database cluster.
__________________
Alex "Ibn" Beckers
AC1 Online Community Relations


So in light of the recent failures on the DT/VT (database) cluster which then used the 'new hardware for TD etc', where are we now ?

TD had a major failure and new hardware was bought. It was all ready to go then the DT/VT cluster failed so they used the new hardware to replace that cluster. :eek:

So where is the new database cluster for TD ? :confused: :confused:

[Kachina]

I would point out that many of us work week ends and Tuesday is our day off. :mad: I remember when updates etc were done from midnite to 6 AM. What happened to that? :confused: