So when does SE plan to credit our accounts for the multiple ongoing service issues?

[Baingoleth]

What Square Enix needs to do is make an official acknowledgment. We know the news section constantly reads "DDoS attacks here, DDoS attacks there." but we really need to hear from a Dev with some reinforcement. Otherwise people will continue to blame Square.

Lots of posts by the SE NOC here:

http://forum.square-enix.com/ffxiv/forums/625-Status

Seeing as they are the people at SE who directly deal with NTT when NTT's service is impacting the game, that is official enough for me.

[Kerwin]

Lots of posts by the SE NOC here:

http://forum.square-enix.com/ffxiv/forums/625-Status

Seeing as they are the people at SE who directly deal with NTT when NTT's service is impacting the game, that is official enough for me.

While that is enough for some... As you can plainly see, that does not seem to be enough for many others in this thread and on the forums.

[Alien_Gamer]

While that is enough for some... As you can plainly see, that does not seem to be enough for many others in this thread and on the forums.

Lol, then it will never be enough. He is the official representative for them regarding network issues, you aren't going to get a more official statement than what he gives. Yoshi may be the guy in charge but he won't be giving any different information than what Noc is posting.

[Kerwin]

Lol, then it will never be enough. He is the official representative for them regarding network issues, you aren't going to get a more official statement than what he gives. Yoshi may be the guy in charge but he won't be giving any different information than what Noc is posting.

https://na.finalfantasyxiv.com/lodes...7d341dbcf750b1

There we go. That wasn't so hard. A little extra response goes a long way.

[Maero]

Well to be fair the ddos attacks are not the only issues at hand, it is also the very bad coding and the not so great servers SE still use, lol
Granted yes ddos attacks are not a SE thing but we still have issues from time to time when not attacked

[Catwho]

Well to be fair the ddos attacks are not the only issues at hand, it is also the very bad coding and the not so great servers SE still use, lol
Granted yes ddos attacks are not a SE thing but we still have issues from time to time when not attacked

SE's back end is pretty standard. Most databases can only handle about 5000-10000 connections at once, which is why SE's server cap is 5000 players at a time, with the lobby server slowing things down to 100 new connections per minute. The servers themselves are pretty stable, only crashing completely once ever few weeks. When it's their fault, SE admits it was a "server programming error" in the recovery notes, usually some bug that caused a memory leak or somesuch.

The network infrastructure itself isn't that great, but again it's also industry standard. Every character is a constant query upon 3-4 servers as we navigate between duties, instances, party finder, etc. They could lift the query cap on the database, but past experimentation has found that the database gets gummed up at 5000+ characters, and that matches my own experience in a development environment whereupon a relationship database chokes and dies beyond 10,000 simultaneous queries. We keep our Oracle databases capped at 10K as well.

[Claviusnex]

SE's back end is pretty standard. Most databases can only handle about 5000-10000 connections at once, which is why SE's server cap is 5000 players at a time, with the lobby server slowing things down to 100 new connections per minute. The servers themselves are pretty stable, only crashing completely once ever few weeks. When it's their fault, SE admits it was a "server programming error" in the recovery notes, usually some bug that caused a memory leak or somesuch.

The network infrastructure itself isn't that great, but again it's also industry standard. Every character is a constant query upon 3-4 servers as we navigate between duties, instances, party finder, etc. They could lift the query cap on the database, but past experimentation has found that the database gets gummed up at 5000+ characters, and that matches my own experience in a development environment whereupon a relationship database chokes and dies beyond 10,000 simultaneous queries. We keep our Oracle databases capped at 10K as well.

Actually there are many systems that handle many more connections and DB queries with ease. For the stuff I worked on most of my career having 20K-50K concurrent users per image was not uncommon. I also worked with some large database design and we could easily get well beyond 10K transactions per second. It typically isn't the DB or operating system software that is the limiting factor but the hardware and application itself. Given the little I know about FF XIV's environment I would say they are hardware constrained and some of that constraint is due to poor application design. I would also say some of the hardware constraints are also based on how much SE is willing to spend in relation to profit margins. So forgoing the DDoS issue which is not their problem SE could probably improve game performance if they were willing to invest in it.

I just want to add onto the DDoS attacks a bit. While SE is using NTT for their backbone access it is important to realize they are not the only backbone provider to be attacked. L3 has also had DDoS problems over this same period of time. L3 also had a routing error that affected Comcast customers during this same time frame. So the current problems really are beyond the control of SE.

[Alien_Gamer]

Well to be fair the ddos attacks are not the only issues at hand, it is also the very bad coding and the not so great servers SE still use, lol
Granted yes ddos attacks are not a SE thing but we still have issues from time to time when not attacked

SE servers were upgraded in the move, the hardware isn't a bottleneck in this issue. The game's back end is horribly antiquated but also irrelevant to this issue. The problem is with the connection between the datacenter and customers; the game itself has remained fairly stable during this. So to be fair, this is not SE fault and we deserve no compensation from them.

[Maero]

SE servers were upgraded in the move, the hardware isn't a bottleneck in this issue. The game's back end is horribly antiquated but also irrelevant to this issue. The problem is with the connection between the datacenter and customers; the game itself has remained fairly stable during this. So to be fair, this is not SE fault and we deserve no compensation from them.

And i never said i wanted compensation.lol
I thought for sure all they had done is move the servers, i guess i never read that they actually upgraded them.
I still think they could of left servers in Montreal and have more in Cali. My connection was so much better with it being in Montreal

[Valkyrie_Lenneth]

And i never said i wanted compensation.lol
I thought for sure all they had done is move the servers, i guess i never read that they actually upgraded them.
I still think they could of left servers in Montreal and have more in Cali. My connection was so much better with it being in Montreal

They did not move the servers, there is no way to move racks of servers over 3000 miles, and have them tested and up and running in 24-48 hours.
The reason they moved their DC out of montreal in the first place was due to electrical problems. The servers were overcapacity, and the DC could not provide stable power any longer.