[EU] Unusually high latency

[Krisom]

By unusual I don't mean the unusual physical misplacement of Europe servers being stationed somewhere in Canada. Before I give you a nice trace route I would like to show you where Europe is


Europe

And now I can give you an example of what is not Europe

Not Europe

Now that we've got that out of the way let us look at some other issue.

Tracing route to 199.91.189.45 over a maximum of 30 hops

  1    <1 ms    <1 ms    <1 ms  BTHomeHub.home [192.168.1.254]
  2    13 ms    13 ms    13 ms  217.32.141.36
  3    12 ms    13 ms    13 ms  217.32.141.62
  4    16 ms    15 ms    16 ms  213.120.161.114
  5    16 ms    15 ms    16 ms  31.55.164.195
  6    16 ms    16 ms    16 ms  31.55.164.109
  7    15 ms    15 ms    15 ms  acc2-10GigE-0-3-0-3.bm.21cn-ipp.bt.net [109.159.
248.242]
  8    23 ms    23 ms    24 ms  core1-te0-3-0-16.ilford.ukcore.bt.net [109.159.2
48.156]
  9    20 ms    20 ms    20 ms  peer2-xe11-0-0.telehouse.ukcore.bt.net [109.159.
254.130]
 10    20 ms    20 ms    21 ms  t2c3-xe-1-1-2-0.uk-lon1.eu.bt.net [166.49.211.18
0]
 11    21 ms    20 ms    20 ms  166-49-211-38.eu.bt.net [166.49.211.38]
 12   187 ms   190 ms   252 ms  xe-1-3-0.mtl10.ip4.tinet.net [89.149.184.74]
 13   183 ms   184 ms   184 ms  ormuco-gw.ip4.tinet.net [216.221.156.110]
 14   190 ms   216 ms   187 ms  192.34.76.2
 15   191 ms   187 ms   188 ms  199.91.189.234
 16   183 ms   182 ms   181 ms  199.91.189.45

Trace complete.

I would like to add that this has only started yesterday night and is still here today. As always the server placement has always been wrong.

[Risae]

Ha. Made me laugh.

Anyway, backbone is getting flooded on weekend which causes packetloss and highlatency. There is nothing Square or you can do, aside from using a VPN to avoid those.

[Asdamine]

There is nothing Square or you can do[...]

Err, you mean SE can't open up European located servers/data-centres?

[Raist]

Err, you mean SE can't open up European located servers/data-centres?

Not unless they want to spend a butt-load of cash on it. In other words... it's probably not going to happen for a long time, if at all.

It's not just a matter of putting up servers in one of their European offices. They also need to have a dedicated high-speed line to synch them to the Japan datacenter as well, and possibly the Canada one too. They've been fighting with connectivity issues between Canada and Japan for about a month now as is. Throw a third site in the mix and things could get even uglier.

Now, if they would give in and allow more regionally locked services, they could trim the costs (and logistic nightmares) dramatically... but that's no likely gonna happen any time soon.

[Axgard]

They also need to have a dedicated high-speed line to synch them to the Japan datacenter as well, and possibly the Canada one too.

What for? There is no cross-realm mechanics at all.

PS: About prices, just for example - DC in UK pricing per square metre that ranges from €197 per month up to €507 per month ($317-$816 or 31.255¥ - 80.442¥).

[Reithomeige]

In the passed two weeks indeed there is lag galore happening. So bad that I DC at times. But it's impossible to do a Duty finder on Mondays.

Having created a tracert I see an average ms of 29 ms to 36 ms on EU soil. As soon as I reach this IP [4.69.153.229] "Level 3 Communications, Inc" I hit: 157 ms.

I ran this tracert after a DC at 10AM GMT. I am aware that the result is nothing alarming like having 300+ms. Nevertheless, the lag is not pleasant.

so using a proxy fixes everything?

[Raist]

What for? There is no cross-realm mechanics at all.

PS: About prices, just for example - DC in UK pricing per square metre that ranges from €197 per month up to €507 per month ($317-$816 or 31.255¥ - 80.442¥).

Who knows just what they are synching between the data centers, but it was announced several times in the Lodestone that they were having a lot of trouble with it, and it was denying players access. Maybe it's DF related, or account authentication when zoning. But it was causing issues with stability for the service.

Edit:
Found this post from a few weeks ago in the Lodestone:
http://na.finalfantasyxiv.com/lodest...682b20b31f9e6c

[Recovery] Notice of Network Recovery (Oct. 22)
During the following time, we were experiencing instability with our network connections, which caused instability with connections from North American and European regions to the Japanese data center and vice versa.

We are pleased to announce that the issue has been resolved. We apologize for any inconvenience.

[Date & Time]
From Oct. 21, 2013 1:30 p.m. to Oct. 22, 2013 3:15 a.m. (PDT)

[Details]
- Instability with connections from North American and European regions to the Japanese data center
- Instability with connections from Japan to the North American/European data center

[Cause]
- Internet instability

[Affected Service]
- FINAL FANTASY XIV: A Realm Reborn

For whatever reason, they need the two data centers to have a reliable connection. And it's not just the Data Centers, but the Clients as well. The JP servers are still needed in the background for all of us (look at your connections in netstat or resmon, and you'll see your client opens a line to a server in JP even though you are playing on a server in NA).

[Risae]

Err, you mean SE can't open up European located servers/data-centres?

The fact that they didn't in the first place means that they indeed do not care about the EU players.

[Krisom]

Hello, thanks for the responses. I'm back down to 120ms (80-90ms with tunneling). Must have been something up for a short while.