Any word on what is causing a handful of us to lag?

[l1nk]

I dont understand u guys with all that connection issues. I found it hilarious today:
https://www.youtube.com/watch?v=BVXYLzIV3Cs
https://www.youtube.com/watch?v=SoDpuqCgXwg

[Raist]

So just a coincidence that it started the day they made major changes to their data center?

No... there is a ton of activity on the internet at any given moment, and we become more and more dependent on the internet daily for so many aspects of our lives--just streaming alone has ramped up our consumption drastically with all the apps for things like sports and news, not to mention things like Netflix. You are also in the Great Lakes area if I remember right? Which means you are getting routed through areas that routinely exhibit problems. And, as I mentioned, the peering agreements your ISP has in place are weighted heavily towards Level3 and AT&T--the two most frequently reported ISP's at the moment. Didn't I mention that in my previous post?

And no... that does not make me an apologist for SE. The facts are the facts...if anything, I'm more a realist I guess? There is documentation spanning back into the 90's over these issues. Filings with the FCC. Forum posts all over the place for various services--and not all of them games. There is a known infrastructure problem in the US...it is why the Net Neutrality movement started in the first place. The problem is people flat out refuse to hold their ISP accountable...when they are the ones that contributed most to the situation we are in. Think about it...how much has the bandwidth increased since Clinton was President? (back then the premium residential was typically no more than 3Mbits, 6 and 10 were still the typical max when Bush Jr. took over) How many times have the standards been revamped? (back then, DOCSIS 1.x was deployed and 2.0 was coming in...3.1 is now deployed). Would it surprise you to know that some local ISP's are still using equipment that has not gone beyond a standard established between 1998 and 2000?

The internet can go down at any point for any reason...just like a flight can get delayed/cancelled, traffic can get jammed, or your mail/packages don't arrive on time. That doesn't mean it is the fault of the hotel you made your reservation with, the coliseum hosting that sporting event, or Amazon that shipped your package on time. It is because something happened in transit to cause your plans to get monkeyed up.

[Raist]

So you are OK with paying for a product but not being able to use it? It is this kind of thinking that allows businesses to get away with the crap they do. Game developers are among the worst offenders of producing crap and charging people for it. This fanboy style defense is exactly why they get away with it and I see it on every forum of every game that has some kind of major issues.

I am not venting about how I don't want to sub anymore. I am venting about how their awful support is going to lose them my money... making sure it is clear to them in multiple formats.

And you keep paying your ISP for their service, which apparently may be failing you. Why aren't you putting your money where your mouth is with them?

I and others did with TWC here about two years ago (for this game, have gone to them off and on for nearly 20 years now with this sort of thing...and it has always worked). We actually found a DOCSIS 1.1 CMTS in play at our headend...and the outrage ensued (it was a device from 1998). The CMTS is basically what "terminates" our cable transmission so it can get onto a normal network (data on cable is packed differently, as it is designed for video transmission). Basically, it was only capable of speeds less than half of what was being pushed into the market---and it wasn't even able to keep up with 15Mbit connections anymore because everyone is trying to move to the higher tiers and exhausting it's throughput--not to mention it was only bonding 4 channels when what they were selling calls for 8 channel modems.

The point is... they were caught deliberately overselling and under-supporting their market, and we held them to task for it.

[Raist]

I think I've mixed up my threads, I've been talking about the Chaos data centre move and for any confusion caused I apologise.

But to answer your question, if the lag was completely SE's end then doesn't it seem reasonable that everyone would have it? And there's not massive amounts of people complaining in game, on places like Reddit or even on these forums (I can see there are complaints, but not compared to the actual number of players) so it really suggests the majority of players AREN'T having an issue?

I know that's no help at all to you, but the best advice is still to contact your Internet provider and tell them you're having a problem connecting to certain networks. They can route you around troubled areas if you can get high enough up in their support for them to actually change anything.

LOL. Easy to do... there's only like 20 threads on the same topic now, and only like 2 pages of posts in the sticky.

Regardless, what you said is technically true. Whether it was to Europe, Japan, or right next door...the same principles apply when it comes to the internet. To get to a server I know is physically right down the street (our team built/installed it in 2006), I go all the way up into North Carolina and back...the data travels north of 250 miles round trip, just to hit a server 8 miles away--and it is even hosted on TWC as well!

[Moonraker]

I posted a detailed post over on the official Lag Issue thread where it appears that, at least from my analysis, the issue is partly due to the Frankfurt network backbone hops but a large part seems to be actually on the KDDI host network for the European Data Center (Chaos).

The relevant post is here which includes some PingPlotter images showing the problem which is happening each primetime since the move and more details from what I can see.

I did include instructions at the end if you want to test it also. Pingplotter is free (has a paid option/s but usable for 20 days or simpler free version) or just using simple Ping testing built into Windows etc.

Network issues are always tough to pinpoint but when the users ping is lower and fairly consistent, it points towards other issues like pack loss and/ or possibly actual server load issues. From what is seen so far, it looks like the network is either not coping correctly with the primetime traffic load or some other reason why connections have dropped packets or similar symptoms.

it doesn't mean it is not a wider network issue with the player IS{ etc. but it does at least help to try and narrow the issues down and is something that SE should be able to address and fix.

[Raist]

There is packet loss occurring earlier in your route though...mostly around the Level3 portions. Congestion is ramping up at the opentransit/Level3 exchange...well before it gets to KDDI (which is fairly consistent with other reports). That is an election made from your ISP's policies, not SE.

For a comparison, here is some data on my connection from here in the states:

C:\Windows\system32>tracert neolobby06.ffxiv.com

Tracing route to neolobby06.ffxiv.com [195.82.50.9]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms LPTSRV [10.10.100.1]
2 43 ms 27 ms 30 ms cpe-75-176-160-1.sc.res.rr.com [75.176.160.1]
3 26 ms 33 ms 34 ms cpe-024-031-198-005.sc.res.rr.com [24.31.198.5]
4 16 ms 12 ms 13 ms clmasoutheastmyr-rtr2.sc.rr.com [24.31.196.210]
5 24 ms 25 ms 28 ms be33.drhmncev01r.southeast.rr.com [24.93.64.180]
6 55 ms 56 ms 57 ms bu-ether45.asbnva1611w-bcr00.tbone.rr.com [107.14.19.44]
7 63 ms 31 ms 47 ms 0.ae0.pr0.dca10.tbone.rr.com [66.109.6.183]
8 54 ms 58 ms 61 ms ash-b2-link.telia.net [195.12.255.229]
9 32 ms 31 ms 32 ms ash-bb4-link.telia.net [213.155.130.2]
10 128 ms 127 ms 134 ms ffm-bb2-link.telia.net [80.91.246.63]
11 126 ms 127 ms 129 ms ffm-b10-link.telia.net [80.91.247.189]
12 127 ms 130 ms 140 ms kddi-ic-301630-ffm-b10.c.telia.net [62.115.32.110]
13 129 ms 129 ms 129 ms 195.82.60.29
14 129 ms 131 ms 128 ms 195.82.61.14
15 126 ms 129 ms 128 ms 195.82.50.234
16 129 ms 129 ms 130 ms 195.82.50.9

Trace complete.

C:\Windows\system32>ping neolobby06.ffxiv.com -n 100

Ping statistics for 195.82.50.9:
Packets: Sent = 100, Received = 100, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 125ms, Maximum = 173ms, Average = 131ms

While there is a bit more jitter than I'd like, the latency is at a somewhat manageable level, and not a massive issue with packet loss (100/100 made it). The difference is in the routing. They are both using routes to get to KDDI's ASN, but the one I am given is considerably more stable.

**Note this is while my mom is streaming to the Roku downstairs on the wi-fi, and I am also in game on the wi-fi waiting on DF...and the numbers are still pretty solid from the US.

You are showing excessive jitter early in the route that may trigger retransmits, as well as packet losses within your ISP's network. In the longer sample you have over 14% in combined losses tracked before you even hit the Level3/KDDI exchange. Each of those hops is showing lost packets when they were pinged--indicating there is heavy congestion impacting response times (also evidenced by the extended spikes). If you could get a pathping to complete, you might actually see some delays estimated on the forwarding side as well because of high utilization. When I ran a pathping just now, my route came up clean:

C:\Windows\system32>pathping neolobby06.ffxiv.com

Tracing route to neolobby06.ffxiv.com [195.82.50.9]
over a maximum of 30 hops:
  0  G75VXLAP [10.10.100.10]
  1  LPTSRV [10.10.100.1]
  2  cpe-75-176-160-1.sc.res.rr.com [75.176.160.1]
  3  cpe-024-031-198-005.sc.res.rr.com [24.31.198.5]
  4  clmasoutheastmyr-rtr2.sc.rr.com [24.31.196.210]
  5  be33.drhmncev01r.southeast.rr.com [24.93.64.180]
  6  bu-ether45.asbnva1611w-bcr00.tbone.rr.com [107.14.19.44]
  7  0.ae0.pr0.dca10.tbone.rr.com [66.109.6.183]
  8  ash-b2-link.telia.net [195.12.255.229]
  9  ash-bb4-link.telia.net [213.155.130.2]
 10  ffm-bb2-link.telia.net [80.91.246.63]
 11  ffm-b10-link.telia.net [80.91.251.250]
 12  kddi-ic-301630-ffm-b10.c.telia.net [62.115.32.110]
 13  195.82.60.29
 14  195.82.61.14
 15  195.82.50.234
 16  195.82.50.9

Computing statistics for 400 seconds...
            Source to Here   This Node/Link
Hop  RTT    Lost/Sent = Pct  Lost/Sent = Pct  Address
  0                                           G75VXLAP [10.10.100.10]
                                0/ 100 =  0%   |
  1    2ms     0/ 100 =  0%     0/ 100 =  0%  LPTSRV [10.10.100.1]
                                0/ 100 =  0%   |
  2   16ms     0/ 100 =  0%     0/ 100 =  0%  cpe-75-176-160-1.sc.res.rr.com [75.176.160.1]
                                0/ 100 =  0%   |
  3   20ms     0/ 100 =  0%     0/ 100 =  0%  cpe-024-031-198-005.sc.res.rr.com [24.31.198.5]
                                0/ 100 =  0%   |
  4   13ms     0/ 100 =  0%     0/ 100 =  0%  clmasoutheastmyr-rtr2.sc.rr.com [24.31.196.210]
                                0/ 100 =  0%   |
  5   25ms     0/ 100 =  0%     0/ 100 =  0%  be33.drhmncev01r.southeast.rr.com [24.93.64.180]
                                0/ 100 =  0%   |
  6   32ms     0/ 100 =  0%     0/ 100 =  0%  bu-ether45.asbnva1611w-bcr00.tbone.rr.com [107.14.19.44]
                                0/ 100 =  0%   |
  7   33ms     0/ 100 =  0%     0/ 100 =  0%  0.ae0.pr0.dca10.tbone.rr.com [66.109.6.183]
                                0/ 100 =  0%   |
  8   30ms     0/ 100 =  0%     0/ 100 =  0%  ash-b2-link.telia.net [195.12.255.229]
                                0/ 100 =  0%   |
  9   31ms     0/ 100 =  0%     0/ 100 =  0%  ash-bb4-link.telia.net [213.155.130.2]
                                0/ 100 =  0%   |
 10  138ms     0/ 100 =  0%     0/ 100 =  0%  ffm-bb2-link.telia.net [80.91.246.63]
                                0/ 100 =  0%   |
 11  125ms     0/ 100 =  0%     0/ 100 =  0%  ffm-b10-link.telia.net [80.91.251.250]
                                0/ 100 =  0%   |
 12  129ms     0/ 100 =  0%     0/ 100 =  0%  kddi-ic-301630-ffm-b10.c.telia.net [62.115.32.110]
                                0/ 100 =  0%   |
 13  129ms     0/ 100 =  0%     0/ 100 =  0%  195.82.60.29
                                0/ 100 =  0%   |
 14  130ms     0/ 100 =  0%     0/ 100 =  0%  195.82.61.14
                                0/ 100 =  0%   |
 15  127ms     0/ 100 =  0%     0/ 100 =  0%  195.82.50.234
                                0/ 100 =  0%   |
 16  128ms     0/ 100 =  0%     0/ 100 =  0%  195.82.50.9

Trace complete.

[l1nk]

Square - is there any chance it will be fixed before 3.1 launch?

[Raist]

Square - is there any chance it will be fixed before 3.1 launch?

May be better to direct that question to your ISP.

[l1nk]

My ISP keeps telling me 'works on our side, have a good day'.

[Raist]

My ISP keeps telling me 'works on our side, have a good day'.

Than they are likely only looking at your localized nodes. You may still be working with the first tiers of support--need a higher level. Tier3 is the group with the means to conduct a more in depth look at the situation, as it is usually deeper in then tiers 1 & 2 are allowed to tinker.

Here is an article with some details on the levels.of multi-tiered suppor. It is more in the scope of software projects, but the same principles apply to in the networking world. The structure is pretty universal, and is even used in places like the service department at a car dealership.

http://project-management.com/unders...-desk-support/