Regarding Reports on Recent Connection Problems with the NA Data Center

[Laraul]

I have Cox im AZ and have not been noticing any issues with latency and the game. No lag spikes. No 90001 errors.

A 30 minute span during prime time. As you see I have no packet loss.


I'm unclear why some people are having problems and others not.

A few observations.

1. If it were NTT servers that were at fault then the latency problems would not be limited to just one ISP. All ISPs would be effected.
2. I don't see how an Cox support operator would know details for why just one game isn't working.
3. Back when the game passed through Level3 servers, a few people had these same problems. People complained that Level3 servers were garbage and they we're at fault. Like people are blaming NTT now.
4. A lot of people like me aren't experiencing problems. But you never here from us because why visit these threads unless we start having trouble.

[Syrehn]

Snip.

The reason you're not having an issue and others are is because they're hitting the congested NTT nodes during peak hours that are failing under the load and you're not. Simple as that.

It's not limited to just Cox, there are other threads with other people from other ISP's having the exact same problem when their connection is travelling through specific NTT nodes.

[Nega_Nomicon]

I have Cox im AZ and have not been noticing any issues with latency and the game. No lag spikes. No 90001 errors.

A 30 minute span during prime time. As you see I have no packet loss.


I'm unclear why some people are having problems and others not.

A few observations.

1. If it were NTT servers that were at fault then the latency problems would not be limited to just one ISP. All ISPs would be effected.
2. I don't see how an Cox support operator would know details for why just one game isn't working.
3. Back when the game passed through Level3 servers, a few people had these same problems. People complained that Level3 servers were garbage and they we're at fault. Like people are blaming NTT now.
4. A lot of people like me aren't experiencing problems. But you never here from us because why visit these threads unless we start having trouble.

"But you never here from us because why visit these threads unless we start having trouble." - So why are you here? We've already established that it's not affecting everybody which is what makes this situation frustrating and strange. I have exceptional hardware and I pay my ISP for the highest speeds they offer. If you did more research you'd learn that it's affecting Cox and other ISPs. If I play any other online game it runs beautifully as it should. This issue is entirely up to Square Enix to solve or they're going to lose a lot of subscribers from the west coast, southern, regions.

[Bluefve]

I have Cox im AZ and have not been noticing any issues with latency and the game. No lag spikes. No 90001 errors.

A 30 minute span during prime time. As you see I have no packet loss.


I'm unclear why some people are having problems and others not.

A few observations.
1. If it were NTT servers that were at fault then the latency problems would not be limited to just one ISP. All ISPs would be effected.

Wrong. MMOs don't work like that. Internet Data Centers often have numerous routes and uplinks, depending on geolocation and IP (for the intent of optimizing connections and providing more bandwidth. This could be a situation where one particular uplink is messed up, and why not everyone sees it.

2. I don't see how an Cox support operator would know details for why just one game isn't working.

True, but an ISP Network Engineer would have better insight and understanding for enterprise-level network architecture and their own infrastructure. The engineer I spoke to talked about his own experience working with Blizzard on a similar issue, back in 2013, for WoW.

3. Back when the game passed through Level3 servers, a few people had these same problems. People complained that Level3 servers were garbage and they we're at fault. Like people are blaming NTT now.

Level3 does have bad routing. Its worse in certain geographic areas though, like France.

4. A lot of people like me aren't experiencing problems. But you never here from us because why visit these threads unless we start having trouble.

Right, no one said EVERYONE is experiencing the problem. But the internet is a complicated beast, and a large number of people ARE experiencing a problem.

Two additional observations from your own data that I've noticed, is that...
1) Mostly (if not all) everyone who's having this problem is from California, CA. (You yourself are in AZ, which may not be affected)
2) Mostly (if not all) everyone who's reporting the problem are pinging 204.2.229.9 (You are pining 204.2.229.103 -- could mean everything, could mean nothing)

I don't know how relevant these details are, but they are nonetheless details.

My Pingplotter results taken just now.

Target Name: 204.2.229.9
IP: 204.2.229.9
Date/Time: 1/10/2018 9:19:17 PM - 1/10/2018 9:29:17 PM

Hop Sent PL% Min Max Avg Host Name / [IP]
1 472 0 0.55 18.01 1.56 SYNOLOGYROUTER [192.168.1.1]
2 472 0 4.72 23.56 9.00
3 472 0 4.70 24.54 9.65 ip68-4-11-194.oc.oc.cox.net [68.4.11.194]
4 472 0 5.25 30.27 11.06 ip68-4-11-16.oc.oc.cox.net [68.4.11.16]
5 472 0 7.47 68.47 14.58 68.1.1.63 [68.1.1.63]
6 472 0 8.26 53.58 15.94 ae-7.a00.lsanca20.us.bb.gin.ntt.net [129.250.194.165]
7 472 1 17.92 40.26 21.84 ae-3.r01.lsanca20.us.bb.gin.ntt.net [129.250.2.233]
8 472 1 8.12 34.10 14.00 ae-8.r23.lsanca07.us.bb.gin.ntt.net [129.250.6.48]
9 472 3 34.06 59.62 38.16 ae-12.r22.snjsca04.us.bb.gin.ntt.net [129.250.4.150]
10 472 3 32.97 55.35 37.39 ae-19.r01.snjsca04.us.bb.gin.ntt.net [129.250.3.27]
11 472 8 35.81 61.52 41.08 ae-1.a00.snjsca04.us.bb.gin.ntt.net [129.250.2.228]
12 472 8 37.46 68.59 41.03 xe-0-0-42-1.a00.snjsca04.us.ce.gin.ntt.net [129.250.195.42]
13 472 23 36.84 57.31 41.21 204.2.229.242 [204.2.229.242]
14 472 5 35.93 98.00 41.41 204.2.229.9 [204.2.229.9]

Would be interesting to see more people's PingPlotter results, too, versus plain traceroutes.

[Laraul]

1) Mostly (if not all) everyone who's having this problem is from California, CA. (You yourself are in AZ, which may not be affected)

Several posts we're from people in AZ, including the original poster. It is not limited to just people in California.

2) Mostly (if not all) everyone who's reporting the problem are pinging 204.2.229.9 (You are pining 204.2.229.103 -- could mean everything, could mean nothing)

I pinged my server, which is 204.2.229.103. I assume 204.2.229.9 is the pinger's server. I don't know for certain

I was simply trying to say that the problem is likely an issue local to the user. If you are experiencing frequent disconnects and/or lag spikes over a number of days, call Cox and have them come out and do a line test. Try also checking your modem's web management server (if it has one). I believe it's usually found by typing "http://192.168.100.1" into a web browser. Lastly you can always reboot/reset/replace your router (if possible). Most people have already tried some of these already.

The last time I was experiencing frequent disconnects and lag spikes, it turned out that line that runs to my block was mauled on by some animal. And it wasn't discovered until they did a visual line check of the line. This was causing a dip in line signal power resulting in a poor connection. It took them two or three days to find and repair it.

[XenoCranel]

Several posts we're from people in AZ, including the original poster. It is not limited to just people in California.


I pinged my server, which is 204.2.229.103. I assume 204.2.229.9 is the pinger's server. I don't know for certain

I was simply trying to say that the problem is likely an issue local to the user. If you are experiencing frequent disconnects and/or lag spikes over a number of days, call Cox and have them come out and do a line test. Try also checking your modem's web management server (if it has one). I believe it's usually found by typing "http://192.168.100.1" into a web browser. Lastly you can always reboot/reset/replace your router (if possible). Most people have already tried some of these already.

The last time I was experiencing frequent disconnects and lag spikes, it turned out that line that runs to my block was mauled on by some animal. And it wasn't discovered until they did a visual line check of the line. This was causing a dip in line signal power resulting in a poor connection. It took them two or three days to find and repair it.

That is totally unrelated to this issue.

[Kerwin]

Several posts we're from people in AZ, including the original poster. It is not limited to just people in California.


I pinged my server, which is 204.2.229.103. I assume 204.2.229.9 is the pinger's server. I don't know for certain

I was simply trying to say that the problem is likely an issue local to the user. If you are experiencing frequent disconnects and/or lag spikes over a number of days, call Cox and have them come out and do a line test. Try also checking your modem's web management server (if it has one). I believe it's usually found by typing "http://192.168.100.1" into a web browser. Lastly you can always reboot/reset/replace your router (if possible). Most people have already tried some of these already.

The last time I was experiencing frequent disconnects and lag spikes, it turned out that line that runs to my block was mauled on by some animal. And it wasn't discovered until they did a visual line check of the line. This was causing a dip in line signal power resulting in a poor connection. It took them two or three days to find and repair it.

This is a tech forum post, if you are not having problems then take your posts elsewhere. Otherwise you clog the thread with your useless statements.

[Chortle]

Several posts we're from people in AZ, including the original poster. It is not limited to just people in California.


I pinged my server, which is 204.2.229.103. I assume 204.2.229.9 is the pinger's server. I don't know for certain

I was simply trying to say that the problem is likely an issue local to the user. If you are experiencing frequent disconnects and/or lag spikes over a number of days, call Cox and have them come out and do a line test. Try also checking your modem's web management server (if it has one). I believe it's usually found by typing "http://192.168.100.1" into a web browser. Lastly you can always reboot/reset/replace your router (if possible). Most people have already tried some of these already.

The last time I was experiencing frequent disconnects and lag spikes, it turned out that line that runs to my block was mauled on by some animal. And it wasn't discovered until they did a visual line check of the line. This was causing a dip in line signal power resulting in a poor connection. It took them two or three days to find and repair it.

ABSOLUTELY DO NOT DO THIS!

Cox will charge you fees for sending someone out.

Also, the issue is NOT on Cox's end as it has been stated multiple times. What time of day did you check your latency? Most of the people here have issues between 6-10pm (sometimes later) ONLY. Every other time of day there seems to be no issue.

Please do not give people advice that could potentially cost them money. When a substantial amount of players from the southwest are experiencing an issue... chances are it's not "a local issue to the user" and it's definitely not because our lines were "mauled on by some animal".

There are plenty more people having this issue than those commenting on this thread. Someone posted about it on reddit yesterday even.

https://www.reddit.com/r/ffxiv/comme...ghly_unstable/

[drebbe]

...
There are plenty more people having this issue than those commenting on this thread. Someone posted about it on reddit yesterday even.

https://www.reddit.com/r/ffxiv/comme...ghly_unstable/

Going to copy what was said at reddit here so SE can clearly see it.

This has been going on for nearly a month, and it looks like it's only recently come to light that the issue is dropped packets with the company SE contracts to host their servers. To quote one user,

"Cox got back to me. Confirmed on NTT.net's side, not a local network or Cox issue -- quite likely one of their several uplinks has has been designated for Cox users (perhaps, just for LA/OC) and is smaller than the real throughput traffic of Cox customers. Therefore, during peak times, it get's congested and drops packets and only users which have been assigned that particular route (based on ISP and perhaps geolocation) are affected and no one else.

Cox is unwilling to contact NTT.net directly, since they are the direct provider for SE. Cox urges users to pressure SE into getting their direct provider, NTT, to resolve the problem.

NTT.net is unwilling to investigate, unless a client directly requests it (AKA SE)-- per their Network Operations Team via noc@us.ntt.net.

Square Enix, please get your Network Engineers to validate based on provided data. I'm willing to cooperate further, and I'm sure other players will be willing to as well."

[Ladon]

Several posts we're from people in AZ, including the original poster. It is not limited to just people in California.


I pinged my server, which is 204.2.229.103. I assume 204.2.229.9 is the pinger's server. I don't know for certain

I was simply trying to say that the problem is likely an issue local to the user. If you are experiencing frequent disconnects and/or lag spikes over a number of days, call Cox and have them come out and do a line test. Try also checking your modem's web management server (if it has one). I believe it's usually found by typing "http://192.168.100.1" into a web browser. Lastly you can always reboot/reset/replace your router (if possible). Most people have already tried some of these already.

The last time I was experiencing frequent disconnects and lag spikes, it turned out that line that runs to my block was mauled on by some animal. And it wasn't discovered until they did a visual line check of the line. This was causing a dip in line signal power resulting in a poor connection. It took them two or three days to find and repair it.

Just to add what the others have already said, please don't do this, it is not good advice and shows a fundamental lack of understanding how to read trace route results.

The trace routes being posted here clearly show issues after the peering hand-off to NTT. If this was an ISP or last mile issue, you would see high latency and/or packet loss the first couple of hops.


With that said, just want to add supporting evidence that it clearly looks to be a capacity issue with NTT during peak hours. I had the opportunity to play some this morning PT, off peak, and saw no issues and my trace route is totally clean:




Compared to last night where issues start to ramp up around 6pm PT, and then hit unplayable levels around 7:30PM PT: