2002 earlier today and now awful awful lag

ive reset the pc, modem ect... logged into other games and what not and not had any lag what so ever. but for some reason after having the login issues earlier now i can log in but the lag is horrible. cant hardly do dungeons even killing mobs for quests is becoming an issue. what is going on? have never had any issues with the game before.

Me too! I just came on to post this exact same issue... i can't do any pvp with my FC... horrendous lag in a game that didn't lag for me the last 8 months... it started today.

Southern California
Charter Communications (ISP)
Primal Data Center
Hyperion Server

its crazy bad, i was going to try to push for 50 this weekend and its made it almost impossible.

Also happening to me. I couldn't log in for most of the day, and now that I can, it takes 20-30 seconds just to change gear sets. This is unplayable.

Same thing happened to me extreme lag I can hardly even gather

I am relieved to know I am not the only one having this problem. I have reset my modem, reset my router, did several ping and speed tests to montreal canada (where the servers are located) IP tracked goblin server connections, called my internet provider, played several other online games to test... and it's only final fantasy that is lagging like CRAZY getting DC'd with 2002 and lagging every 1-3 seconds. it started this morning and is progressively getting worse throughout the day. SE please fix your mess.

I even downloaded and paid for a subscription of Battleping to see if that would fix it. Nope nope and nope.

Connection was fine for me all day but suddenly about 60-90 mintues ago I started getting massive lag spikes. It would work fine for about 15 seconds and then about 15 seconds of extreme lag to the point where everything would freeze and loop animations then suddenly do everything at once (just to be clear I know it's not a graphical issue).

Test connection and it's fine, can even watch a stream on Twitch at over 720p with no lag or other interruptions. It's possible it's something between my ISP and the game servers, but everything else seems to be working fine.

One odd things I've noticed is that my commands still seem to go through, and so most of my DPS rotations still land even though nothing on my screen is moving other than my own characters. It's as if the server is getting data correctly but not sending it at the right rate (which could also indicate traffic congestion where it's getting caught "in queue").

Definitely making the game unplayable though and hoping it gets fixed soon.

Well... whatever they did or didn't do during the data center maintenance... certainly changed something.... the game is unplayable.. so I logged off... had a raid tonight and one tomorrow... doesn't look like i'm going to make it.
I hope they at least provide a response that they know something is wrong for some people.

Horrid lag/spikes/delay.

Can't play the game as it is. I hope SE acknowledges the existence of this issue and gets fix'd soon.

That's all i want to see is a "we know about it and are looking into it" so i can check tomorrow

the cause of the drop outs/90k is cogentco seeing as they have from anywhere 15%-60% packet loss on there hops. the issue isnt with servers but its been with cogentco network which u would see if u use winmtr to tracert to server and leave it on for couple hrs it will show u the packetloss so u can easily see problem. issue only effects na/eu servers because of being routed through the bad path/ bad hop which cogentco has




only way to fix it is contact your isp and ask for lvl3 tech support and get them to reroute your intern around the bad hop which is atlas.cogentco.com


se doesnt have to acknowledge it the one who has to fix the routing issue is your isp

Quote:

---

Originally Posted by kaya_suzumiya

the cause of the drop outs/90k is cogentco seeing as they have from anywhere 15%-60% packet loss on there hops. the issue isnt with servers but its been with cogentco network which u would see if u use winmtr to tracert to server and leave it on for couple hrs it will show u the packetloss so u can easily see problem. issue only effects na/eu servers because of being routed through the bad path/ bad hop which cogentco has



only way to fix it is contact your isp and ask for lvl3 tech support and get them to reroute your intern around the bad hop which is atlas.cogentco.com


se doesnt have to acknowledge it the one who has to fix the routing issue is your isp

---

I'm sorry but this is not the case. The fact that this is happening to hundreds of players across all of NA makes it much more than just a local ISP. If it was indeed a local ISP provider, you would only see complaints from a specific location on the map and the same ISP provider. This however is not the case. It is safe to assume there are atleast 20+ different ISP provider players with all of the same issue.

This leads it to one conclusion: Based off of the starting of this problem, it clearly started acting up after SE's emergency maint. So it is indeed SE's fault for the 2002, and the terrible latency and lag.

cool, an answer =) thanks Kaya. ill see if i can get that done through my isp. is it normal for that to happen just all of a sudden the way it did? i dont know a whole lot about packet losses, hops and what not.

Quote:

---

Originally Posted by mattlp63

cool, an answer =) thanks Kaya. ill see if i can get that done through my isp. is it normal for that to happen just all of a sudden the way it did? i dont know a whole lot about packet losses, hops and what not.

---

Don't bother doing that. This is solely a Square Enix Issue. Her answer would be advice i'd give to singular, or same ISP users that all had the problem. but I've already counted 8+ provider players with the same issues. Comcast, Wave Broadband, AT&T, GCI, Horizon, Real Fast internet, Century Link, Condo Internet.

Quote:

---

Originally Posted by Mizukaru

I'm sorry but this is not the case. The fact that this is happening to hundreds of players across all of NA makes it much more than just a local ISP. If it was indeed a local ISP provider, you would only see complaints from a specific location on the map and the same ISP provider. This however is not the case. It is safe to assume there are atleast 20+ different ISP provider players with all of the same issue.

This leads it to one conclusion: Based off of the starting of this problem, it clearly started acting up after SE's emergency maint. So it is indeed SE's fault for the 2002, and the terrible latency and lag.

---

He/She wasn't claiming it was a local ISP. Cogent Communications is not a localized ISP. They are one of the MAJOR transit providers that most every local ISP enters into agreements with to carry data across long distances to other ISP segments. There are a small handful of such providers compared to the vast array of local ISP's. They all interconnect at specific exchanges to hand data off to the various backbones to get us across the country, across national borders, and across the oceans.

Cogent's network map:

http://www.cogentco.com/files/images...rkmap_page.png


TATA's network map:

http://www.tatacommunications.com/si...map%282%29.png


link to level3's map(wasn't posted as a single image, but a series of layered images):
http://maps.level3.com/default/#.U1x-W1c2lW8

So you can have customers from any number of local ISP's affected by one specific group's routes throughout a region if that route is hitting overloaded segments and they all have subscribed to that route, which is the case here. Your local ISP buys access to routes with different partners... it is up to them to either enforce any QoS they may have with your current routing partner to get that route cleaned up, or switch you to an alternate route.

Quote:

---

Originally Posted by Mizukaru

Don't bother doing that. This is solely a Square Enix Issue. Her answer would be advice i'd give to singular, or same ISP users that all had the problem. but I've already counted 8+ provider players with the same issues. Comcast, Wave Broadband, AT&T, GCI, Horizon, Real Fast internet, Century Link, Condo Internet.

---

Have them submit tracerts... you may find that they are getting routed across one, two, or three of the main offending routing partners in the US: cogentco, level3, or TATA (as6453).

so then what are our options? call our isp and ask what exactly? or is it something that might be fixed on its own? or by cogento?

so if it is this website cognetco or level3 or ATAT, what exactly is causing this issue? did SE stop providing for them? did something happen? will this issue stay indefinitely? This is a concern to me, because I use wave broadband and their IT is dumb as rocks and internet is shitastic, but it's the only provider I have in this apartment complex

so How am I going to be able to find what exactly wave broadband is using and tell them to stop using it? That seems pretty impossible seeing how most of their support members are booklet drones.

or list of providers that will work with FF14, when i move to a new apartment in 3 months

Hmm, I did find I'm getting routed through a bad hop through Cogenta but not sure what I can do to fix it. Pretty sure if I were to ask for "tier3 support" they'd have no idea what I'm talking about, or even if they did I'm not sure they would actually patch me through instead of just sending a request.

I (I guess thankfully) haven't had much dealings with ISP networking support, anyone have any suggestions for getting it resolved?

Sudden surges in traffic are usually the cause. Sometimes they erroneously flag such traffic as suspicious and it gets assigned a lower priority, causing packets to get delayed or dropped a lot. We saw this happen about two years ago in FFXI with BT/O2 customers in the UK, again back around Beta4 through initial launch of FFXIV. In both cases, they identified the misclassification within their own routing and worked with SE to implement a workaround for their specific network. Basically, they found SE wasn't putting something in the headers to prevent them from shaping the packets as low priority. Don't know exactly what the remedy was... hopefully they got SE to update the headers somehow. May be wishful thinking... but one can hope.

As for what to do... gather data (ie tracerts) showing that your routes may be over congested or otherwise you suspect you may be getting throttled (will have highly erratic response times at specific hops, possibly intermittent timeouts) and forward them to your ISP's tier3 support, requesting escalation to investigate the matter. That's usually enough to get the ball rolling, but sometimes you may want to throw in screen shots of your modem logs and signal pages if you can--especially useful if you see a lot of errors or a wide fluctuation in power/signal levels between the good and bad time frames.

It's gotten so bad with TWC in some areas, they don't even reply back anymore... they just adjust our routing within a week of sending them the reports.

Quote:

---

Originally Posted by UnlikelyHero

Hmm, I did find I'm getting routed through a bad hop through Cogenta but not sure what I can do to fix it. Pretty sure if I were to ask for "tier3 support" they'd have no idea what I'm talking about, or even if they did I'm not sure they would actually patch me through instead of just sending a request.

I (I guess thankfully) haven't had much dealings with ISP networking support, anyone have any suggestions for getting it resolved?

---

They all have a Tier3 support. It has to do with the networking structure. Each tier of agents is responsible for a specific level of support. Tier1 is basically the guys that answer the phone... the ones that get you to check local connections, reboot your devices, etc. When they send technicians on site, then you are moving into Tier2 territory, possibly into Tier3. After that, as you are getting into issues on the poles, at the exchanges, etc. you are moving into Tier3 and 4 (sometimes referred to as Maintenance and/or Engineering). But, it's an industry standard... they will know what you mean by Tier3 (or, at least they should).

so it seems like for goblin server issues, it's related to Ormuco... so would i just ask them to reroute through a different provider?

http://i62.tinypic.com/2ln8t2.jpg

this is what I got from my tracer


here's my second test:

http://i59.tinypic.com/xbl4cg.jpg

seems like the MS ping is bouncing around like fucking crazy.



There is one thing that I hate more than bad internet... SE's customer support.

above is identical to mine as well....

Charter Communications said they don't run lines past the first 6 and the rest is all Atlas.Cogentco.com. Looks like it's not our issue, it's Atlas.cogentco and I have no idea if they even know or even care.
I'm just waiting for some confirmation from SE.

I contacted Cogent Communications and told them about their ip address/hop 38.122.42.34. The operator said they are aware of the problem with the line and they are currently trying to fix it. They couldn't give me a time estimate as to how long it would take. All they said was that there were people on site, trying to isolate the line, and fix it asap.

Quote:

---

Originally Posted by Zell

I contacted Cogent Communications and told them about their ip address/hop 38.122.42.34. The operator said they are aware of the problem with the line and they are currently trying to fix it. They couldn't give me a time estimate as to how long it would take. All they said was that there were people on site, trying to isolate the line, and fix it asap.

---

I looked up their website and found this on their service status page

Quote:

---

Welcome to the Cogent Communications status page. Customers located on the east coast of the United States may be experiencing latency or packet loss due to a vendor fiber cut. The vendor is on site and is working to repair the cut, but there is no ETR at this time. The master case is HD5724309.

---

I'm guessing what's happened is that a lot of the traffic going through the cut area has had to be rerouted which is causing congestion issues (explaining why even us non-East coast people are having issues as well).

I just sent an email to Cognetco about the tracert images I took, and about houw thousands of gamers are having issues with it. if you guys take 6 minutes to all send them emails, I'm sure they'll get to fixing it quickly

Quote:

---

Originally Posted by Mizukaru

so How am I going to be able to find what exactly wave broadband is using and tell them to stop using it? That seems pretty impossible seeing how most of their support members are booklet drones.

or list of providers that will work with FF14, when i move to a new apartment in 3 months

---

Best way to get them involved is to submit traces to your servers showing signs of potential congestion/packet loss along your route. Ideally, you would want some to other services as well if you can--provides more compelling evidence of a problem to help nudge them into taking action. While a tracert isn't evidence in and of itself because of how it works (it pings each hop in your route a few times, whereas your game data just gets forwarded from hop to hop), it provides a roadmap of where your data is going and can point out where certain segments are enforcing clampdowns due to congestion or otherwise suffering other problems. It's more like a test for smoke... that leads them to where to look for fire.

To generate a trace, you will need either the URL (like www.google.com), or the actual IP address. To get the address for your server, you can open Resource Monitor (just type the command "resmon", no quotes, in your Run or Search box and hiting the Enter key) and then log fully into the game. While in game, alt-tab back to ResMon and expand one of the Network/TCP related categories and look for the connection that ffxiv.exe is using. The Canadian connection will have an IP starting with 199, the Japanese one will start with 124. Once you have tat information, you can run a trace.

The process of running a trace is actually pretty easy, but some are not comfortable with it... I get that. But if you've gone through the process of getting the actual server IP, you technically should be able to stop there and just give that IP address to your ISP's Tier3 guys. They <should> be capable of terminaling in to either your modem or your headend closest to your modem and generating their own reports from there for testing. Mileage can vary though, and you may need to actually do the legwork of running a trace and emailing it to them to get them off their duffs.

Windows has it's own tool built in to run a trace, so you don't really need to get a third-party tool (although, they can give a lot of information, a simple tracert should be enough to peak their interest... it's a very common tool still used in IT). Open a command prompt. You can do this by typing the command "cmd" (again, no quotes) in your Search/Run box and hitting the Enter key. This should pull up a window for entering text commands. Typically it has a black background with white text. Formatted something like this:

Code:

---

Microsoft Windows [Version 6.2.9200]
(c) 2012 Microsoft Corporation. All rights reserved.

C:\Windows\System32>

---

Before continuing, you may want to drag the window's width a little wider to avoid some of the wrap around text... can make the results look a little scrambled sometimes. Drag a corner of the window and just drag it like you would normally to resize a window. You can make all future command prompts wider by changing the default width. Click the black box icon up in the upper left corner of the window and select Defaults, then enter a new width value for the window size and click the "OK" button. Typically a value of 100 is enough to avoid the wrap around. Anywho... back to the how-to:

Your cursor should be blinking at you, waiting for a command. Make sure your focus is in the window (just click anywhere in the window), and type the "tracert" command, a space, then either the URL or the IP address you want to trace, and press the Enter key. Something like this:

tracert www.google.com

or

tracert 199.91.189.31

It will start pinging each hop along your route 3 times, and list the response times for each ping along with the name of the router/server as well as it's ISP. When done, it should look something like this:

Code:

---

Microsoft Windows [Version 6.2.9200]
(c) 2012 Microsoft Corporation. All rights reserved.

C:\Windows\System32>tracert www.google.com

Tracing route to www.google.com [74.125.131.105]
over a maximum of 30 hops:

  1    1 ms    <1 ms    <1 ms  LPTSRV [10.10.100.1]
  2    34 ms    20 ms    26 ms  cpe-098-025-064-001.sc.res.rr.com [98.25.64.1]
  3    13 ms    12 ms    10 ms  cpe-024-031-198-005.sc.res.rr.com [24.31.198.5]
  4    27 ms    28 ms    26 ms  24.93.64.124
  5    33 ms    35 ms    30 ms  107.14.19.20
  6    31 ms    31 ms    30 ms  ae-0-0.pr0.dca20.tbone.rr.com [66.109.6.183]
  7    29 ms    30 ms    61 ms  74.125.49.181
  8    30 ms    46 ms    32 ms  209.85.252.46
  9    32 ms    30 ms    30 ms  72.14.236.148
 10    41 ms    44 ms    49 ms  72.14.234.168
 11    43 ms    45 ms    43 ms  209.85.251.228
 12    *        *        *    Request timed out.
 13    44 ms    43 ms    43 ms  vc-in-f105.1e100.net [74.125.131.105]

Trace complete.

C:\Windows\System32>

---

When you see a * result like that for a response time, that means the ping timed out (took longer than about 3 seconds to respond... or didn't respond at all). When you see it with 3 stars like it did at hop 12, it isn't necessarily a cause for concern. Some routers will have a rule to flat out not respond to this specific type of test. What is of concern though is if that same router responds sometimes, but not at others (like you get a time/*/time response, or you ping it once and get * * * then ping it again and it responds). That is basically what you are looking for--erratic behavior.

Now, by erratic, I don't mean the slight variances you see at each line. It's going to fluctuate a little. The red flags are the intermittent timeouts, and a large difference in response times... like a swing from 30ms to 90ms, then back to 35 on the same line. Sometimes it's really out there... like 25 to 1500 than down to 90. If you start seeing swings where it varies by more than 50% between the best and last times, there may be something going on there. You will want to run a couple traces to compare them... you may find it looks good, then it looks bad, then it looks good again. These are the kinds of things that need to be brought to their attention---something is causing communication to bog down periodically, and it may at times cause data to get dropped and be retransmited. This can bring about the rubberbanding/lagged responses if it is happening enough to get you flagged for slow start in the TCP/IP protocol, and under extreme conditions it can cause disconnnections.

Now that you have been able to sniff out some potential problems... how do you get that lovely data to you ISP? Email is a great tool. Some will even let you paste it directly into a chat window if your contact is that way. But, you need to get it into the clipboard to do that. You can right-click anywhere in that window and a menu will pop up. Click the option to "Select all", and it will highlight all text on screen and within the buffer (it buffers a good bit of text that has rolled off the screen). Now, just press the Enter key and it copies all that text to the clipboard. Now you can just paste it into any text editor, email, chat window, etc. like you normally would (either that program's Paste option from the edit menu, or the Windows shortcut combination of CTRL-V).

Now, if you find you are doing this often, there is a way to just run the trace directly into a text file. It's basically done through the same process, but you append switches to make it dump the results into a text file instead of showing them on screen. You can save the commands into a .bat file that you can just run when you want, and it will dump it all into a .txt file that you can copy out of, or just email the whole thing. I'll put a sample up here that you can use for a guide, edit it as you see fit:

Code:

---

ECHO OFF
CLS
ECHO. >>C:\WUTEMP\Netstats\FFXIV_TRACE.TXT
ECHO %time% %date% >>C:\WUTEMP\Netstats\FFXIV_TRACE.TXT
ECHO Tracing path to 199.91.189.31
TRACERT 199.91.189.31 1>>C:\WUTEMP\Netstats\FFXIV_TRACE.TXT
ECHO Traces/Pings Complete
ECHO. >>C:\WUTEMP\Netstats\FFXIV_TRACE.TXT
pause
start c:\WUTEMP\NETSTATS\FFXIV_TRACE.TXT

---

Just open Notepad, copy that text in, then save it to your desktop or other convenient location as a .bat file. Make sure to open the File menu and click "Save As...", then change the file type to "All files (*.*)" under the file name field, and give it a name ending in .bat (like FFXIV_Trace.bat). Depending on your environment settings, it may save it as FFXIV_Trace.bat.txt if you don't change it to "All Files", and then it won't recognize it as an executable file and will just open it in notepad when you run it.

This is basically what the batch file is doing:
ECHO OFF turns off the echoing of typed text, keeps the screen from getting cluttered
CLS clears the screen
ECHO. >>C:\WUTEMP\Netstats\FFXIV_TRACE.TXT inserts a blank line into the file FFXIV_TRACE.TXT, which is created or saved in the folder path C:\WUTEMP\Netstats.. edit the path and file name to whatever you like
ECHO %time% %date% >>C:\WUTEMP\Netstats\FFXIV_TRACE.TXT adds the current time and date to the start of te trace log
ECHO Tracing path to 199.91.189.31 this is just a notice that will show on your screen... if you trace a different IP, it will still show this one if you don't edit it. This is just there to remind me, in case I forget to update it if my IP changes so I know I need to update the file.
TRACERT 199.91.189.31 1>>C:\WUTEMP\Netstats\FFXIV_TRACE.TXT this is calling the trace just like you normally would do in a command prompt, but it is redirecting the output instead to he text file
ECHO Traces/Pings Complete just another notice that shows up on your screen to let you know the trace completed. Did this since all output is sent to the text file--you don't see the results on screen
ECHO. >>C:\WUTEMP\Netstats\FFXIV_TRACE.TXT just inserting a blank line at the end of the session. Add the blank lines at start/end because it will append the results each time it is run. So if you run it twice, you get two traces, each with a time/date stamp, with a few blank lines between them
pause just pauses the script, and asks you to press any key before continuing. Did this because I will sometimes run this while in game, and the next line would cause the game to loose focus
start c:\WUTEMP\NETSTATS\FFXIV_TRACE.TXT this is a call to open the log file so you can review the results. After doing this, the command window will close on it's own


Alright! think this may be setting a record for a wall of text. Note that the folder path and the name of the text file are listed with a lot of commands. You will need to edit them each to keep it all together, or you will get weird results. May be easier to just create a WUTEMP folder at the root of your C: drive, and then add a NETSTATS folder inside that. The first time you run the script it will create the new text file, and each time you run it it will append the results to the same folder. If you want to archive the logs, just change the name of the log file (I just change it to the date it was run), and then it will just create a new file the next time you run it.

ok just woke up, nice to see you guys kept at it haha. ran a few tracerts and they did not have any problems at all. Hopped on the game and everything seems fine.

Final push to 50 can continue...for now.

I just woke up and I'm still getting 90k+2002, anyone else?

everything you said except your data center (im in whatever Gilgamesh is) is the same for me and i am also having these problems.

This sucks, so I called Cognetco this morning and they said they resolved their issue. but I am still getting the same lag and bullshit latency.

The lag is still here...and as a healing class the game is unplayable. The lag makes me unable to heal anybody and this causes wipes...and even when i play my dps classes the lag kills me with attacks I cant see...I cannot play this game till this is resolved. I have played since the relaunch and its never been this bad.

I have noticed this as well its about 7 pm central time and the Lag of doom happened the past two days. I cant even craft without the lag killing the experience. Come on fix the server issues already. :mad:

there addressing the issue with the maintance on http://forum.square-enix.com/ffxiv/t...18#post2072818

it isnt server issue its issue with routing to the server.