So ... I'm not affected by this, but a friend is, so I have a interest in seeing this resolved. Unfortunately that means that I can't really do any debugging from my end of things.

From reading this thread, I've gathered that there are 2 main groups that are experiencing this issue ( or at least are vocal about it): FIOS and Malaysia.

To give a relevant networking crash course: The way the net works is that it isn't just a straight connection between you and the Final Fantasy server - because there is no direct cable connecting you and them. Instead, your packets travel through multiple routing server hubs. Some call em nodes, some call em hops, just keep in mind that there are more things involved than just "you" and "final fantasy server".

Now, due to the fact that only a specific set of people are experiencing this issue, I will assume that the problem isn't directly on the Final Fantasy server, but rather on a routing server inbetween - that would explain why this problem seems largely isolated to people who have their location in common.* ( It is only possible for this to be a Final Fantasy server problem if their firewall or router is improperly configured for a specific IP set, however I doubt that as they don't seem to employ IP blocking. It would be pointless overhead. )

So, getting back to the matter at hand: debugging the problem.

The goal is to determine the point of failure.

While this won't immediately solve the issue, it will let you contact your ISP with concrete information to ask for a fix ( something like "My connection to canada has to go through server <hostname><ip> and it seems to go unresponsive every 2 minutes." )

They might still be difficult about it and not take you serious, but by having some actual data they might be more cooperative.


The traceroute is a excellent idea, but you essentially need to keep repeating it until the traceroute itself hits the point of failure - which would be a timeout.
It would also need to be repeated multiple times to get hard data on this ( is it always the same route server that fails, or maybe a set of them ? ).

More info on tracert can be found here: http://support.microsoft.com/kb/162326

To assist in gathering the data, I've written a small batch script.
Create a new text file somewhere on your computer, for example at C:\ for easy access.
Name it something like nettest.bat, when prompted that it may make the file unusable, click yes.
Right click the file and select edit, a text-editor will open.
Paste the following:
Code:
@echo off
set /a "wait = 5"
set "server=neolobby02.ffxiv.com"
:while1
echo Checking server. Output goes to log.txt
echo %date% %time% >> log.txt
tracert -d %server% >> log.txt
echo Press CTRL + C To Stop!
echo Waiting %wait% seconds ...
timeout /t %wait% /nobreak
goto :while1
endlocal
Now just double-click the file and it will start querying the final fantasy lobby server. All US/EU servers in that center, so you don't need to query the actual game server itself.

What this does is:
It will automatically, every 5 seconds do a tracert command to the lobby server, saving the output in a log.txt file.

So, let this run for a while. Maybe 20 or 30 minutes. Close it with CTRL + C and then search the log.txt file for occurences of "*" instead of a time - the star denotes a timeout.

Please report back with information such as:
- How long you've let the test run
- Which IP/Server had the most * / timeout
- General frequency of * occuring.

Make sure you also state where you are from, so the data can be attributed to the proper group.

I'm hoping that we can see a common trend among multiple people from a group, so that we can narrow it down to a single, or a handful of servers causing issues.