It is easy to say, all is fine at my side, it has to be you - but that's not half the truth...
Server
DoP - Grid 1 showed up laggy yesterday. Yes, that's what we've all seen. I don't remember exactly, but 7-9 people disconnected due to network problems. More than 40 were online - having more or less lag.
We had also an issue with teamspeak and our homepage. Maybe it's important to know that these two things are hosted on a totally different server. I don't really know but I was told that one is hosted in Cologne and the other in Frankfurt.
So analyses may start with an eye on the facts.
First of all I don't want to conceal that the server hosting the LfS-Dedicated Hosts is used for other things too. Yesterday there were some people uploading some backups via FTP. Although the ftp-connections are limited to a maximum, they could cause lag to some people. Those backups are made from Monday till Friday in the evening, so that may be the reason why troy experienced lag more than once in the past.
But this is no explanation for the trouble with teamspeak!
Next thing is to check the connection from point too point. A simple ping to "wherever" doesn't really help. (The answers from Windows are in german language. If you perform the commands on your machines you'll get them in your preferred one. The words doesn't matter too much in my explanation.)
A ping to google.de may show:
Ping-Statistik für 173.194.32.216:
Pakete: Gesendet = 4, Empfangen = 4, Verloren = 0
(0% Verlust),
Ca. Zeitangaben in Millisek.:
Minimum = 21ms, Maximum = 21ms, Mittelwert = 21ms
A ping to the BurnoutCrew-Server shows from my current location:
Ping-Statistik für 81.20.142.84:
Pakete: Gesendet = 4, Empfangen = 4, Verloren = 0
(0% Verlust),
Ca. Zeitangaben in Millisek.:
Minimum = 18ms, Maximum = 23ms, Mittelwert = 19ms
Not much information when everything is ok.
Sad but true, the internet isn't a thing of Clients an Servers as we see with a simple ping. A network packet is travelling to different points in the world before it - hopefully - finally reaches the server. A simple thing to have a look after that routing is to perform a trace route (
tracert in Windows console). This should give you some information about the time a packet looses during its way through the net.
In case of google it looks like this in my case:
Command:
tracert www.google.de
Routenverfolgung zu google.de [173.194.39.31] über maximal 30 Abschnitte:
1 <1 ms <1 ms <1 ms 192.168.250.254
2 1 ms 1 ms 1 ms 217.89.123.97
3 10 ms 8 ms 9 ms 80.154.240.37
4 16 ms 14 ms 15 ms hh-ea7-i.HH.DE.NET.DTAG.DE [62.154.33.2]
5 14 ms 14 ms 14 ms 80.150.170.98
6 15 ms 15 ms 15 ms 209.85.249.124
7 15 ms 15 ms 15 ms 209.85.245.165
8 15 ms 15 ms 15 ms ham02s13-in-f31.1e100.net [173.194.39.31]
Ablaufverfolgung beendet.
And for our server:
tracert 81.20.142.84
Routenverfolgung zu 81.20.142.84 über maximal 30 Abschnitte
1 <1 ms <1 ms <1 ms 192.168.250.254
2 1 ms 1 ms 1 ms 217.89.123.97
3 9 ms 9 ms 9 ms 80.154.240.37
4 18 ms 18 ms 17 ms 217.239.50.81
5 18 ms 17 ms 17 ms 80.157.131.218
6 19 ms 18 ms 18 ms core-sto2-po9.netcologne.de [81.173.192.9]
7 19 ms 19 ms 19 ms rtkds-sto-po2.netcologne.de [87.79.16.58]
8 28 ms 19 ms 19 ms backbone-nc-cgn.surfplanet.de [85.88.20.62]
9 19 ms 18 ms 18 ms 81.20.142.84
Ablaufverfolgung beendet.
We can see most loss of time happens from point 2 to 3 and 4. This is the local in-house router to the local provider. Step 8 and 9 are localized in the data center, well step 9 is the server itself.
In case of lag or connection problems I would like to see this trace route to get more than an idea where the time loss happens.
I tested this trace route during a bigger ftp-upload without limited traffic speed. The connection is less stable than before. The worst report I found was this:
Routenverfolgung zu 81.20.142.84 über maximal 30 Abschnitte
1 <1 ms <1 ms <1 ms 192.168.250.254
2 4 ms 1 ms 1 ms 217.89.123.97
3 10 ms 18 ms 27 ms 80.154.240.37
4 16 ms 16 ms 57 ms 217.239.50.81
5 26 ms 34 ms 17 ms 80.157.131.218
6 31 ms 35 ms 18 ms core-sto2-po9.netcologne.de [81.173.192.9]
7 18 ms 18 ms 19 ms rtkds-sto-po2.netcologne.de [87.79.16.58]
8 19 ms 19 ms 19 ms backbone-nc-cgn.surfplanet.de [85.88.20.62]
9 52 ms 38 ms 32 ms 81.20.142.84
Ablaufverfolgung beendet.
Not that bad anyway but a possible cause for lag experiences.
Ok, this was what you can do to check your connection to the server and help me to find problems if there are some at server side.
What we do to get the server as stable as possible:
1) FTP will be disabled
2) Tools for Sign-In and Qualify and so on will be stopped
3) There are many LfS-Hosts on this machine. Most of them will be stopped as well.
So hopefully no one has to disconnect due to network problems but I feel like some people will have problems as every year.