From: messages [mailto:messages-bounces@vassalengine.org] On Behalf Of Joel
Uckelman
Sent: Friday, April 03, 2015 7:19 AM
To: The mailing list for VASSAL
Subject: Re: [messages] [Technical Support & Bugs] Very slow response
I noticed just now that I have a machine for which the route from it to vassalengine.org goes via Telia but not via AT&T—this happens to be my server in Tucson which is on a Cox business DSL line. This is what I see from traceroute:
I tried transferring a 6MB file from there to vassalengine.org and got around 890KB/s, which seems quite good. This makes me think the problem is not with Telia, either.
I have two friends on AT&T (one on U-verse, and the other on DSL I believe) that have been having intermittent problems with lag on the central Vassal server for several weeks now. Here are some TRACERT results:
In the meantime we’ve been using the P2P connection to connect everyone up together using a forwarded port, but ran into some problems with 6 people at once and no problems with only 4 people. Is there any additional documentation on using the P2P connection so we can try to smooth that out a bit more?
Someone posted on the U-verse support forums last month
(forums.att.com/t5/Troubleshoot- … -p/4203601[1]) but they
just got a copy/paste response and nothing else since then.
1 * * * Request timed out.
2 * * * Request timed out.
3 23 ms 22 ms 23 ms 99.134.205.78
4 27 ms 26 ms 26 ms 12.83.114.17
5 50 ms 37 ms 37 ms 12.122.117.97
6 34 ms 40 ms 35 ms 192.205.33.42
7 91 ms 62 ms 61 ms ash-bb4-link.telia.net [62.115.137.72]
8 137 ms 140 ms 141 ms prs-bb2-link.telia.net [80.91.251.246]
9 133 ms 148 ms 136 ms prs-b8-link.telia.net [213.155.136.209]
10 142 ms 142 ms 141 ms online-ic-305116-prs-b8.c.telia.net [62.115.40.8
6]
11 135 ms 138 ms 137 ms 45x-s31-1-a9k2.dc3.poneytelecom.eu [195.154.1.35
]
12 140 ms 141 ms 140 ms one.vassalengine.org [62.210.178.7]
Trace complete.
I’ve tried wired connection, I’ve tried DMZplus mode, still no solution. I have contacted AT&T and sent them the trace route, a link to this thread, and everything I’ve tried. I will update with a response.
Has anyone found a solution for this yet? This is very frustrating!
I had been running smooth without major porblems until a couple of days ago. Now, I’m back to having problems even though I thought the direct ethernet cable had fixed it. As such, I reported it to AT& T tonight.
General instructions for reporting to At&T via chat:
1)Go to att.com/esupport/index.jsp
2) On the right of the screen there will be a little “chat now” button. click on it
3) Answer a few questions and it will start a chat session.
4) The agent was helpful but wanted me to run speed tests and clear cache, etc. What seemed to get the point across was when I referenced this forum and topic. ( I gave them the link https://forum.vassalengine.org/t/very-slow-response/7535/1)
The issue is known, and they said they are working on it. If you are having the issue though and are an At&T customer, then it seems the only thing you can do to hurry it along is report it in this method.
By the by, find a bill. They will want your Uverse account number.
Any luck? It’s gotten really bad in recent days. I have ow talked to three differecent AT&T support people. Their offical position seems to be that it is not an At&T problem, but if I’d like to pay a one-time $50 fee , they wil reasearch the problem. I really don’t want to to do this expecially since I believe they will end up saying it is an “application” problem.
I doubt that the problem is a duff nameserver. Our server’s IP address has been correct in every traceroute posted here, including the traceroutes posted by people having the problem. If it really were caused by a bad nameserver, then you’d expect at least one of the traceroutes to show a DNS problem—either a failure to resolve vassalengine.org or an incorrect IP address for it.
I’m waiting for a reply from AT&T. I’ll keep you posted about what I hear.
I’m also an AT&T customer and having the same issues. Any news or update
here?
I’ve heard nothing back from AT&T. The last email I sent was 29 April.
I’ll try once again later today, and if that doesn’t get a reply, I’ll
try to get their attention on Twitter a second time.
I’m still trying, with no success thus far, to get the attention of someone at AT&T who can help us. Has anyone who is their customer had any luck with calling AT&T’s tech support?
We had an AT&T player last week and the lag was so bad we could not play. He turn on his AT&T hotspot and used that and it worked fine. It was only the home network that was causing the extreme lag.
Interesting. I had abysmally slow response with both my U-Verse home network and using my AT&T phone as a hotspot, but when I switched to my (Verizon) work phone as a hotspot, performance was immediately and drastically better.
I have a few friends at AT&T; I’ll see if they have any suggestions for a good path to follow to try and get a resolution.
Here are the traceroutes for both connections. Interestingly, the tracert on Verizon was slower and seemed to have more issues than the AT&T one, but actual server response within VASSAL was dramatically better.
1 <1 ms <1 ms <1 ms 172.20.10.1
2 114 ms 30 ms 35 ms 172.26.96.161
3 36 ms 38 ms 35 ms 172.18.236.228
4 29 ms 35 ms 33 ms 12.249.2.17
5 47 ms 159 ms 88 ms 12.83.170.18
6 161 ms 137 ms 37 ms gar24.attga.ip.att.net [12.122.141.181]
7 281 ms * 114 ms 192.205.33.42
8 94 ms 44 ms 69 ms ash-bb4-link.telia.net [62.115.137.74]
9 257 ms 177 ms 142 ms prs-bb2-link.telia.net [80.91.251.103]
10 387 ms 134 ms 138 ms prs-b8-link.telia.net [213.155.131.17]
11 259 ms 141 ms 136 ms online-ic-305116-prs-b8.c.telia.net [62.115.40.86]
12 162 ms 138 ms 144 ms 45x-s31-1-a9k2.dc3.poneytelecom.eu [195.154.1.35]
13 148 ms 137 ms 138 ms one.vassalengine.org [62.210.178.7]
I spoke with a higher-tier rep from AT&T for about 45 minutes tonight. He was very friendly and helpful, but unfortunately he didn’t have access to go much deeper than the traceroutes and such we’re already doing. He suspects that it’s something in the way the traffic is routed after it leaves the AT&T network, because the response times on the first few hops are generally good, so the next step is probably investigating some kind of peering relationship between AT&T and whatever other ISPs they route traffic through to get to vassalengine.org. He’s checking with his manager tomorrow to see what next steps we might be able to take; however, he acknowledged that in order to get attention at a higher level, it will likely take more than one or two customers reporting an issue and some posts on an internet message board.
I’ll post whatever I hear back, but in the meantime, if you are an AT&T customer that’s having a performance problem with VASSAL, please follow the steps recommended in revans’ post above and report the issue. Every little bit helps, and if we can squeak loud enough (POLITELY please), we may be able to get some grease.