Amazon's cloud issues..UO's nearly unplayable

[LetheGL]

Anyone else have horrid lag since the switch? Running tracert for Atlantic shows 11+ hops around EUROPE before hitting the Amazon servers on the WEST coast even though Atlantic is supposed to be an EAST coast server. Seems such a shame to have a fun game made nearly unplayable because of poor connections.

http://stratics.com/community/threads/help-needed-to-track-down-disconnection-problem.304285/

http://stratics.com/community/threads/constant-server-disconnects.308055/

http://stratics.com/community/threads/connection-issues.308954/

http://stratics.com/community/threads/lag-in-game-but-ping-stable.310487/

 

[Petra Fyde]

That's not necessarily because of the Cloud. For some months a few years ago most mainland Europa players were directed to Europa and Drachenfels via Washington curtesy of Sprintlink. It's not the cloud that's at fault, it's the routing to it.

 

[Lady Storm]

I have been seeing weird lag... i would type speach to someone and the time difference between what i do and what happens is at the very least to a 10 second or more time gap.
Actions in game have been to the point i dont want to leave the bank unless absolutely needed. This is random times and I cant pinpoint the cause.

 

[kongomongo]

If that was a general problem with Amazons cloud service noone would use it.

 

[Lady Storm]

I have to agree with Petra, many many years ago AT&T had a major router go belly up and we ended up circling the country to get access to UO. Made for some mad lag.
Also many Cable and satilite wed providers have had troubles that add in to the lag issue.

 

[LetheGL]

If that was a general problem with Amazons cloud service noone would use it.

Makes sense..Guess I'm just grumpy and aggravated. Anyways, I just downloaded the WTFast trail and I'm loving it so far. Seems to make a big difference.

 

[claudia-fjp]

If that was a general problem with Amazons cloud service noone would use it.

You say that like we have a say in the matter. Companies use it to save money, it's about the bottom line for them. End users wouldn't use it but we have no choice. There has been several noticable problems since the switch.

 

[Ender]

You say that like we have a say in the matter. Companies use it to save money, it's about the bottom line for them. End users wouldn't use it but we have no choice. There has been several noticable problems since the switch.

I think he meant companies wouldn't use it. And they probably wouldn't. Netflix uses it I think, and it works great.

 

[cazador]

The switch has brought me from the 18-21ms pings to 80-140ms I'm sure it's just my internet though :rollseyes:


Sent from my iPhone using Tapatalk

 

[Cetric]

same, i went from 30-45ms to 80-150ms avg when that change happened

 

[Promathia]

UO is the only MMO I play where I lag, and it really started happening when the switch was made.

 

[Flutter]

same

 

[LetheGL]

Well, I'm glad to see I'm not the only one aggravated with the crappy pings!

 

[Orgional Farimir]

Same. Add in the fact that uo assist doesn't work with windows 8 is why I canceled my 2 accounts

 

[claudia-fjp]

I think he meant companies wouldn't use it. And they probably wouldn't. Netflix uses it I think, and it works great.

There is a big difference between what netflix uses it for and what EA uses it for. Sure clouds work great for storing movies and tv (until content you paid for is yanked ala Prep and Landing by Disney on Amazon) but GAMING on the cloud is a whole other beast and I'm not impressed and very unhappy with performance. Even logging in is a trial these days sometimes.

 

[CovenantX]

I don't think it's the 'Cloud' causing the problems, more so the hops on the way to it. I move fine most of the time, the only real lag issue I had (did coincidentally occur after the cloud move) But it was caused by the whole Alter.net issue which was NOT the cloud.

Lately I've been laggy off and on in spikes, but not consistently, It could be the cloud, but I doubt it, as I've not been laggy more so before or after the move, with the exception of the alter.net issue I talked about above.

Also, if it were the cloud, do you think Lower-ping or WTFast would help at all ? If it does, it's probably not the cloud.

 

[Lythos-]

I went from pinging 12-15 to mid 60's. Even using a proxy I am still over double what I'm used to. Without it I literally take a few steps and lag on a good day and on bad days I take a few steps and freeze for a few.

Point is, we should all have smooth gameplay with no need to pay extra for a proxy to be able to play a 16 year old game that's supposedly based on our own coast...

 

[kongomongo]

Running tracert for Atlantic shows 11+ hops around EUROPE before hitting the Amazon servers on the WEST coast even though Atlantic is supposed to be an EAST coast server

I know this discussion is going pretty much nowhere since "lag" is such a simple word for something like a TCP/IP connection. Anyways, since noone even cared to post a traceroute I'll take the above quote which shows that in this particular instance the "cloud" has NOTHING to do with the high latency, as the latency already builds up on the route to the cloud. So the "lag" happens before you're even hit the EC2 service.

And saying EC2 just works for Netflix?? What do you base that claim on? For reference: http://aws.amazon.com/game-hosting/

tl;dr
Probably routing problem, you can't fix it, UO devs cant fix it. Run a tracert and try to contact the hop that is causing lag is probably your best bet.

Oh and I ping 10ms, so I am _pretty_ sure it's not the server...

flame on!

 

[Lady Storm]

Kongomongo calm down many know that its not the game or the cloud...
It would surprise all how many bounces it takes for a signal from your comp to the cloud or vice versa. It's no where near even a few ... Any one could be the trouble.

 

[Lefty]

Same. Add in the fact that uo assist doesn't work with windows 8 is why I canceled my 2 accounts

I'm running Win 8.1 and UOAssist is working fine. Perhaps you need to run as administrator.

 

[Orgional Farimir]

I'm running Win 8.1 and UOAssist is working fine. Perhaps you need to run as administrator.

Tried that. It works just fine on my old computer running xp, but when I loaded it on my new computer with windows 8 uoa won't load.

 

[Goldberg-Chessy]

I have recently been experiencing some minor lag on Atlantic.
Not enough to make the game unplayable but pretty annoying and it has gotten me killed a couple times.

 

[LetheGL]

I know this discussion is going pretty much nowhere since "lag" is such a simple word for something like a TCP/IP connection. Anyways, since noone even cared to post a traceroute I'll take the above quote which shows that in this particular instance the "cloud" has NOTHING to do with the high latency, as the latency already builds up on the route to the cloud. So the "lag" happens before you're even hit the EC2 service.

And saying EC2 just works for Netflix?? What do you base that claim on? For reference: http://aws.amazon.com/game-hosting/

tl;dr
Probably routing problem, you can't fix it, UO devs cant fix it. Run a tracert and try to contact the hop that is causing lag is probably your best bet.

Oh and I ping 10ms, so I am _pretty_ sure it's not the server...

flame on!

I agree it's probably not the cloud itself but the route to it that's the problem. Although if EA wasn't using Amazon's cloud we wouldn't be using this piss-poor route to the cloud servers. So it's pretty much a problem caused by using the cloud...
Trace routes and these are actually really good today. All my problems come from TeliaSonera International where the route hops arounds Europe. Those are typically 80-100+ ms each.

SONOMA

4 8 ms 10 ms 11 ms rrcs-98-103-112-25.central.biz.rr.com [98.103.112.25]
5 15 ms 15 ms 15 ms be22.pltsohae01r.midwest.rr.com [65.189.179.128]

6 13 ms 15 ms 15 ms be25.clmkohpe01r.midwest.rr.com [65.29.1.28]
7 26 ms 27 ms 23 ms 107.14.19.60
8 24 ms 28 ms 26 ms ae-0-0.cr0.chi10.tbone.rr.com [66.109.6.20]
9 24 ms 23 ms 22 ms ae0.pr1.chi10.tbone.rr.com [107.14.17.192]
10 24 ms 55 ms 24 ms chi-bb1-link.telia.net [213.248.76.97]
11 86 ms 86 ms 84 ms sjo-bb1-link.telia.net [213.155.132.180]
12 85 ms 86 ms 86 ms a100row-ic-300117-sjo-bb1.c.telia.net [213.248.87.118]
13 86 ms 87 ms 87 ms 205.251.229.179
14 93 ms 92 ms 94 ms 205.251.229.233
15 93 ms 95 ms 93 ms ec2-54-241-130-239.us-west-1.compute.amazonaws.com [54.241.130.239]

Trace complete.

ATLANTIC

4 10 ms 9 ms 10 ms rrcs-98-103-112-25.central.biz.rr.com [98.103.112.25]
5 16 ms 15 ms 15 ms be22.pltsohae01r.midwest.rr.com [65.189.179.128]

6 17 ms 15 ms 16 ms be25.clmkohpe01r.midwest.rr.com [65.29.1.28]
7 25 ms 28 ms 27 ms ae-4-0.cr0.chi30.tbone.rr.com [66.109.6.68]
8 28 ms 27 ms 26 ms ae-0-0.cr0.chi10.tbone.rr.com [66.109.6.20]
9 24 ms 24 ms 24 ms ae0.pr1.chi10.tbone.rr.com [107.14.17.192]
10 26 ms 24 ms 25 ms chi-bb1-link.telia.net [213.248.76.97]
11 35 ms 35 ms 36 ms nyk-bb1-link.telia.net [213.155.131.240]
12 45 ms 54 ms 43 ms ash-bb3-link.telia.net [213.155.134.149]
13 66 ms 54 ms 85 ms ash-b1-link.telia.net [80.91.248.159]
14 36 ms 36 ms 36 ms vadata-ic-157232-ash-bb1.c.telia.net [62.115.9.66]
15 37 ms 47 ms 37 ms 72.21.220.55
16 51 ms 37 ms 37 ms 72.21.222.89
17 * * * Request timed out.
18 * * * Request timed out.
19 * * * Request timed out.
20 37 ms 37 ms 37 ms ec2-107-23-95-234.compute-1.amazonaws.com [107.23.95.234]

Trace complete.

 

[Winter]

... Trace routes and these are actually really good today. All my problems come from TeliaSonera International where the route hops arounds Europe. Those are typically 80-100+ ms each.
...

That is a lot of hops, but your ping times don't look that bad to me. Any of the hops in my route test aren't as good as your worst, and look fine for PvM, not sure about PvP.

The "Request timed out" hops are probably servers that don't allow pings, and you won't ever get a ping time.

 

[kongomongo]

I agree it's probably not the cloud itself but the route to it that's the problem. Although if EA wasn't using Amazon's cloud we wouldn't be using this piss-poor route to the cloud servers. .

Yea true.

Looking at your trace telia is to blame. With my route we had 2 months of terrible ping AND packet loss because of an over congested route. I had great success using a free vpn service at that time to alter the route and circumvent the bad hop.

 

[G.v.P]

Where can one find the new server names to test tracert?
http://stratics.com/community/threads/shard-ip-addresses.249180/

How do you perform a trace route?
Start>Command Prompt>
--> tracert [address], hit enter/return
Example: tracert 107.23.147.156

How do you perform a ping?
Start>Command Prompt>
--> ping -n [#] [address], hit enter/return
Example: ping -n 100 107.23.147.156 (this will perform a ping #100 times)

 

[G.v.P]

Hm.

Route To Chesapeake:
tracert 107.23.147.156

[...]
Hop # 7: 10 ms, 11 ms, 10 ms, 72.21.221.78
Hop # 8: 14 ms, 16 ms, 13 ms, 72.21.220.31
Hop #9: 11 ms, 11 ms, 11 ms, 205.251.245.55
Hop #10, 11, 12: *, *, *, Request timed out.
Hop #13: 17 ms, 10 ms, 13 ms, ec2-107-23-147-156.compute-1.amazonaws.com

Ping to Chesapeake:
ping -n 100 107.23.147.156

Ping: 100/100, 0% loss, 10 ms-214 ms, 20 ms avg.

Looks like everything is pretty consistent for me...I usually ping 80-120 ms in the other game I usually play .

 

[Winter]

Instead of using the command line to ping the servers, the UO Trace app is very handy, very nice. I use it for non-UO servers too when my ISP has issues.

Here is the EA Help link to the tool with instructions: http://help.ea.com/en/article/how-to-uo-trace/