Rubber banding + Lag...

[hiimcody1]

EDIT: WOOOOO 24HRS TDM full, zero crashes xD

Congrats. I'll give you an hour before it bites the dust

[soja]

You must be hinting at a game update in an hour. That's the only way it goes down!

[Edge100x]

I like your optimism! We could really use a game update.

[marrr]

What is the current CPU usage on that machine?

CPU usage: 40.81%
Free physical memory: 114251mb

After playing a bunch tonight..
The server ran fine for 5 maps with 1 round each. 6th map MP_Resort and 7th MP_Journey = lag. I will see if the 8th map is lagged out too.

[soja]

What is the current CPU usage on that machine?

CPU usage: 40.81%
Free physical memory: 114251mb

After playing a bunch tonight..
The server ran fine for 5 maps with 1 round each. 6th map MP_Resort and 7th MP_Journey = lag. I will see if the 8th map is lagged out too.

Sounds like the issue most people are having, id have to say wait for a patch D:

[Edge100x]

40% is OK, and shouldn't cause usage-related problems, but it's little higher than what we'd like to see. We're still working on hardware rollouts to bring CPU usage down across the board and you should start seeing lower numbers during peak hours.

[soja]

John, have you ever experienced issues with "too many" servers running on one machine? Not running into hardware usage issues, but running into software issues?

[Mish1]

Our cpu usage at prime time is around 80% on a shared server.

It seems as if the longer the server runs populated then the lag increases. By the 5th or 6th map in the rotation the lag is very bad. Some maps seem to lag worse than others which is why we have removed them from our server. Those maps are Parcel Storm, Lancang Dam and Hainan Resort.

[woowoo]

Our cpu usage at prime time is around 80% on a shared server.

It seems as if the longer the server runs populated then the lag increases. By the 5th or 6th map in the rotation the lag is very bad. Some maps seem to lag worse than others which is why we have removed them from our server. Those maps are Parcel Storm, Lancang Dam and Hainan Resort.

Could be a memory leak which as a side effect can affect cpu usage.

[.=QUACK=.Major.Pain]

Could be a memory leak which as a side effect can affect cpu usage.

Sounds like it. The longer the server runs, the more it leads to an overload.

[10-78 Busterking]

Could be a memory leak which as a side effect can affect cpu usage.

Sounds like it. The longer the server runs, the more it leads to an overload.

I am having the same problem with 2 Arma3 servers, then support agreed with me that there was a memory leak. They need to be shut down at least once a day if not more.

I think it's all related to how NFO has their VDS setup, maybe overloading their system with servers.

[woowoo]

Could be a memory leak which as a side effect can affect cpu usage.

Sounds like it. The longer the server runs, the more it leads to an overload.

I am having the same problem with 2 Arma3 servers, then support agreed with me that there was a memory leak. They need to be shut down at least once a day if not more.

I think it's all related to how NFO has their VDS setup, maybe overloading their system with servers.

My guess its a problem with BF4. We saw the exact same thing with both BF3 and BF:BC2.

[Spray]

Could be a memory leak which as a side effect can affect cpu usage.

Sounds like it. The longer the server runs, the more it leads to an overload.

I am having the same problem with 2 Arma3 servers, then support agreed with me that there was a memory leak. They need to be shut down at least once a day if not more.

I think it's all related to how NFO has their VDS setup, maybe overloading their system with servers.

We certainly don't overload any of our systems. Our automated load balancing won't allow a machine to stay 24 hours with over 40% usage, we get alerts when usage is high, and we simply don't stand for it. We couldn't guarantee the best performance if we used practices like that.

[Edge100x]

I am having the same problem with 2 Arma3 servers, then support agreed with me that there was a memory leak. They need to be shut down at least once a day if not more.

Memory leaks are entirely in software and due to implementation bugs. ArmA3 is particularly bad about using large amounts of CPU and memory.

We restart servers once a day when they're empty automatically, which works best for most games. You can also have restarts performed more often. There's a control for this on the "Server control" page.

I think it's all related to how NFO has their VDS setup, maybe overloading their system with servers.

We do not overload our VDSes. In fact, we assign dedicated HT cores and memory, so there's a hard (and low) cap on how many customers can be put onto each machine.

Mish1, that would likely have been a one-night high number due to a sudden rush of people. 80% is far beyond our allowed threshold, so your machine would definitely have been automatically load-balanced today. If you see a number that high again, please contact us, and we can manually move you to another machine.

[Wraygun]

Getting lots of lag with R3. Anyone else?

I don't k now if it's network lag or game hitches/spikes.