[CS:GO] VPS Server: Rubberbanding

[sobs.ml]

Hi,

Recently I started experiencing random rubberbanding across my servers (seems to be a correlation with players connect but hard to tell).
I talked to the support and they ended up changing machine to ensure hardware wasn't the issue. The problem is still happening and I've been told by the support to ask here in hope to find a solution to my problem.

Both my RAM and CPU are more than enough: even with full servers the CPU stands at around 70%.

I need help as soon as possible.
Thanks in advance!

[sobs.ml]

Hi,

Recently I started experiencing random rubberbanding across my servers (seems to be a correlation with players connect but hard to tell).
I talked to the support and they ended up changing machine to ensure hardware wasn't the issue. The problem is still happening and I've been told by the support to ask here in hope to find a solution to my problem.

Both my RAM and CPU are more than enough: even with full servers the CPU stands at around 70%.

Edit: Wasn't sure in which section to put it since it belong'd in two since the source of the rubberbanding is unknown.
Original Thread: viewtopic.php?f=44&t=13737

I need help as soon as possible.
Thanks in advance for your support!

[soja]

Managed or unmanaged?

Windows or Linux?

What CPU is the VDS currently on?

How big are the servers?

Tickrate of the servers having issues?


at 70% CPU you are pretty much at the top end limit of where you want to be for good performance. You may want to consider adding an additional two HT cores to your package.

[Edge100x]

One of the first things you'll want to check here is the net_graph output. You'll want to look for packet loss and choke there, and see how much the server FPS is dipping.

[sobs.ml]

Managed or unmanaged?

Windows or Linux?

What CPU is the VDS currently on?

How big are the servers?

Tickrate of the servers having issues?


at 70% CPU you are pretty much at the top end limit of where you want to be for good performance. You may want to consider adding an additional two HT cores to your package.

Unmanaged?
Windows 2012 R2
Standard for VPS
from 20 to 32 slots
from 64 to 102.4

@edge
Yes of course it's the first thing I checked and nothing seems to change on the net_graph when the spikes/rubberband happen this is why I came here.
I know how an overloaded server looks like and none of these symptoms show up.

[Edge100x]

Have you tried turning off logging (and anything else writing to disk) temporarily to see if it is I/O related?

[sobs.ml]

Have you tried turning off logging (and anything else writing to disk) temporarily to see if it is I/O related?

Good idea!
Mind telling me how to do so? I know about "sv_logs" but that's pretty much it, never was in a situation where I needed to disable logging.

[Edge100x]

The rcon command "log off" will do this.