notfred messing with my internet

Cypher-

[H]ard|Gawd
Joined
Dec 5, 2007
Messages
1,321
For some reason notfred is seriously messing with my internet. Ive been having bad latency over the last few days and figured it was just my isp. I have narrowed down the problem to notfred. This is what I get when I ping my router with it running:

Ping statistics for 192.168.1.1:
Packets: Sent = 21, Received = 21, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 0ms, Maximum = 57ms, Average = 16ms

without:

Ping statistics for 192.168.1.1:
Packets: Sent = 26, Received = 26, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 0ms, Maximum = 1ms, Average = 0ms


Any ideas on how to fix this? I tested my XP virtual machine and it is not effecting the internet at all so I can safely say that notfred is the problem.
 
I'm clueless about network stuff, but with notfreds running on my setup (Win7x64) I'm getting an average of 2ms between my computer and router over WiFi.

What did you mean about an XP VM? Running XP in VMware doesn't affect latency? Is the XP VM putting any serious load on your system? Try putting your system at 100% load without using notfreds and check latency again.
 
The only time I ever have latency issues with the Linux SMP client is when a workunit is in the process of being uploaded. Is that what you're seeing, or is this lag occurring all the time when the VM is running even when it isn't uploading or downloading? If it's happening all the time, try going into your router settings and enabling its QoS (quality of service) feature.
 
SMP units are big, they will take a bit to upload. Unless you have a good router that will load balance, it could mess things up for a bit.
 
I worry about this problem with my new place. The internet service is 128k upload and download and based on what I have seen so far with this local phone co-op that may be best-case speeds. I wonder how long it will take to upload/download work units. I may have to get some traffic shaping and such if it becomes a problem. I only have 1 SMP VM running, the rest is GPU.
 
my guess is the lags coming from the way you have the VM's setup on your system.. if they are bridged or direct connect.. if they are bridged its possible its causing the lag..
 
my guess is the lags coming from the way you have the VM's setup on your system.. if they are bridged or direct connect.. if they are bridged its possible its causing the lag..
They still shouldn't cause lag no matter what unless there's an active connection.
 
It also doesnt help that most mobos have RealTrash LAN controllers that blow goats to start with.
 
They still shouldn't cause lag no matter what unless there's an active connection.


im just coming up with idea's for people to branch off of to find a solution because i never had an issue when i use to run the VM's.. but i was using NAT instead of bridged connections..
 
I have a Comcast with a 16/2 line. The WU's shouldn't take more then a few minutes to upload. This latency is whenever notfred is running period and doesnt matter if its uploading a WU or not. In any case the ping was from my pc to my router so the internet connection doesnt really come into play.

I currently have notfred set up using a bridged connection. I also have a virtual machine of XP set up using the same bridged connected and there is no latency what so ever when I run that. Even if I try using NAT there is still no improvement.

I just found out that our school offers a free MSDNAA subscription so I'm downloading Windows 7x64 professional to upgrade from RC1. Sometime soon I'll format and hopefully that will fix it if there's just something wacky with my pc.

EDIT:
jimbowdang said:
What did you mean about an XP VM? Running XP in VMware doesn't affect latency? Is the XP VM putting any serious load on your system? Try putting your system at 100% load without using notfreds and check latency again.

We have a winner. When I have xp at idle there is no problem. I put prime95 on it (and this is only using 1 core since its 32bit) and if I have that running the problem comes back. As soon as I stop it everything is fine. If I run prime95 on my own system I do not have the problem either so the problem is being caused by the vmplayer being under load.

Anyone ever hear of this?
 
Last edited:
i suggested this to evil last night.. try looking at your NIC configuration and see if there is an option called optimized cpu usage or something along those lines.. the only system i have that uses that option is turned off right now since im to lazy to fix the memory issue with it.. but if it does you could try messing with that and see what happens..
 
If the computer is bogging down when the VM's are at full load then you could try useing "WinAFC" to set the VM's priority to low.
This should free up CPU cycles for other work as needed.

Luck ......... :D
 
If the computer is bogging down when the VM's are at full load then you could try useing "WinAFC" to set the VM's priority to low.
This should free up CPU cycles for other work as needed.

Luck ......... :D

For some reason I cannot get WinAFC to pick up vmware-vmx. However, task manager>vmware-vmx>priority below normal seems to have worked wonders.

Thanks for all the help guys.
 
Back
Top