Jump to content

Changing MTU?


Marcin541

Recommended Posts

You can't change it? I've seen many programs that allow you to. If it isn't changing it' date=' what is it doing to it?[/quote']

If your ISP has set MTU to 1492 there is no way you can overide it to 1500

you can change it lower eg 576, but who wants that, that only give a lower speed.

Path MTU detection works fine, so there is no reason to mess with these settings.

VanBuren :)

Link to comment
Share on other sites

  • 4 weeks later...
:D :D I have a program that is called a Rwin calculator it is made by Big Amp if some body will help me to post the file anyone is welcome to it maybe I can do a screen shot of it and somebody may recognize it this program has mtu mss and rwin :D :D
Link to comment
Share on other sites

:D :D I have a program that is called a Rwin calculator it is made by Big Amp if some body will help me to post the file anyone is welcome to it maybe I can do a screen shot of it and somebody may recognize it this program has mtu mss and rwin :D :D

Cool, just zip it and click Additional Options... when you write your post, that allow you to Attach files to your post.

Alot of tweak programs has those features too, DR.Tcp has MTU, TCP optimizer has also MTU, Cablenut has MSS thats MTU-40.

My experiance is that there is no need to use those features if you have MTU detection enabled, its enabled by default in most OS.

VanBuren :)

Link to comment
Share on other sites

Happy :D Holidays van the man that is very correct however if your isp has turned off the icmp ping options in their net work how can you effectively calculate latency for max rwin values in other  words that is exactly what my isp has done to me as to not recieve the denial of service attacks that our network has been plaqued with in recent months or they could be just pumping smoke up my ass also I know if you change mtu in the tcp optimizer program it will ask you to return to a mtu of 1500 but I can only set the amount of data to be sent at 995 bits of data or it will start to defrag at any amount higher,  strange could you educate me on latency and Rwin or  :?::?: explain 1492 mtu for PPPoe which is set by the isp  or 576 for dial up I really do'nt know or understand some of these formulas and they have kind of all melted together after 2 years of trying to get top performance out of this &{;" thing

Link to comment
Share on other sites

Happy :D Holidays van the man that is very correct however if your isp has turned off the icmp ping options in their net work how can you effectively calculate latency for max rwin values in other  words that is exactly what my isp has done to me as to not recieve the denial of service attacks that our network has been plaqued with in recent months or they could be just pumping smoke up my ass also I know if you change mtu in the tcp optimizer program it will ask you to return to a mtu of 1500 but I can only set the amount of data to be sent at 995 bits of data or it will start to defrag at any amount higher,  strange could you educate me on latency and Rwin or  :?::?: explain 1492 mtu for PPPoe which is set by the isp  or 576 for dial up I really do'nt know or understand some of these formulas and they have kind of all melted together after 2 years of trying to get top performance out of this &{;" thing

I suggest you use a ping value between 200 and 300 and see what works best for you, try 200 and raise the value to 230 and see if it goes any faster.

As you probl understand its hard to find a good latency value to use by downloading several files until your out of bandwidth, but there is no other way to go beside just testing values. My experiance is that a value between 200 and 300 is optimal.

I never heard of a ISP service before that defrag packets bigger then 995 byte, try cleaning up your registry, in TCP optimizer where you see MTU, delete the value you got there so the field is blank and click save. Also in cablenut program, click delete cablenut tweaks and click save to registry and reboot.

Now all settings should be as windows default.

run this analyzer and post result URL here so i can see it

http://www.dslreports.com/tweaks

here is some info regarding MTU http://www.dslreports.com/faq/7801

and here is some info befind calculating RWIN http://www.dslreports.com/faq/2024

quote "RWIN is determined mathematically. The required variables for the equation are Maximum Potential Speed (MPS) of the line, Maximum Segment Size (MSS) of the packet and Return Trip Time (RTT)(which most people measure with "ping" tests). Depending on the type of connection, a multiplier is usually applied to RTT to account for bad ping days. But generally the simple construction of the equation would be (MPS * Average Worst RTT)/8. The result is then converted to the nearest multiple of MSS."

Edit: here is a example  http://www.dslreports.com/faq/6181

quote"Example: MSS is 1460, RTTavg is 125ms, and advertised download speed is 1500kbps.

125 x 1.5 = 187.5

187.5 x 1500kbps = 281250

281250 divide by 8 = 35156

35156 divide by 1460 (MSS) = 24.07

Round up to next even number = 26

26 x 1460 = 37960 (Optimal RWIN)."

hope this helps and happy holidays Fred :)

VanBuren :)

Link to comment
Share on other sites

thank you van it will take a while to digest all of this info but I am willing to apply all of this info so as not to be left in the dark like i said after 2 years of trying to come up with the best rwin for my paticular computer it is worth the invested time :D

Link to comment
Share on other sites

there sure is a lot of strange stuff going on but I will prevail I have'nt been this challenged in a long time thanks again Van theMan

:::.. Download Stats ..:::

Connection is:: 2015 Kbps about 2 Mbps (tested with 1013 KB)

Download Speed is:: 246 KB/s

Tested From:: http://www.testmy.net/

Bottom Line:: 36 times faster than 56K you can download 1MB in 4.16 second(s)

Validation Link:: https://testmy.net/cgi-bin/get.cgi?Test_ID=WOCNP1TXF

could not get the tweak to work server busy I guess will keep on trying though I was able to get some ping times from a site that I will not even mention the name HaHa ;)

======================================================================================

=== VisualRoute ® 2005 Server Edition (v9.0a) report on Dec 29, 2004 11:10:23 PM ===

======================================================================================

Report for www.testmy.net [67.19.36.6]

Analysis: 'www.testmy.net' [6.67-19-36.reverse.theplanet.com]  was found in 12 hops (TTL=54). It is running

a HTTP server on port 80 (Apache/1.3.33 (Unix) mod_auth_passthrough/1.8 mod_log_bytes/1.2

mod_bwlimited/1.4 PHP/4.3.10 FrontPage/5.0.2.2635 mod_ssl/2.8.22 OpenSSL/0.9.7a).

-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

| Hop | %Loss | IP Address    | Node Name                                | Location        | Tzone  | ms | Graph      | Network                                                      |

-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

| 0  |      | 161.58.180.113 | WIN10115.visualware.com                  | *                |        |    |            | Verio, Inc. VRIO-161-058                                      |

| 1  |      | 161.58.176.129 | -                                        | ?Englewood, CO  |        | 0  | x          | Verio, Inc. VRIO-161-058                                      |

| 2  |      | 161.58.156.140 | -                                        | ?Englewood, CO  |        | 0  | x          | Verio, Inc. VRIO-161-058                                      |

| 3  |      | 129.250.28.206 | xe-1-2-0-3.r20.asbnva01.us.bb.verio.net  | Ashburn, VA, USA | -05:00 | 0  | x          | Verio, Inc. VRIO-129-250                                      |

| 4  |      | 129.250.5.34  | p16-0-1-1.r21.dllstx09.us.bb.verio.net  | Dallas, TX, USA  |        | 31 |      x-    | Verio, Inc. VRIO-129-250                                      |

| 5  |      | 129.250.28.167 | ge-1-0-0.a00.dllstx09.us.ra.verio.net    | Dallas, TX, USA  |        | 31 |      x-    | Verio, Inc. VRIO-129-250                                      |

| 6  |      | 129.250.28.186 | ge-1-2.a00.dllstx04.us.ra.verio.net      | Dallas, TX, USA  |        | 31 |      x    | Verio, Inc. VRIO-129-250                                      |

| 7  |      | 157.238.228.38 | ge-9-3.a00.dllstx04.us.ce.verio.net      | Dallas, TX, USA  |        | 31 |      x-    | Verio, Inc. VRIO-157-238                                      |

| 8  |      | 70.85.127.30  | dist-vlan31.dsr3-2.dllstx3.theplanet.com | ?Dallas, TX, USA |        | 36 |      -x--- | ThePlanet.com Internet Services, Inc. NETBLK-THEPLANET-BLK-13 |

| 9  |      | 70.85.127.76  | dist-vlan22.dsr1-2.dllstx2.theplanet.com | ?Dallas, TX, USA |        | 31 |      x    | ThePlanet.com Internet Services, Inc. NETBLK-THEPLANET-BLK-13 |

| 10  |      | 12.96.160.8    | dsr2-2-v1.dllstx4.theplanet.com          | ?Dallas, TX, USA |        | 31 |      x    | THEPLANET.COM INTERNET SERVICES THEPLANE725-160              |

| 11  |      | 67.18.116.70  | gig1-0-1.tp-car9-2.dllstx4.theplanet.com | ?Dallas, TX, USA |        | 36 |      -x--- | ThePlanet.com Internet Services, Inc. NETBLK-THEPLANET-BLK-11 |

| 12  |      | 67.19.36.6    | www.testmy.net                          | ?Dallas, TX, USA |        | 31 |      x-    | ThePlanet.com Internet Services, Inc. NETBLK-THEPLANET-BLK-11 |

-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Link to comment
Share on other sites

WEB100 Kernel Variables:

Client: localhost/127.0.0.1

AckPktsIn: 249

AckPktsOut: 0

BytesRetrans: 19320

CongAvoid: 142

CongestionOverCount: 0

CongestionSignals: 12

CountRTT: 168

CurCwnd: 5520

CurMSS: 1380

CurRTO: 360

CurRwinRcvd: 26280

CurRwinSent: 5840

CurSsthresh: 4140

DSACKDups: 0

DataBytesIn: 0

DataBytesOut: 589260

DataPktsIn: 0

DataPktsOut: 427

DupAcksIn: 67

ECNEnabled: 0

FastRetran: 11

MaxCwnd: 16560

MaxMSS: 1380

MaxRTO: 510

MaxRTT: 280

MaxRwinRcvd: 26280

MaxRwinSent: 5840

MaxSsthresh: 8280

MinMSS: 1380

MinRTO: 280

MinRTT: 80

MinRwinRcvd: 26280

MinRwinSent: 5840

NagleEnabled: 1

OtherReductions: 0

PktsIn: 249

PktsOut: 427

PktsRetrans: 14

X_Rcvbuf: 107520

RcvWinScale: 7

SACKEnabled: 3

SACKsRcvd: 78

SendStall: 0

SlowStart: 19

SampleRTT: 170

SmoothedRTT: 130

X_Sndbuf: 107520

SndWinScale: 0

SndLimTimeRwin: 0

SndLimTimeCwnd: 10321546

SndLimTimeSender: 2566

SndLimTransRwin: 0

SndLimTransCwnd: 1

SndLimTransSender: 1

SndLimBytesRwin: 0

SndLimBytesCwnd: 589260

SndLimBytesSender: 0

SubsequentTimeouts: 0

SumRTT: 19420

Timeouts: 1

TimestampsEnabled: 0

WinScaleRcvd: 0

WinScaleSent: 7

DupAcksOut: 0

StartTimeUsec: 681475

Duration: 10324505

c2sData: 2

c2sAck: 2

s2cData: 9

s2cAck: 2

half_duplex: 0

link: 100

congestion: 1

bad_cable: 0

mismatch: 0

spd: 0.00

bw: 0.54

loss: 0.028103044

avgrtt: 115.60

waitsec: 0.36

timesec: 10.00

order: 0.2691

rwintime: 0.0000

sendtime: 0.0002

cwndtime: 0.9998

rwin: 0.2005

swin: 0.8203

cwin: 0.1263

rttsec: 0.115595

Sndbuf: 107520

Checking for mismatch condition

(cwndtime > .3) [0.99>.3], (MaxSsthresh > 0) [8280>0],

(PktsRetrans/sec > 2) [1.4>2], (estimate > 2) [0.54>2]

Checking for mismatch on uplink

(speed > 50 [0>50], (xmitspeed < 5) [0.29<5]

(rwintime > .9) [0>.9], (loss < .01) [0.02<.01]

Checking for excessive errors condition

(loss/sec > .15) [0.00>.15], (cwndtime > .6) [0.99>.6],

(loss < .01) [0.02<.01], (MaxSsthresh > 0) [8280>0]

Checking for 10 Mbps link

(speed < 9.5) [0<9.5], (speed > 3.0) [0>3.0]

(xmitspeed < 9.5) [0.29<9.5] (loss < .01) [0.02<.01], (mylink > 0) [3.0>0]

Checking for Wireless link

(sendtime = 0) [2.0E=0], (speed < 5) [0<5]

(Estimate > 50 [0.54>50], (Rwintime > 90) [0>.90]

(RwinTrans/CwndTrans = 1) [0/1=1], (mylink > 0) [3.0>0]

Checking for DSL/Cable Modem link

(speed < 2) [0<2], (SndLimTransSender = 0) [1=0]

(SendTime = 0) [2.0E-4=0], (mylink > 0) [3.0>0]

Checking for half-duplex condition

(rwintime > .95) [0>.95], (RwinTrans/sec > 30) [0>30],

(SenderTrans/sec > 30) [0.1>30], OR (mylink <= 10) [3.0<=10]

Checking for congestion

(cwndtime > .02) [0.99>.02], (mismatch = 0) [0=0]

(MaxSsthresh > 0) [8280>0]

estimate = 0.54 based on packet size = 10Kbits, RTT = 115.6msec, and loss = 0.028103044

The theoretical network limit is 0.54 Mbps

The NDT server has a 105.0 KByte buffer which limits the throughput to 7.09 Mbps

Your PC/Workstation has a 25.0 KByte buffer which limits the throughput to 1.73 Mbps

The network based flow control limits the throughput to 1.09 Mbps

Client Data reports link is 'T1', Client Acks report link is 'T1'

Server Data reports link is '10 Gig', Server Acks report link is 'T1

all of this info is froma web site I will put the link into upon CA3LE GUY's approval it is so strange I get the same results every time now that I have followed your instructions Van the Man

Link to comment
Share on other sites

Delete all tweaks like i said in my previous post, then run a new TCP dump from Web 100

it says your MSS is 1380 and that give you a MTU of 1420

lets see if you get more then that after deleting all the tweaks.

When we know the MSS without tweaks i can make a custom cablenut file for you, just tell me what cap you have

VanBuren :)

Link to comment
Share on other sites

:D Good morning Van my my caps are1200kdwn and256kup I could'nt get any connectivety when I wiped the slate clean so I went back and put

mtu@1500

ttl@64

rwin65535

mtu discovery@yea

seletiveaks@yes

max dupak's@3

trun off rfc 1323

this is what I got

WEB100 Kernel Variables:

Client: localhost/127.0.0.1

AckPktsIn: 232

AckPktsOut: 0

BytesRetrans: 21900

CongAvoid: 129

CongestionOverCount: 0

CongestionSignals: 11

CountRTT: 151

CurCwnd: 4380

CurMSS: 1460

CurRTO: 360

CurRwinRcvd: 65535

CurRwinSent: 5840

CurSsthresh: 2920

DSACKDups: 0

DataBytesIn: 0

DataBytesOut: 575240

DataPktsIn: 0

DataPktsOut: 394

DupAcksIn: 68

ECNEnabled: 0

FastRetran: 11

MaxCwnd: 13140

MaxMSS: 1460

MaxRTO: 720

MaxRTT: 270

MaxRwinRcvd: 65535

MaxRwinSent: 5840

MaxSsthresh: 5840

MinMSS: 1460

MinRTO: 330

MinRTT: 80

MinRwinRcvd: 62615

MinRwinSent: 5840

NagleEnabled: 1

OtherReductions: 0

PktsIn: 232

PktsOut: 394

PktsRetrans: 15

X_Rcvbuf: 65535

RcvWinScale: 2147483647

SACKEnabled: 3

SACKsRcvd: 73

SendStall: 0

SlowStart: 18

SampleRTT: 170

SmoothedRTT: 130

X_Sndbuf: 65535

SndWinScale: 2147483647

SndLimTimeRwin: 0

SndLimTimeCwnd: 10266210

SndLimTimeSender: 11389

SndLimTransRwin: 0

SndLimTransCwnd: 1

SndLimTransSender: 1

SndLimBytesRwin: 0

SndLimBytesCwnd: 575240

SndLimBytesSender: 0

SubsequentTimeouts: 0

SumRTT: 18100

Timeouts: 0

TimestampsEnabled: 0

WinScaleRcvd: 2147483647

WinScaleSent: 2147483647

DupAcksOut: 0

StartTimeUsec: 260101

Duration: 10303712

c2sData: 2

c2sAck: 2

s2cData: 2

s2cAck: 2

half_duplex: 0

link: 100

congestion: 1

bad_cable: 0

mismatch: 0

spd: 0.00

bw: 0.56

loss: 0.027918782

avgrtt: 119.87

waitsec: 0.00

timesec: 10.00

order: 0.2931

rwintime: 0.0000

sendtime: 0.0011

cwndtime: 0.9989

rwin: 0.5000

swin: 0.5000

cwin: 0.1003

rttsec: 0.119868

Sndbuf: 65535

Checking for mismatch condition

(cwndtime > .3) [0.99>.3], (MaxSsthresh > 0) [5840>0],

(PktsRetrans/sec > 2) [1.5>2], (estimate > 2) [0.56>2]

Checking for mismatch on uplink

(speed > 50 [0>50], (xmitspeed < 5) [0.30<5]

(rwintime > .9) [0>.9], (loss < .01) [0.02<.01]

Checking for excessive errors condition

(loss/sec > .15) [0.00>.15], (cwndtime > .6) [0.99>.6],

(loss < .01) [0.02<.01], (MaxSsthresh > 0) [5840>0]

Checking for 10 Mbps link

(speed < 9.5) [0<9.5], (speed > 3.0) [0>3.0]

(xmitspeed < 9.5) [0.30<9.5] (loss < .01) [0.02<.01], (mylink > 0) [3.0>0]

Checking for Wireless link

(sendtime = 0) [0.00=0], (speed < 5) [0<5]

(Estimate > 50 [0.56>50], (Rwintime > 90) [0>.90]

(RwinTrans/CwndTrans = 1) [0/1=1], (mylink > 0) [3.0>0]

Checking for DSL/Cable Modem link

(speed < 2) [0<2], (SndLimTransSender = 0) [1=0]

(SendTime = 0) [0.0011=0], (mylink > 0) [3.0>0]

Checking for half-duplex condition

(rwintime > .95) [0>.95], (RwinTrans/sec > 30) [0>30],

(SenderTrans/sec > 30) [0.1>30], OR (mylink <= 10) [3.0<=10]

Checking for congestion

(cwndtime > .02) [0.99>.02], (mismatch = 0) [0=0]

(MaxSsthresh > 0) [5840>0]

estimate = 0.56 based on packet size = 11Kbits, RTT = 119.87msec, and loss = 0.027918782

The theoretical network limit is 0.56 Mbps

The NDT server has a 63.0 KByte buffer which limits the throughput to 4.17 Mbps

Your PC/Workstation has a 63.0 KByte buffer which limits the throughput to 4.17 Mbps

The network based flow control limits the throughput to 0.83 Mbps

Client Data reports link is 'T1', Client Acks report link is 'T1'

Server Data reports link is 'T1', Server Acks report link is 'T1'

Link to comment
Share on other sites

:D check these scores out it is 50% better

:::.. Download Stats ..:::

Connection is:: 3233 Kbps about 3.2 Mbps (tested with 1496 KB)

Download Speed is:: 395 KB/s

Tested From:: http://www.testmy.net/

Bottom Line:: 58 times faster than 56K you can download 1MB in 2.59 second(s)

Validation Link:: https://testmy.net/cgi-bin/get.cgi?Test_ID=EAICDS6CG :D

Link to comment
Share on other sites

×
×
  • Create New...