Mdthbs

Questions about authorship rank and academic politics

Prime parity peregrination

What spells could a Bard identify from a Pact of the Tome Warlock's Book of Shadows?

Was it really unprofessional of me to leave without asking for a raise first?

Miss Toad and her frogs

What is the highest number of sneak attacks that a Pure/High Level Rogue (Level 17+) can make in one round?

Why do I need two parameters in an HTTP parameter pollution attack?

Why do the keys in the circle of fifths have the pattern of accidentals that they do?

Convergenge or divergence of series with e

Who are these Discworld wizards from this picture?

How hard is it to sell a home which is currently mortgaged?

Word for the ability to deliver clever replies?

How did researchers use to find articles before the Internet and the computer era?

Did Wakanda officially get the stuff out of Bucky's head?

Can I ask to speak to my future colleagues before accepting an offer?

Can someone break into a Leomund's Tiny Hut via the Ethereal Plane?

Does Anosov geodesic flow imply asphericity?

Inquiring about the possibility of a job

Can two or more lightbeams (from a laser for example) have visible interference when they cross in mid-air?

Why are there so many religions and gods?

Why did this meteor appear cyan?

Do space suits measure "methane" levels or other biological gases?

Can a single server be associated with multiple domains?

Reverse of diffraction

fast retransmit does not work

Why does this simple Perl client not work?Why beep does not work in Ubuntu?Udev hwdb rule does not workCUPS Error Printer is not RespondingX11 forwarding to Mac does not workTrackpad and keyboard no longer work on any linux distribution but work on GRUB / Windows and BIOS menuUbuntu 14.04 does not save screen resolutionAck/Ag does not return search result without *Cannot receive DHCP OFFERpkill does not work on Ubuntu 16.04

.everyoneloves__top-leaderboard:empty,.everyoneloves__mid-leaderboard:empty,.everyoneloves__bot-mid-leaderboard:empty{ margin-bottom:0;
}

1

Fast retransmit says that if 3 dupacks have been received by sender, then the sender should retransmit the lost packet. But my tcpdump output shows that it didn't work like this. It didn't retransmit the packet even more than 10 dupacks have been received.
Why didn't it triggerfast retransmit? Does anyone has idea about this? Thanks

cat /proc/version:
Linux version 4.2.0-42-generic (buildd@lgw01-55) (gcc version 4.8.4 (Ubuntu 4.8.4-2ubuntu1~14.04.3) ) #49~14.04.1-Ubuntu SMP Wed Jun 29 20:22:11 UTC 2016

sysctl:
net.ipv4.tcp_congestion_control = cubic

ubuntu tcp ack

share|improve this question

edited Aug 29 '16 at 0:54

chengtcli

asked Aug 28 '16 at 5:08

chengtclichengtcli

633 bronze badges

bumped to the homepage by Community♦ 1 hour ago

This question has answers that may be good or bad; the system has marked it active so that they can be reviewed.

• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  In our systems, net.ipv4.tcp_timestamps=0 and net.ipv4.tcp_sack=0
  
  – chengtcli
  Aug 30 '16 at 1:38
  

  
  
  
  

add a comment | 

1

Fast retransmit says that if 3 dupacks have been received by sender, then the sender should retransmit the lost packet. But my tcpdump output shows that it didn't work like this. It didn't retransmit the packet even more than 10 dupacks have been received.
Why didn't it triggerfast retransmit? Does anyone has idea about this? Thanks

cat /proc/version:
Linux version 4.2.0-42-generic (buildd@lgw01-55) (gcc version 4.8.4 (Ubuntu 4.8.4-2ubuntu1~14.04.3) ) #49~14.04.1-Ubuntu SMP Wed Jun 29 20:22:11 UTC 2016

sysctl:
net.ipv4.tcp_congestion_control = cubic

ubuntu tcp ack

share|improve this question

edited Aug 29 '16 at 0:54

chengtcli

asked Aug 28 '16 at 5:08

chengtclichengtcli

633 bronze badges

bumped to the homepage by Community♦ 1 hour ago

This question has answers that may be good or bad; the system has marked it active so that they can be reviewed.

• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  In our systems, net.ipv4.tcp_timestamps=0 and net.ipv4.tcp_sack=0
  
  – chengtcli
  Aug 30 '16 at 1:38
  

  
  
  
  

add a comment | 

Fast retransmit says that if 3 dupacks have been received by sender, then the sender should retransmit the lost packet. But my tcpdump output shows that it didn't work like this. It didn't retransmit the packet even more than 10 dupacks have been received.
Why didn't it triggerfast retransmit? Does anyone has idea about this? Thanks

cat /proc/version:
Linux version 4.2.0-42-generic (buildd@lgw01-55) (gcc version 4.8.4 (Ubuntu 4.8.4-2ubuntu1~14.04.3) ) #49~14.04.1-Ubuntu SMP Wed Jun 29 20:22:11 UTC 2016

sysctl:
net.ipv4.tcp_congestion_control = cubic

ubuntu tcp ack

share|improve this question

edited Aug 29 '16 at 0:54

chengtcli

asked Aug 28 '16 at 5:08

chengtclichengtcli

633 bronze badges

share|improve this question

edited Aug 29 '16 at 0:54

chengtcli

asked Aug 28 '16 at 5:08

chengtclichengtcli

633 bronze badges

share|improve this question

edited Aug 29 '16 at 0:54

chengtcli

asked Aug 28 '16 at 5:08

chengtclichengtcli

633 bronze badges

asked Aug 28 '16 at 5:08

chengtclichengtcli

633 bronze badges

asked Aug 28 '16 at 5:08

chengtclichengtcli

633 bronze badges

1 Answer
1

active

oldest

votes

0

Please check "tcp_thin_dupack"

sysctl net.ipv4.tcp_thin_dupack

net.ipv4.tcp_thin_dupack = 0

tcp_thin_dupack - BOOLEAN

    Enable dynamic triggering of retransmissions after one dupACK

    for thin streams. If set, a check is performed upon reception

    of a dupACK to determine if the stream is thin (less than 4

    packets in flight). As long as the stream is found to be thin,

    data is retransmitted on the first received dupACK. This

    improves retransmission latency for non-aggressive thin

    streams, often found to be time-dependent.

    For more information on thin streams, see

    Documentation/networking/tcp-thin.txt

share|improve this answer

answered Aug 28 '16 at 7:31

0x0C40x0C4

33511 silver badge66 bronze badges

• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  it seems that the tcp_thin_dupack is able to trigger fast retransmit with one DUP ack, so it can improve retransmission latency. But this seems can't explain why fast retransmit was not triggered with multi DUP acks.
  
  – chengtcli
  Aug 28 '16 at 8:29
  
  
  

  
  
  
  

add a comment | 

Your Answer

StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "106"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);

StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});

function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: false,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: null,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});


}
});

draft saved

draft discarded

Sign up or log in

StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});

Sign up using Google

Sign up using Facebook

Sign up using Email and Password

Post as a guest

Name

Email

Required, but never shown

StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2funix.stackexchange.com%2fquestions%2f306167%2ffast-retransmit-does-not-work%23new-answer', 'question_page');
}
);

Post as a guest

Name

Email

Required, but never shown

0

Please check "tcp_thin_dupack"

sysctl net.ipv4.tcp_thin_dupack

net.ipv4.tcp_thin_dupack = 0

tcp_thin_dupack - BOOLEAN

    Enable dynamic triggering of retransmissions after one dupACK

    for thin streams. If set, a check is performed upon reception

    of a dupACK to determine if the stream is thin (less than 4

    packets in flight). As long as the stream is found to be thin,

    data is retransmitted on the first received dupACK. This

    improves retransmission latency for non-aggressive thin

    streams, often found to be time-dependent.

    For more information on thin streams, see

    Documentation/networking/tcp-thin.txt

share|improve this answer

answered Aug 28 '16 at 7:31

0x0C40x0C4

33511 silver badge66 bronze badges

• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  it seems that the tcp_thin_dupack is able to trigger fast retransmit with one DUP ack, so it can improve retransmission latency. But this seems can't explain why fast retransmit was not triggered with multi DUP acks.
  
  – chengtcli
  Aug 28 '16 at 8:29
  
  
  

  
  
  
  

add a comment | 

0

Please check "tcp_thin_dupack"

sysctl net.ipv4.tcp_thin_dupack

net.ipv4.tcp_thin_dupack = 0

tcp_thin_dupack - BOOLEAN

    Enable dynamic triggering of retransmissions after one dupACK

    for thin streams. If set, a check is performed upon reception

    of a dupACK to determine if the stream is thin (less than 4

    packets in flight). As long as the stream is found to be thin,

    data is retransmitted on the first received dupACK. This

    improves retransmission latency for non-aggressive thin

    streams, often found to be time-dependent.

    For more information on thin streams, see

    Documentation/networking/tcp-thin.txt

share|improve this answer

answered Aug 28 '16 at 7:31

0x0C40x0C4

33511 silver badge66 bronze badges

• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  it seems that the tcp_thin_dupack is able to trigger fast retransmit with one DUP ack, so it can improve retransmission latency. But this seems can't explain why fast retransmit was not triggered with multi DUP acks.
  
  – chengtcli
  Aug 28 '16 at 8:29
  
  
  

  
  
  
  

add a comment | 

Please check "tcp_thin_dupack"

sysctl net.ipv4.tcp_thin_dupack

net.ipv4.tcp_thin_dupack = 0

tcp_thin_dupack - BOOLEAN

    Enable dynamic triggering of retransmissions after one dupACK

    for thin streams. If set, a check is performed upon reception

    of a dupACK to determine if the stream is thin (less than 4

    packets in flight). As long as the stream is found to be thin,

    data is retransmitted on the first received dupACK. This

    improves retransmission latency for non-aggressive thin

    streams, often found to be time-dependent.

    For more information on thin streams, see

    Documentation/networking/tcp-thin.txt

share|improve this answer

answered Aug 28 '16 at 7:31

0x0C40x0C4

33511 silver badge66 bronze badges

sysctl net.ipv4.tcp_thin_dupack

net.ipv4.tcp_thin_dupack = 0

tcp_thin_dupack - BOOLEAN

    Enable dynamic triggering of retransmissions after one dupACK

    for thin streams. If set, a check is performed upon reception

    of a dupACK to determine if the stream is thin (less than 4

    packets in flight). As long as the stream is found to be thin,

    data is retransmitted on the first received dupACK. This

    improves retransmission latency for non-aggressive thin

    streams, often found to be time-dependent.

    For more information on thin streams, see

    Documentation/networking/tcp-thin.txt

share|improve this answer

answered Aug 28 '16 at 7:31

0x0C40x0C4

33511 silver badge66 bronze badges

answered Aug 28 '16 at 7:31

0x0C40x0C4

33511 silver badge66 bronze badges

answered Aug 28 '16 at 7:31

0x0C40x0C4

33511 silver badge66 bronze badges