Help me figure out why I have gaming latency

Unfortunately I don't understand and therefore I can't configure the instances! It's written nowhere that you have to add ifb4 according to the configuration... I don't know what to say

@moeller0 @dlakelan
My ISP asks me: In order to narrow down the problem, please send us the MTR of your game server and three speed tests. What is your ping on the game?
My question is: What is the command line to answer their question? mtr ( add Ip ) ? Add timestamp? How do you do it?

"mtr" is an alternative to traceroute.
(Installable package for OpenWrt.)

opkg update 
opkg install mtr

mtr actually shows high latency for the connection step from Europe to USA (Stockholm to Washington) for that ea.com address that you showed in the sceenshots

root@router1:~# mtr easo.ea.com

                                My traceroute  [v0.95]
router1 (78.27.101.244) -> easo.ea.com (159.153.64.173)      2022-11-05T08:31:45+0200
Keys:  Help   Display mode   Restart statistics   Order of fields   quit
                                             Packets               Pings
 Host                                      Loss%   Snt   Last   Avg  Best  Wrst StDev
 1. 159-255-240-2.bb.dnainternet.fi         0.0%    47    1.4   1.3   1.1   1.5   0.1
 2. hel7-sr22.net.dnaip.fi                  0.0%    47    1.5   2.8   1.4  35.8   5.3
 3. 213.249.107.90                          0.0%    47    7.3  12.2   7.0  49.2   9.6
 4. ae64.edge1.Stockholm1.Level3.net        0.0%    47   14.6   9.2   7.0  21.0   3.6
 5. ae2.3602.ear3.Washington1.level3.net   89.1%    47  115.7 115.7 115.6 116.1   0.2
 6. 4.79.23.202                             0.0%    47  116.7 116.5 116.3 117.6   0.2
 7. 159.153.92.10                           0.0%    47  117.0 117.0 116.6 117.3   0.1
 8. 159.153.92.78                           0.0%    47  116.7 118.0 116.6 173.4   8.3
 9. easo.ea.com                             0.0%    47  116.5 116.5 116.4 116.7   0.1

Based on that example, my latency is small from me to my ISP (DNA) and even upto the next country Sweden. Then the transatlantic link causes latency, and finally the internal connections in USA are again quick.

Might be the same for you. Not much for a local ISP to do.

Ps.
If your your question was specifically about the mtr command line for screenshot / copypaste, the reporting mode might be easiest:
20 pings and report: mtr -r -c 20 easo.ea.com

root@router1:~# mtr -r -c 20 easo.ea.com

Start: 2022-11-05T09:18:46+0200
HOST: router1                     Loss%   Snt   Last   Avg  Best  Wrst StDev
  1.|-- 159-255-240-2.bb.dnainter  0.0%    20    1.2   1.3   1.1   1.6   0.1
  2.|-- hel7-sr22.net.dnaip.fi     0.0%    20    2.6   1.7   1.4   3.3   0.5
  3.|-- 213.249.107.90             0.0%    20    7.3   9.2   7.1  22.5   4.8
  4.|-- ae64.edge1.Stockholm1.Lev  0.0%    20    7.2   8.0   7.1  13.8   1.7
  5.|-- ae2.3602.ear3.Washington1 90.0%    20  115.8 115.7 115.6 115.8   0.2
  6.|-- 4.79.23.202                0.0%    20  116.5 116.5 116.3 116.9   0.1
  7.|-- 159.153.92.10              0.0%    20  117.2 117.0 116.8 117.3   0.1
  8.|-- 159.153.92.78              0.0%    20  116.7 116.7 116.6 117.0   0.1
  9.|-- bfbc2-ps3.gos.ea.com       0.0%    20  116.4 116.5 116.4 116.6   0.0
2 Likes

Too bad I don't play too much on PC. but while playing online I noticed this same and permanent latency, this famous delay that makes me play clearly in the past and never in the present ... Yes you are right, I'd better reread my thread from the beginning and collect the questions and information that everyone has asked, besides I'm embarrassed not to have done, there are many contributors who helped me here and I drowned I lost my head in the test, forgive me for this negligence involuntary :frowning:

My connections to migrate from vdsl to FTTS, I went from 70 mb to 450 mb and gain some ms!
Unfortunately the latency phenomenon has increased.

Logically, when there is a bottleneck somewhere on the network, the latency phenomenon increases. Because if we take as example a pipe with a relatively small diameter (bottleneck) it is easier to pass 70 people than 450.
On the other hand, if this diameter is really bigger, the 70 or 450 people will be able to circulate without jostling each other or creating a traffic jam.
Excuse me for the metaphor.

I may be wrong in this too simplistic explanation

A reliable test with MTR should last at least ten minutes.
The MTR output displays data for two areas of analysis: lost packets and latency. Since a single test with MTR is unidirectional, it would be advisable to trace in both directions.
We realize that the results are very often different in one direction or the other.
There is a good chance that the return path is very very different but how to prove it? Especially to my ISP!

Sorry for the confusion.
But is a possible way forward:

  1. install sqm-scripts and luci-app-sqm
  2. configure and test sqm
  3. add the interfaces from sqm into cake-autorate_config.sh

I like mtr -ezb4w -c 100 1.1.1.1 just replace the ip address, this will run for 100 seconds and plot a report after that time. For interactive use where you see updates in real-time use mtr -ezb 1.1.1.1
You can install and run mtr from your router's command line interface (via ssh).

1 Like

Thank you ...

add sqm interfaces in cake-autorate_config.sh

My concern is clearly here.

In dl ifb4wan and for ul ifb4br-lan
But nothing happens :frowning:
In luci a works I set sqm 10 /10 but the config.sh file has no effect

1 Like

This doesn't seem right. How have you setup SQM in LuCi? Screenshot might be helpful.

1 Like

I am at work unfortunately I have no image to transmit to you immediately.
I can tell you that I configured simply I activated sqm by checking the box, choose interface wan and inserted the bandwidth of 10000 in both directions, I have of course activated cake and piece of cake then save and apply

Is it advisable to make a game during the test?
If yes, is it necessary to locate the game's IP server and test on the same server that I play live?

Well, let's start with your static non-autorate sqm configuration.

Please post the output of:

  1. tc -s qdisc with sqm enabled
  2. a link to a completed waveform bufferbloat test: https://www.waveform.com/tools/bufferbloat

With that in hand let's figure out how/if static sqm works as expected. Since autorate builds on top of your initial cake configuration this is the fundament we need to be sure works.

1 Like

Let's not start with equal rates for both directions as that will not allo us to figure out whether we got the two directions properly assigned, so please repeat the waveform test also with sqm set to:
download: 10000
upload: 5000

root@CakeAutoRate:~# tc -s qdiscc
Object "qdiscc" is unknown, try "tc help".
root@CakeAutoRate:~# tc -s qdisc
qdisc noqueue 0: dev lo root refcnt 2
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
qdisc fq_codel 0: dev eth0 root refcnt 2 limit 10240p flows 1024 quantum 1518 ta                                                                                                                                                             rget 5ms interval 100ms memory_limit 4Mb ecn drop_batch 64
 Sent 72527557 bytes 90167 pkt (dropped 0, overlimits 0 requeues 1)
 backlog 0b 0p requeues 1
  maxpacket 0 drop_overlimit 0 new_flow_count 0 ecn_mark 0
  new_flows_len 0 old_flows_len 0
qdisc noqueue 0: dev lan1 root refcnt 2
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
qdisc noqueue 0: dev lan2 root refcnt 2
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
qdisc noqueue 0: dev lan3 root refcnt 2
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
qdisc noqueue 0: dev lan4 root refcnt 2
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
qdisc cake 8014: dev wan root refcnt 2 bandwidth 5Mbit besteffort triple-isolate                                                                                                                                                              nonat nowash no-ack-filter split-gso rtt 100ms noatm overhead 44
 Sent 17021857 bytes 36574 pkt (dropped 681, overlimits 22537 requeues 0)
 backlog 0b 0p requeues 0
 memory used: 48458b of 4Mb
 capacity estimate: 5Mbit
 min/max network layer size:           28 /    1476
 min/max overhead-adjusted size:       72 /    1520
 average network hdr offset:           14

                  Tin 0
  thresh          5Mbit
  target            5ms
  interval        100ms
  pk_delay       4.69ms
  av_delay       2.37ms
  sp_delay          2us
  backlog            0b
  pkts            37255
  bytes        18000010
  way_inds            1
  way_miss          225
  way_cols            0
  drops             681
  marks               0
  ack_drop            0
  sp_flows            2
  bk_flows            1
  un_flows            0
  max_len         10463
  quantum           300

qdisc ingress ffff: dev wan parent ffff:fff1 ----------------
 Sent 46499946 bytes 42659 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
qdisc noqueue 0: dev br-lan root refcnt 2
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
qdisc cake 8015: dev ifb4wan root refcnt 2 bandwidth 10Mbit besteffort triple-is                                                                                                                                                             olate nonat wash no-ack-filter split-gso rtt 100ms noatm overhead 44
 Sent 44826548 bytes 40984 pkt (dropped 1675, overlimits 57284 requeues 0)
 backlog 0b 0p requeues 0
 memory used: 138976b of 4Mb
 capacity estimate: 10Mbit
 min/max network layer size:           46 /    1476
 min/max overhead-adjusted size:       90 /    1520
 average network hdr offset:           14

                  Tin 0
  thresh         10Mbit
  target            5ms
  interval        100ms
  pk_delay       32.7ms
  av_delay       10.4ms
  sp_delay          2us
  backlog            0b
  pkts            42659
  bytes        47319332
  way_inds          143
  way_miss          282
  way_cols            0
  drops            1675
  marks               0
  ack_drop            0
  sp_flows            1
  bk_flows            1
  un_flows            0
  max_len         10430
  quantum           305

root@CakeAutoRate:~#

1 Like

So according to this information the interfaces to use in autorate are:

dl_if=ifb4wan # download interface
ul_if=wan      # upload interface

Please try to copy these as is into your cake-autorate_config.sh and restart cake-autorate.sh.

Your test result (9.2/4.19 indicates that the dl_if and ul_if lines above should be correct).

EDIT: I screwed up, you were doing fine, the cake configuration matches the speedtest result...

1 Like

ok i have make the change, so how can i check this now ?

Are you running cake-autorate as a service? If so restart it:

service cake-autorate restart

And try:

tail -f /var/log/cake-autorate.log

DATA; 2022-11-05-16:32:12; 1667662332.733965; 1667662332.733663; 18; 17; 0; 0; 1667662332.728640; 1.1.1.1; 259; 4844; 5100; 256; 13117; 4844; 5100; 256; 13117; 0; 0; dl_idle; ul_idle; 20000; 20000
DATA; 2022-11-05-16:32:12; 1667662332.783878; 1667662332.783487; 18; 17; 0; 0; 1667662332.778860; 1.0.0.1; 259; 4847; 5200; 352; 13117; 4847; 5200; 352; 13117; 0; 0; dl_idle; ul_idle; 20000; 20000
DATA; 2022-11-05-16:32:12; 1667662332.836727; 1667662332.836443; 18; 17; 0; 0; 1667662332.831510; 8.8.8.8; 259; 6406; 6450; 44; 13117; 6406; 6450; 44; 13117; 0; 0; dl_idle; ul_idle; 20000; 20000
DATA; 2022-11-05-16:32:12; 1667662332.885972; 1667662332.885658; 18; 17; 0; 0; 1667662332.880920; 8.8.4.4; 259; 6000; 6100; 100; 13117; 6000; 6100; 100; 13117; 0; 0; dl_idle; ul_idle; 20000; 20000
LOAD; 2022-11-05-16:32:12; 1667662332.894523; 1667662332.894339; 83; 17

so i need to change, sqm set to:
download: ?
upload: ?

ok i have put DL 50000
UL 50000
autorate actived AUTO

test waveform result: https://www.waveform.com/tools/bufferbloat?test-id=63027199-518b-4f33-a456-d75f2098766b

LOAD; 2022-11-05-16:42:09; 1667662929.610880; 1667662929.610697; 0; 0
LOAD; 2022-11-05-16:42:09; 1667662929.811444; 1667662929.811261; 28; 0
LOAD; 2022-11-05-16:42:10; 1667662930.011915; 1667662930.011734; 0; 0
LOAD; 2022-11-05-16:42:10; 1667662930.212495; 1667662930.212314; 0; 0
LOAD; 2022-11-05-16:42:10; 1667662930.413091; 1667662930.412904; 144; 42
LOAD; 2022-11-05-16:42:10; 1667662930.614192; 1667662930.613965; 265; 522
LOAD; 2022-11-05-16:42:10; 1667662930.814786; 1667662930.814591; 0; 2