I had trouble getting the tool running. I tried a couple diff linux OS (lubuntu and ubuntu) but for whatever reason the tool was not working, getting the same errors with the oodle thing.

For those unfamiliar with linux and who have trouble getting the tool working on the archiso, here is how I got the tool working again.

• Set up network switch as specific in Soreepeong's instructions
• Follow this tutorial to set up a complete debian vm with enough virtual HD space: https://techlabs.blog/categories/debian-linux/create-a-debian-linux-virtual-machine-using-hyper-v
• You can tweak some of the other settings (like running on startup, RAM allocation)
• Once debian VM is running, follow all the on-screens prompts to fully install the system (there are a lot of steps but it guides you through it. make sure you remember your password cuz you'll need it every time)
• Once debian is fully installed, open the terminal app and then follow Soreepeong's instructions to download and start the tool. However, put sudo before every command (i.e. sudo iptables ...)
• Make sure you install the GCC libraries for x32 because the tool seems to need this:
  sudo apt install g++ gcc-multilib g++-multilib
• In addition to sudo before each command, You need to be in root admin in Terminal, otherwise you will get error messages.
• To get in root admin, can use the commands:

su
sudo su -

• This puts you in root. You will know because the username in Terminal will change to this:
  

Hopefully this helps the technically inept ppl like myself

So I get disconnected while in queue right now because it gives me the following error as the game loads me in:

I have never encountered this erorr before even while using outdated opcodes, so maybe it is related to the latest commit? Not sure. Could still be related to opcodes though.

Is the route passing 124.150.157.0/24 through the vm needed? It only seems to be used for the lodestone and not any of the servers and I can't access the lodestone when using it.

Hello! Just wanted to let you know that the hotfix today makes this not work at the present time, but I am sure slight tweaks will have it up and running again. Thank you!

I've configured the software to run on my linux machine and have setup a static IP on my ps4, and it's worked for months. But as of the latest version, it hasn't worked anymore. I get the same amount of latency on action presses as I normally do. However, it does seem the program is capturing my inputs since it is logging stuff, so in theory it's still correctly configured.

I've tried several times making it download the opcodes again since they were updated, nothing changed, really.
Here's a screenshots of some of the logs as I try hitting dummies in-game:

Also, one bizarre behavior change I've noticed is that: normally after setting up the static IP on my PS4 and setting the gateway to point to my laptop, it will not have a working internet connection until I run the script. However, after today, it does seem to work normally with the static IP pointing to my machine regardless of me running the script. Now I'm not a networking person and I do not know if this has anything to do with anything, but it is something that changed so I'm stating it here.

I apologize if I failed to cite any steps or issues I've had, but if it's something I'm doing wrong, please tell me so I can fix it.
Could it be because I'm not using a VM with no memory and just my own linux machine? I've tried running the iptables commands following the step by step guide again but it changes nothing, either.

Hello, I'm relatively new to all this. Can someone please help me? I'm trying to run XivMitmLatencyMitigator on my PS5 through a MacBook pro; however, the one and only time I see Mac mentioned is halfway through step 8:

Mac: Run a Terminal, and the run the following.
sudo route -n add -net 204.2.229.0/24 <vm_ip>
sudo route -n add -net 124.150.157.0/24 <vm_ip>
sudo route -n add -net 183.111.189.0/24 <vm_ip>
sudo route -n add -net 195.82.50.0/24 <vm_ip>

Am I missing something here? Is this all that needs to be done before finalizing step 9? If not, is that explained anywhere else? Every other step has to do with Windows or Linux.

Please Delete this issue, (Missinformed)
(Sorry for the inconvenience

I tried several distros, ubuntu, arch, debian. same problems.
Could someone make a tutorial that works?

https://stackoverflow.com/a/48370480

Will there be an update to the setups using virtual machine for the PS4? Been having a lot of connection and disconnect problems to the servers in this latest update. Used the https://github.com/Soreepeong/XivMitmLatencyMitigator perfectly until this latest update that ended today (08/03/2021) and I would like to know if it only takes time for everything to settle or if a different configuration will be required. Same for future updates.

I've done step by step and even managed to connect to the game a few times, but it lasts at most for about 10 minutes until the game's connection is lost and then I'm required to do the whole procress (after setting the VM up) to log in again.

First time posting, so I apologize if this isn't the right place to ask this sort of question.

Here is my VM for reference.

Edit: uploaded the image

I was wondering if this would be updated for Endwalker. Thanks for the good work on this and Alexander as always!

Hello, it's been a while since I played and when trying to use the tool now I find an error message. I looked at recent issues and couldn't find one similar, I play on console so I don't really understand how I'd get ahold of "ffxiv.exe" for the VM to work. Any help is appreciated, here's the error:

I'm using Hyper-V on Windows 10 Pro with Archlinux if that helps.

I try running this, but get this error when I try to log on:

Looks like it is defaulting to CN servers? I am really not sure...

Pretty sure just by putting that in the issue title I've become that which I hate, but ...

Have you tried containerising this in any way, so that people on Linux/Mac/whatever can just run a container on the PC they're playing on?

I messed around with it a little bit but the codebase is pretty rigid about its forwarding rules at present.

Cheers,

Arc

Hey there, I'm intending on running FFXIV directly on Linux. Is it possible to run this on the same system as the game? If so, would I just follow steps 6 & 7(Or only 7) & all set? Or should I still route through a VM?

I have been running this on my RPi3 for about a week with no issues (MCH IS HAPPY!) but this afternoon when I went to login following the 6.15 patch, it wouldn't load. I set my PS4 back to normal networking and it works fine. Here is the output from the script via journalctl

pi@pihole:~ $ sudo systemctl status xivalex
* xivalex.service - XIV Alexander Daemon
     Loaded: loaded (/etc/systemd/system/xivalex.service; enabled; vendor preset: enabled)
     Active: active (running) since Tue 2022-06-07 16:24:54 EDT; 1min 57s ago
   Main PID: 4606 (XIVAlexanderMIT)
      Tasks: 4 (limit: 2060)
     Memory: 11.9M
        CPU: 871ms
     CGroup: /system.slice/xivalex.service
             |-4606 /bin/bash - /usr/local/bin/XIVAlexanderMITM.sh
             |-4608 /usr/bin/python3 /usr/local/bin/XIVAlexanderMITM.py
             |-4783 /usr/bin/python3 /usr/local/bin/XIVAlexanderMITM.py
             `-4784 /usr/bin/python3 /usr/local/bin/XIVAlexanderMITM.py

Jun 07 16:26:42 pihole XIVAlexanderMITM.sh[4784]:     bundle_data = bytearray(zlib.decompress(bundle_data))
Jun 07 16:26:42 pihole XIVAlexanderMITM.sh[4784]: zlib.error: Error -3 while decompressing data: incorrect header check
Jun 07 16:26:42 pihole XIVAlexanderMITM.sh[4784]: 2022-06-07 16:26:42,913        4784        INFO        D->U discarded 00
Jun 07 16:26:43 pihole XIVAlexanderMITM.sh[4608]: 2022-06-07 16:26:43,022        4608(main)        INFO        [4732  ] has exit with status code 0.
Jun 07 16:26:43 pihole XIVAlexanderMITM.sh[4783]: 2022-06-07 16:26:43,288        4783        ERROR        Unknown error occurred while trying to parse bundle
Jun 07 16:26:43 pihole XIVAlexanderMITM.sh[4783]: Traceback (most recent call last):
Jun 07 16:26:43 pihole XIVAlexanderMITM.sh[4783]:   File "/usr/local/bin/XIVAlexanderMITM.py", line 452, in find
Jun 07 16:26:43 pihole XIVAlexanderMITM.sh[4783]:     bundle_data = bytearray(zlib.decompress(bundle_data))
Jun 07 16:26:43 pihole XIVAlexanderMITM.sh[4783]: zlib.error: Error -3 while decompressing data: incorrect header check
Jun 07 16:26:43 pihole XIVAlexanderMITM.sh[4783]: 2022-06-07 16:26:43,293        4783        INFO        D->U discarded 00

I am playing from the EU on the NA datacentre and I discovered a problem where while using this my client will completely stop reacting to all skill inputs after a brief time. This follows this log message:

I was seemingly able to get around the problem by raising EXTRA_DELAY (to 0.125) but it is still quite worrying as it could suddenly turn me into a waste of space during an instance and I'm not sure how reliable my workaround is. The only way to make the client accept skill input again is to log out and back in. Do you have any idea what's happening?

I am trying to make the python script run on a raspberry pi zero W I have connected to my local network. It is running raspberry pi os.
When I try to add a new rule to iptables this is the console output:

$ iptables -t nat -A POSTROUTING -s my-local-ip/24 -o wlan0 -j MASQUERADE
iptables v1.8.2 (nf_tables): Chain 'MASQUERADE' does not exist
Try `iptables -h' or 'iptables --help' for more information.

Any idea on what to do? Am I maybe missing some additional dependencies?

Hello! I'm prefacing this by saying I really don't know where I want to get with this because I don't expect you to troubleshoot this for me, it's a lot of things and you probably have better things to do than deal with my dumb ass, as this could prove time-consuming and frustrating. But I want to provide my experience with setting this up.

I've been desperately trying to make this work all day with a laptop running Windows 10 Home and my PS4 but I keep running into different errors. I'm completely unfamiliar with Linux and virtual machines so from the get go this is a whole new realm for me. Even with the instructions provided I'm struggling a lot, I'm really not a techy type person.

I've tried Hyper-V on Windows 10 Home with a workaround I found on another website since it typically won't work on Home and requires upgrading to Pro, which I can't afford. Said workaround was successful and enabled Hyper-V but when I finally get to step 4-xi to power up the virtual machine I get this: https://puu.sh/HQV9a/21e05bdde0.png which I tried to look up on Google and I just can't make heads or tails of it. Too many terms and things thrown around I don't understand at all.

I tried VMware and VirtualBox too, and I run into several other errors along the way. It's too many things to list and I don't remember most of them at this point, so I'll just speak of my latest attempt. With VirtualBox and an Arch Linux ISO I got to this point: https://puu.sh/HQUJN/a9fabe20f8.png

No errors or anything that I understand to be an error, but I do realize it's still perfectly possible the virtual machine or its connection to the network aren't set up correctly. So after getting there, when I go to my PS4 to set up a new internet connection, I put in what I think (I'm really not sure) are the correct IP Address, Subnet Mask and Primary & Secondary DNS, and for Default Gateway I introduce 10.0.2.15 and then the PS4 just won't get connected to the Internet at all. So I'm really stuck because no matter what I do or where I get to, something makes me hit a brick wall and I feel too dumb at this point to make this work and I'm just defeated, I give up.

But I do want to thank you for making this and XivAlexander (a lot of my friends use it) and providing it for free, and thank you for your time reading my frustrations. And I also want to say my apologies if this was a waste of time to read.

Hello,

When trying to run the script I get the following errors in the screenshot below:

I believe it is caused by the addition of this new file in a recent commit to XivAlexander:
https://github.com/Soreepeong/XivAlexander/blob/6e5d42402f972478fc6893ea6280eedc3d355d49/StaticData/OpcodeDefinition/notes.txt

To test I forked XivAlexander and XivMitmLatencyMitigator, deleted notes.txt and changed "OPCODE_DEFINITION_LIST_URL" in "mitigate.py" to point to the fork and it was now able to update opcodes like normal.

(I'm extremely new to GitHub, so I'm sorry if I did anything wrong.)

P.S. Thank you for creating this tool. I can finally enjoy playing the game after years of being unable to double weave.

Probably aware of this issue, but just in case, when I start this, I receive the following output that may be related to a change in the XIVAlex source:

Taken from the XIVA issue post:

"ActorCast": "0x00a9",
"ActorControl": "0x01c8",
"ActorControlSelf": "0x035a",
"AddStatusEffect": "0x01e9",
"RequestUseAction": "0x02a2",
"RequestUseAction2": "0x02f6",
"SkillResultResponse01": "0x0283",
"SkillResultResponse08": "0x025b",
"SkillResultResponse16": "0x015d",
"SkillResultResponse24": "0x0091",
"SkillResultResponse32": "0x0169",

Hello!

First, allow me to thank you for this amazing tool, allowing PS4 players to experience the same benefits as those who use XIVAlex. Right now, the two seem to be nearly identical in performance, but there is something I have noticed in my own experiences with the LatencyMitigator; whenever I am moving and/or in instances with multiple people (8-man raids included), I will occasionally clip and experience inconsistent performance. This is a result of the c2m latency being much lower in these cases (I have seen it be 10-30ms in a crowded instance, but 80+ ms in a solo instance), which widens the gap between latency and delay.

Supposedly, XIVAlex has a latnency correction feature, and if such a feature could be added to the LatencyMitigator to account for this, it would greatly improve the consistency of this tool. Thank you!

Hi. I already installed xivmitm on my ps5 and works awesome, but the last week I had to move all my room through other one where I have only 1 ethernet cable so I connected to the console. Now the problem is that I don´t have other one for the pc and I can´t find a way to config the vm to work wireless. That´s why I´m making this post, I´ll appreciate if someone can help me :3

Not sure if this is already being researched, but someone I know found the 5.5 opcodes to update the plugin, I do not know how useful they will be for this, but I will post them here anyway.

"ActorCast": "0x03c0",
"ActorControl": "0x03b5",
"ActorControlSelf": "0x03c0",
"AddStatusEffect": "0x01d7",
"AlwaysOnTop": false,
"GameServerIpRange": "124.150.157.0/24,195.82.50.0/24,204.2.229.0/24,127.0.0.0/8,10.0.0.0/8,172.16.0.0/12,192.168.0.0/16,",
"GameServerPortRange": "10000,10001-65535",
"LogAllIpcMessage": false,
"MedianRttCalculationCount": 1,
"RequestUseAction": "0x03b0",
"RequestUseAction2": "0x01a0",
"ShowControlWindow": false,
"ShowLoggingWindow": true,
"SkillResultResponse01": "0x0204",
"SkillResultResponse08": "0x01c7",
"SkillResultResponse16": "0x032a",
"SkillResultResponse24": "0x035c",
"SkillResultResponse32": "0x0d20",
"UseAllIpcMessageLogger": false,
"UseAutoAdjustingExtraDelay": false,
"UseEffectApplicationDelayLogger": false,
"UseHighLatencyMitigation": true,
"UseHighLatencyMitigationLogging": true,
"UseOpcodeFinder": true

This utility is not working with the new patch, but I assume this is for reasons similar to ACT being down after a new patch releases, so I am not too surprised, just an FYI if you were unaware of this. Thank you!

Im getting this error

Hey there, found this tool within the past week and it has been a godsend! I've been running it on a laptop I don't really use, connected to ethernet since the VM wasn't working on wifi. My PS4 is then connected to the internet via the VM.
The tool worked perfectly until today, and I was wondering if that would be due to the 5.58 patch. When I type in the curl command it appears to download, but once I'm in the game it doesn't print any more logs and my latency is as usual. Additionally, every time I press Control-C to exit and try to restart it immediately disconnects me from the game.
Just wanted to make sure that the tool will be working again soon! Can't believe I've lived this whole time without it, thank you very much for the work

Hello! Thanks for the help the other day and I'm sorry to bother again.

I seem to run into some sort of stability or connection issue where suddenly the game will stop processing my actions and everything either completely freezes or, more often, I will see people/enemies moving around a bit and eventually completely stop and myself autoattacking. When this happens and I execute actions I'll see the animation but no damage, the next combo won't light up, etc. I end up having to force close my client and having to set up the connection again or run without the mitigator. I don't run into connection issues without it.

Here's a log of what happens, where you can notice a sharp increase in "delay" at the time I run into the issue.
https://puu.sh/HS8Vv/996279c115.png

Sometimes this happens as soon as 5-10 minutes into playing, sometimes it takes a bit longer, and the other day I ran several roulettes in a row with no issues. I've been mostly playing MCH and DRG (log above is MCH).

I know this is most likely an issue on my end as I'm still not fully sure if my VM and PS4 connection are set up 100% correctly. I can provide any other data you might need.

Edit: I will also add that when I set up the connection for the mitigator, the download speed shown on PS4 when testing the connection tanks pretty low compared to my regular connection, is that normal?

I want to preface that I'm generally unfamiliar with coding and networking.

That aside, I understand that there's a ~500ms delay after every OGCD. This delay makes stuff like triple weaving in most scenarios unfavorable, regardless of ping.

When running the latency mitigator, I experience almost zero delay on OGCD and can triple weave on most jobs without clipping. I was wondering if this was intentional/client-side only or if this is an actual issue. I've attached a .mp4 and .png of me triple weaving on 2.5s GCD and the code from my VM, respectively.

.
Edit: I've set DEFAULT_EXTRA_DELAY to as high as 0.500. There was no distinguishable difference in weaving when compared to a value of 0.075.

Edit: Messed around with value...

Ended up working out for me. Let me know if this is O.K.

I've been speedrunning e9s with this on and off and for some reason it keeps spitting out this:
[5] 2021-02-14 19:54:13.667681 ┎ Not user-originated, and isn't an auto-attack (00f5) [5] 2021-02-14 19:54:13.668266 S2C_ActionEffect: actionId=00f5 sourceSequence=0000 wait=100ms
the id has been different between sessions I remember it being 00ba once for certain (yesterday)

I feel that perhaps the auto attack id is shuffled in this instance perhaps?

I'm playing Dancer and it happens in this instance, can't confirm for any other savages since my group has been focusing on this one.

I assume XivAlexander would have the same issue, attached is the full log from today.
xmlm.20210214155118.5.log

The reason I asks, is could I just boot up a docker VM on my machine directly and use that instead?
Or why not just run the python script directly on my machine?
I'm sure I've missed something, so asking for clarification.

Edit: So the script intercepts the traffic and responds to the client the necessary ack to allow abilities to be used, circumventing waiting for the servers ack, did I get it right?

I can't figure out how to get a VM running that isn't slow and laggy. But I do have a Linux machine. Would it be a bad idea to just run this program straight on the computer and not sandbox it?

When using with a VPN like Mudfish, what is the expected route order? Mudfish creates a TUN interface but I'm unclear on if iptables is supposed to be natting the tunnel or the underlying interface to the default gateway, or if additional rules are required.

There's no encryption as far as I'm aware, it's just a standard layer 3 tunnel. It'd be good to get some clarification on expected routing between a console and the servers when both the mitigator and a "VPN" are running, especially with the intended source and device for masquerading.

To clarify a bit, as far as I can tell the NAT rules for prerouting are fine as is, but postrouting would require the full local subnet as the source and then an entry for both the eth0 (or whatever) interface as well as the vpn interface, so something like this:

iptables -t nat -I PREROUTING -d 124.150.157.0/24 -p tcp -j REDIRECT --to 9999"
iptables -t nat -A POSTROUTING -s 10.0.0.192/26 -o eth0 -j MASQUERADE
iptables -t nat -A POSTROUTING -s 10.0.0.192/26 -o tun0 -j MASQUERADE

I used a random port here as an example, and I'll also note that SE DCs don't use /24 networks, but yolo.

Hasn't done this before (even on big patch days) - at least for me.

Hi, i just downloaded ubuntu but when i run the curl command i get this

Also since i have installed ubuntu HFS.exe and ffxiv.exe dosent work ( as far as i know exe files dosent work on linux machine) anyone has an idea ?, Thanks in advance

I am a new user of XivMitmLatencyMitigator, tyring to use it for version 6.08 of FFXIV.
When trying to use this, I encountered an error definitions.json not found. Aterward, the program is stuck in downloading opcode definition files.
I also tried to remove the definitions.json that seems to be proposed in other thread, but it does not help either.

Please help to solve the issue.

Read the files in https://github.com/Soreepeong/XivAlexander/tree/main/StaticData/OpcodeDefinition and automatically apply corresponding opcodes for connections.

Any idea how to solve this? I installed xivalexander months ago, but with the new steps we have to do to make it work again, I'm a bit lost. I tried to pick another ip from HFS, but it keeps saying the "connection timed out", or the other one if i choose the same ip on HFS than the local one, but I don't know if that's good. I'd really appreciate if someone help me, xivalexander is so damn good.

I've installed the most recent version of arch, since I was getting the "not enouh free space" error using the live cd version.
I keep getting this error after running the curl command for the script

Has anyone here faced the same issue before? How do I get around this?

Hello, I'd just like to point out that the tool is no longer working after the patch. I understand it's supposed to until the new opcodes are found but since xivalexander was already working I thought this would be as well (Apologies if that's not how it works and we just need to wait a little more.)

Here's what's getting printed on my VM just in case, the action request messages don't even appear upon using OGCDs.

Is it required to have a Hardwired connection between the Machine running the VM and the PS4? Or is it possible to do it wirelessly?
(Tested between two PC's as a test run on the same Network, Devices seemed to not connect)

Do not close socket as soon as either direction of stream gets shut down.

until yesterday it was working perfectly. But today I was unable to connect. I redid all the steps, but it still doesn't work.

I'm currently running this python script as a systemd service on my local raspberry pi. It appears to be working properly because the log files are populated with new lines whenever I use an action. The issue is that I can't read any output in real time by using journalctl. I was trying to replicate the virtual machine method described in the guide so that I could keep track of what was happening without looking for the log file itself but it appears only the initial part of the script output is sent to the journal. Is there any way around this?

I'm opening a new thread so as to not clog the one in the XivAlexander page, hopefully that's ok. You and someone else suggested deleting the definitions.json file in order to update the opcodes, but I'm afraid I have no idea where to find that file!
What I've been doing is run the iptables and curl commands on virtualbox every time I launch the game on my PS4. My laptop running the VM has to be running at all times since pressing Ctrl+C to exit the program just makes my game disconnect, and when I'm done for the day I can just pause the VM and put the PC to sleep.
Idk if that is the intended way to run the program but that's what has worked for me. Could you perhaps point me to where that file would be stored, or the correct way to use the program? Thanks so much for your work

Whenever I try to do step 5 of the guide, after following everything else to the letter, I just get greeted by a screen that says

Arch Linux install medium (x86_64, UEFI)
Arch Linux install medium (x86_64, UEFI) with speech
Arch Linux install medium (x86_64, UEFI, copy to RAM)
EFI Shell
Reboot Into Firmware Interface

Choosing any of the first three options just end up in a blank screen. Not sure where to go from here or did I configure something wrong?

I've been running this on an ARM SBC till recently but latest version requires me to provide an exe file and install x86 gcc libraries. I tried running the version prior the latest one and the game crashes with this error in console.

If I understand this correctly you now use the exe file and x86 gcc to automatically find possible new headers? I'd like to know if a workaround for ARM users is possible since I can't run x86 libraries on my system.