do I need to modify my nginx.conf?

events {
    worker_connections 1024;
}

http {
    server {
        listen 80;
        server_name site.luxmundicorp.com;
        return 301 https://$host$request_uri;
    }

    server {
        listen 443 ssl;
        server_name site.luxmundicorp.com;

        ssl_certificate C:/nginx/ssl/cert.pem;
        ssl_certificate_key C:/nginx/ssl/key.pem;
        ssl_protocols TLSv1.2 TLSv1.3;
        ssl_ciphers HIGH:!aNULL:!MD5;

        root C:/nginx/html;
        index index.html;

        location / {
            try_files $uri $uri/ =404;
        }
    }
}

Hmm you using cgnat or something there?

or a vpn?

nord vpn currently running but not connected

As that ip looks more in line with cgnat systems

oh sorry

or vpn blocks.

192.168.1.208

i copied the wrong one

Yeah so put that one

Later you can put it just to be on 80 and forget to use "SSL" TLS in there, Though if it has a redirect aim for the https port as you'll get infinite redirects if you ask traefik to hit say port 80 of nginx there

so in this should I change the http one to port 9443? and relaunch nginx?

Up to you whatever won't overlap with traefik's 80 and 443 binds

I'm just really confused haha, should I have nginx running currently or no?

Yes you should pick a port any port for nginx and reflect that in the rules

So what actually needs modified in the nginx.conf? should I change 443 to 9443?

and then ensure the nginx.yml url ends in 9443?

Well you can't have 2 things occuyping the host's "443" port so only one can take up said port so yes.

I'm so confused, I can't get my nginx to run on 9443 and it keeps spitting out this error 2025/06/03 22:06:38 [emerg] 6384

19476: cannot load certificate key "C:/nginx/ssl/key.pem": BIOnewfile() failed (SSL: error:80000002:system library::No such file or directory:calling fopen(C:/nginx/ssl/key.pem, r) error:10000080:BIO routines::no such file)

Though I can see it is properly running on https at https://portal.luxmundicorp.com

I do really appreciate your patience in helping me as I really do hate networking 😦

You can just have it run on http for nginx you can forgo the "SSL" part as Traefik would handle that for you.

But networking is simple once you get a good understanding~

nginx
events {
    worker_connections 1024;
}

http {
    set_real_ip_from 192.168.1.208;       
    real_ip_header X-Forwarded-For;       
    real_ip_recursive on;  

    server {
        listen 9443;
        server_name site.luxmundicorp.com;

        root C:/nginx/html;
        index index.html;

        location / {
            try_files $uri $uri/ =404;
        }
    }
}

Could do it like this depending if you don't want to worry about certs as Traefik will handle the TLS part for you. Though might want to remove the portal part if you did copy that from me on the /rules/nginx.yml as I thought the portal was another nginx thing you don't want conflicting routes set so just have the site part

hm my shit seems to have OOM'ed something

esi had a bad day

Ah

Yeah...

They‘re doing some pretty significant changes on their backend right now

ok, something is fucking off

had to nuke the cache

redis kept oom'ing because of errors

Eh don't worry zkill broke some stuff too new url! Whee!

Hi everybody. From the SeAT API I try to get all assets for a given character filtered by type/groupID eq 25 (Ships?), but the filter doesn't get applied and I get all assets. Filtering by e.g. quantity eq 1 works without any issue. Request URL as of Swagger:

https://seat.{some-server}.space/api/v2/character/assets/{char_id}?%24filter=type%2FgroupID%20eq%2025

I'm stuck and any help would be appreciated (to not get all assets and filter with jq).

hi, i just install new seat 5 server and it show inactive was i missing something ? I already set admin mail, esi

is this a blade or docker install?

blade

okay

https://eveseat.github.io/docs/installation/manual_installation/ did you follow all the steps?

yep i follow all step

what does service supervisor status give you?

or supervisorctl status

you skipped at least one step in the install instructions

https://eveseat.github.io/docs/installation/manual_installation/#supervisor

i forgot reload it

tks it work now

is there any way to get data more than 3 week ?

More data of what? But most likely no. We can only provide what esi provides.

Hmmm how is your storage looking?

Or is it first deployment?

May I bump this?

I ll check the storage but last time I check was good and is the 4 deployment training to deploy jajaja

did you set password for the database etc?

In the .env yes

after or before starting the stack

before

"exception": "[object] (Seat\\Eseye\\Exceptions\\RequestFailedException(code: 400): invalid_scope: Invalid scope(s) requested: esi-characterstats.read.v1 at /var/www/seat/vendor/eveseat/eseye/src/Access/AccessTokenRefresher.php:152)

I guess they changed something eh

https://esi.evetech.net/v1/killmails/127809339/8f46b6546627b1669cb019421f361495a0d12001/?datasource=tranquility

It's not working on SeAT SRP plugin

https://discord.com/channels/821361165791133716/821361903933718589/1381246041843826809

yeah that one might've been worth an announcement tbh 😄

ah great, now it's "Seat\Eseye\Exceptions\RequestFailedException: invalid_scope: Invalid scope(s) requested: esi-characterstats.read.v1 in /var/www/seat/vendor/eveseat/eseye/src/Access/AccessTokenRefresher.php:152" for days :v

weird

something is pretty clearly fucked even after I ran the update

k, cleared queues, let's see what happens at the next refresh

what the fuck, why is it still nagging about this shit?

Seat\Eseye\Exceptions\RequestFailedException: invalid_scope: Invalid scope(s) requested: esi-characterstats.read.v1 in /var/www/seat/vendor/eveseat/eseye/src/Access/AccessTokenRefresher.php:152

Remove it from the sso profile. You probably still have it after the update

I have updated it, let's see if that helps

updated... still happening

I guess it would be nice if you could, like, remove a scope gracefully...

seems like something has gone wrong with the database migration or similar, hm

@user_614098468218339348 any thoughts?

Hey guys, I could use your help in some docker related issues. I had SeAT running locally, then after the first restart I face some network not found issues 🤔

$ docker compose -f docker-compose.yml -f docker-compose.mariadb.yml -f docker-compose.traefik.yml up
[+] Running 8/8
 ✔ Network seat-docker_seat-internal  Created 
 ✔ Network seat-docker_seat-gateway   Created
 ✔ Container seat-docker-mariadb-1    Created
 ✔ Container seat-docker-cache-1      Created
 ✔ Container seat-docker-front-1      Created
 ✔ Container seat-docker-worker-1     Created
 ✔ Container seat-docker-traefik-1    Recreated
 ✔ Container seat-docker-scheduler-1  Created
Attaching to cache-1, front-1, mariadb-1, scheduler-1, traefik-1, worker-1
Gracefully stopping... (press Ctrl+C again to force)
Error response from daemon: failed to set up container networking: network b5dd2106a68edbaf8572dce176fcc633161040aee09105d7d05e02d179d34a7c not found

Already tried to down/up it again, also restarted Docker. Same result. Do you have any idea how to fix this? It looks like that the seat-frontend network is not properly created, right? Update: When I remove mariadb from the start up command, the error doesn't appear. Update2: I merged the content of the mariadb compose file to the main compose file and now it works 🤔 Maybe some race condition by design, idk

Did you see any errors during the migration? I was still seeing quite a few errors related to this scope even though I had applied the update two or so days ago. I did go into the SSO scope UI page today once I noticed my errors were significantly higher than usual and verified the scope wasn't there and saved the profile on my end manually (with no changes) for the scope dialog, then restarted the docker stack which effectively clears the queue completely. After the stack restart an hour ago I did get a few errors (80 ish) related to that scope, but looks like it's cleared up now, no scope errors in over an hour and jobs are completing. But I do admit I didn't verify the DB to see if the scopes had been removed successfully via migration, I just did the above due to time constraint on my side.

no errors that I could see

Last time CCP removed a scope I manually removed the scope from my profiles before it was updated in the SeAT code so I'm not as fluent with the "typical" behavior.

yeah that's what I did

but not this time around...

:v

I didn't do it before the update and my seat is just fine. Maybe just roll back and try update again?

I'm running a blade install so I guess I can restore from backup

I've had this happen before when the previous stack wasn't taken down correctly (possibly you missed a -f {dockerfile}.yml if you downed the stack or updated it prior without defining all the same files. If you're familiar with docker you could place something like the following in your .env file to include the other compose files so you never miss it. Example (which you'd replace with any of the compose files you're using/need):

COMPOSE_FILE=docker-compose.yml:docker-compose.mariadb.yml:docker-compose.proxy.yml

Once this is done, your command goes from docker compose -f {file}.yml -f {file2}.yml up -d to my preferred at least where individual files don't need specified (e.g. docker compose up -d). The disclaimer being that you need to be mindful of differences when running stuff from documentation since you don't have to specify them like this any longer. https://docs.docker.com/compose/how-tos/environment-variables/envvars/#compose_file

Yes I have it too. I don't quite know whats wrong though. The update did exactly the same thing as the last time they removed a scope. Back then it worked, but not now?

Hey yall i updated to latest images and im getting a ton of tmeout errors. before it was like 20 ish per month now it's more then 6.5K in 3-4 days. IS there anything i need to update?

Also the code apparently doesn't send a list of scopes to CCP, so I don't see how we are requesting the removed scope. It might be an issue on CCPs side, I'm not sure

well fuck 😐

@user_131158012831203328 can you clear the queue using horizon:clear, then manually update a character using esi:update:character <character_id> and see if it processes

I'll try

right now, the supervisor is shut down

CCP said that this issues lasted until around 17:00 UTC but it should be fixed. And for me, it seems to be working again

I just restarted the thing now... let's see

it's VERY angry atm 😄

mostly "attempted too many times" 😄

thats why I usggested to clear the queues

ran the job manually after clearing queues and it looks like it worked

also going through some chars I know has access to citadels that I think seat doesn't know the name of ... and seeing if that resolves some issues (I know it doesn't check everything)

yes so it looks like the issue resolved itself

yeah maybe 😄

did you follow the installation instructions to the t?

yea not te docker one but the manuel one

read that the mail could be blocking it as well

i have set it up with a local smtp server from mailpit with a valid from adress could this be it

https://eveseat.github.io/docs/installation/manual_installation/ you need to follow all the steps, closely

rgr i could try agen to the letter i am running on a mac so i need to make some le way but i have done those step realy closely

use a virtual machine for this shit

it's a lot less painful

yeah gonna switch to my hypernode virtual machine now on my windows pc and to it there

suck thoug got it to work propperly al is great exepct for the jobs not fetching any thing

and my own cli command with eveeys works fine

https://eveseat.github.io/docs/configuration/esi_configuration/

https://eveseat.github.io/docs/configuration/eve_administrative_contact/ this one

that one is important

yeah i have added that last night

and restart?

ofcource

the job out put is all is well within 10ms

and this is with debug mode on

and i am logging eve eye

so i know nothing is going to the eve api

ahh to bad i cant share my cli command that re uses most of you stuff but works

Do the jobs succeed or fail?

some times is success some times fails, no output in any log

Probably best to turn log level to debug, as then we can see if it's a middleware bouncing

yeah i have done that not getting mutch out of it, when i run the whole worker it just goes at light speed saying oke or fail within 10 ms. mostly ok but not getting data. To be honest my cli command works i got my own thing working but, i dont want tore invent the wheel cause you guys made a epic one. I made a wsl2 ubuntu server on my game rig, i am going to do a new install and follow the guide. Gonna start fresh,

Can you show us the logs?

.docs.troubleshooting

@user_614098468218339348, https://eveseat.github.io/docs/troubleshooting/

yeah i could got the eveeye logs and laravel log

How do you want me to show you the logs

they are big

just send them here

the one of today i needed to split up

I did a re install on my wsl machine and it runs bloody good

Hey, did some of you tried to run seat locally using laravel sail? Is this doable? nvm, works 🙂

Is the eve api having issues? Past two days my logs are full of hundreds of mb in Timeout waiting on backend. TemporaryEsiOutageException

I did get 15k failed jobs today. (Normally ~500). But since the update that seems to have resolved itself 😄

Why am I getting {"error":"invalidscope","errordescription":"The requested 'esi-characterstats.read.v1' scope is not valid."} when trying to login to my SeAT with SSO

install the newest update for seat

How do I update it?

just restart it?

https://eveseat.github.io/docs/upgrading/general/

I was getting that error and just updated now I get {"error":"invalidscope","errordescription":"The requested 'esi-ui.open_window.v1' scope is not valid."}

nvm I added the scope on the dev website of eveonline

did it wipe the database with the upgrade

I tried to restore a backup but now the front container just loops, here is a few lines from before it loops.

   INFO  Configuration cached successfully.  


   INFO  Routes cached successfully.  


   INFO  Running migrations.  

  2025_01_31_210832_add_db_citadel_access_cache ..................... 2ms FAIL

In Connection.php line 829:
                                                                               
  SQLSTATE[42S01]: Base table or view already exists: 1050 Table 'citadel_acc  
  ess_cache' already exists (Connection: mysql, SQL: create table `citadel_ac  
  cess_cache` (`citadel_id` bigint not null, `character_id` bigint not null,   
  `last_failed_access` timestamp null, primary key (`citadel_id`, `character_  
  id`)) default character set utf8mb4 collate 'utf8mb4_unicode_ci')            
                                                                               

In Connection.php line 587:
                                                                               
  SQLSTATE[42S01]: Base table or view already exists: 1050 Table 'citadel_acc  
  ess_cache' already exists                                                    
                                                                               

mysqld is alive
PONG
starting web service
Processing plugins from SEAT_PLUGINS
Installing plugins: cryptatech/seat-fitting,warlof/seat-discord-connector
Cannot create cache directory /home/seat/.cache/composer/repo/https---repo.packagist.org/, or directory is not writable. Proceeding without cache. See also cache-read-only config if your filesystem is read-only.

my start command is "podman compose -f docker-compose.yml -f docker-compose.mariadb.yml -f docker-compose.proxy.yml up -d"

Yea I’m not sure what happened cause after the pull it was like a fresh database of seat

I’ll try the database restore again on Monday

Thanks for the help

Don't feel bad podman might have the list of everything that might be better but also introduces modified behavior some stuff that should work doesn't..

Hey there! I just installed SeAT last night and I'm having some intermittent issues with contract notifications. Checking the logs, I seem to be getting 504 timeouts for when it fetches data from ESI. Any thoughts on what I should be checking first? I wonder if I am hitting some type of limit/backoff with ESI.

I should mention that the Discord notifcations come through sparsly. I get maybe about... 30% of them coming in, most of the time I dont get a notification when a contract has been assigned to a character EDIT: Hang on.... I have a thought... Lemme change a few things..

Im taking a guess, its either that you havent added your admin email https://eveseat.github.io/docs/configuration/eve_administrative_contact/, Or maybe something not allowing your "call backs" to the host?

Well, I think I solved the 504 errors. I had enabled all scopes during testing and I guess ESI was getting hammered too hard so it was backing me off. 😄 I slimmed down the scope requirements for my hauler and only have publicData for my customers. I am getting contracts pulled in, just no Discord notifications about being assigned a new contract are being fired off. I have received them before but then they stopped working. When I hit the Test button, a notification to the discord is sent just fine.

Thanks. Admin email was added during the install and nothing I'm aware of is being blocked back to my instance.

Hi, im have some error about assest updating, cause my stuff should be go other station but in website it still in old station

I try add character update in cron but it noti this error

and wier thing why i have two this, one my previous server setup it should be one, now it two

oh dear

You should not need this command scheduled. The one that updates characters is the buckets command

How long have you waited, and are your asset jobs fine

I think this can happen if you launch supervisor twice, but I‘m not an expert on blade server administration

Do the contracts that don’t generate notifications appear in seat when searching the characters contracts? Then we can narrow down if the issue is the notification or the esi side

Yes, they appear.

I will sometimes get updates on the contracts, like when they finish or get accepted but its sparse. I looked in the Horizon logs and saw that there was a contract notification sent at 00:00. Forgot to take a picture of it though. No more failed jobs in Horizon since I narrowed the scopes.

i just back home and recheck it fine rn, but it seem take ~ 2-3h to sync it

supervisor show only 1 run but seem it work normally i wont change it lol

does anyone have the issue with Scope profile being seen as unknown? This is not the new profile, I have it since the server is up and just saw it.

Yeah, mine show up like that too

Hey, just one question: Can we simply remove the owner who originally registered the application in the Developer Portal from the corp? Or would that cause issues? It shouldn't really matter—unless he gets upset and deletes the app. ?

It does not matter what corp the person is in that registered the application as the app registration is tied to the account level. However, if you don’t have control over the application/account it means that person could delete it or refuse to modify it if you need to one day and cause you a potential nightmare.

Yes, I know, thanks.

The titles job always fails for me

anybody know a fix ?

Well, I feel like a dumbass. Slimming down the scopes should have clued me in. I had the following scopes in SeAT, minus read_notifications and wasn't receiving any more contract created notifications in Discord. Once I added that scope back in, good to go now. I was thinking SeAT was only looking at the contracts themselves. Oh whale. That was a fun experience!

Hello i am very new to these things but i managed to get seat up and runnning, however is there something i need to do to get the reactions candidates in the Moons reporter detail view to display? I thought if i updated seat and sde it would but no luck after waitng a couple days

Hmm, cant say I have tried that one in a while. Are you able to send me an example moon report in a DM for a moon that doesnt have it showing

Yes. We should probably clarify somewhere in the docs, that a large percentage of our inbuilt notifications are based on actual in game notifications not the items themselves in SeAT.

is there a way to delete all characters registered at once? I feel like I am missing something, but maybe you have to do it manually?

As an owner or a user?

sorry i never got back to you, you migth not even remmeber at this point what this was about and i don't blame you. Long sotry short, RL happened and then more RL happened and i was recovering from a medical issue , all good now though. Anyhow, i never managed to connect to the DB (and didn't try for obvious reasons). Yesterday is ran into the exact same behaiour with a new user though and managed to fix it without really identifying the issue. He tried to change his main character by deleting characters with a junior officer. In the end, his user account was still there and whenever he registered his characters again the issue occured, where it never pulled any info beyond public data even though showing like it had full scopes. By deleting his user on top of that after pulling all characters we got it to work. Maybe that clarifies the issue for me and once i'm fully back to the point where i can sit on a PC for prolonged times i'll look into why the db refuses my connection attempts. thanks for your time ❤️

I do remember. If it is what I suspect, your observation that deleting the user fixes the problem matches my expectations. If you got the time, we can continue troubleshooting any time

Hi, is it possible to export results from the main search function? Rather than having to sift through hundreds of pages of only 100 lines

I've been wondering the same... 😄

trying to find something from this is counter intuitive

Feeling that pain :v

I hate to imagine Wiblas number

the biggest thing for me is that search is godawful slow and moving between pages like the above takes fucking ages and bogs down the server

I've had it OOM before

because of it

so yeah

Hey fellas, I observed that my sidebar configuration conflicts with other packages. 🤔 If I install e.g. seat-text to my local seat, the sidebar of my own package disappears. When I remove it, the sidebar entries are there again... I just use the line from the docs:

$this->mergeConfigFrom(__DIR__ . '/Config/package.sidebar.php', 'package.sidebar');

Also tried to change the namespace package to something else, but then there is no sidemenu entry for my package at all.

This shouldn't happen. What is your package.sidebar.php?

Also, did you run php artisan config:clear after doing changes

I cleared whole cache after each install sail artisan cache:clear

<?php

return [
    'text' => [
        'name' => 'Something',
        'icon' => 'fas fa-shopping-cart',
        'route_segment' => 'something',
        'permission' => 'something.view',
        'entries' => [
            [
                'name' => 'Sub Menu',
                'icon' => 'fas fa-search-dollar',
                'route' => 'something.index',
                'permission' => 'something.view'
            ],
        ],
    ],
];

so it does conflict because you name your sidebar entry the same as the one in seat-text.

php
<?php

return [
    'use_something_different_here' => [
        'name' => 'Something',
        'icon' => 'fas fa-shopping-cart',
        'route_segment' => 'something',
        'permission' => 'something.view',
        'entries' => [
            [
                'name' => 'Sub Menu',
                'icon' => 'fas fa-search-dollar',
                'route' => 'something.index',
                'permission' => 'something.view'
            ],
        ],
    ],
];

omgosh, because of the text at the top. I thought it was some kind of reserved "type" keyword.

Thanks for the quick hint. I was blind

Temporarily impaired, maybe

Is there any plugins for SeAT 5 for doctrines?

This is the one I use. https://github.com/eveseat-plugins/seat-fitting

tyvm, exactly what I was looking for

I am getting extremely frustrated with SeAT, my seat-front keeps crashing and nothing I do works

Got logs..?

no idea if this is accessible https://pastebin.com/tmTMYe8X

Are you not using a cert for some reason?

Got logs from seat-front or a ``docker compose ps`` ?

:: too many certificates (5) Ah are you persisting the certs properly?

Can you show the traefik section?

in the compose

everything shows as up now, so I guess I made a mistake somewhere.

ml
root@auth:/opt/seat-docker# docker compose ps
NAME                      IMAGE                         COMMAND                  SERVICE     CREATED         STATUS                          PORTS
seat-docker-cache-1       redis:7-alpine                "docker-entrypoint.s…"   cache       5 minutes ago   Up 5 minutes
seat-docker-front-1       ghcr.io/eveseat/seat:latest   "/docker-entrypoint.…"   front       5 minutes ago   Restarting (1) 55 seconds ago
seat-docker-mariadb-1     mariadb:10.11                 "docker-entrypoint.s…"   mariadb     5 minutes ago   Up 5 minutes
seat-docker-scheduler-1   ghcr.io/eveseat/seat:latest   "/docker-entrypoint.…"   scheduler   5 minutes ago   Up 5 minutes                    80/tcp, 8080/tcp
seat-docker-traefik-1     traefik:2.10                  "/entrypoint.sh --ap…"   traefik     5 minutes ago   Up 5 minutes                    0.0.0.0:80->80/tcp, [::]:80->80/tcp, 0.0.0.0:443->443/tcp, [::]:443->443/tcp, 0.0.0.0:8080->8080/tcp, [::]:8080->8080/tcp
seat-docker-worker-1      ghcr.io/eveseat/seat:latest   "/docker-entrypoint.…"   worker      5 minutes ago   Up 5 minutes                    80/tcp, 8080/tcp

I had a cert, but I guess that broke something, again, for the 500th time since Sunday lol

``docker logs seat-docker-front-1``

it seems I'll just have to wait another day and hope this sht is fixed xD

I really didn't want to pay for setup yet again after spending 2.2bil on Alliance Auth. I wanted to learn and do it myself

huh

Guess you should go shoot the person who did It I guess, I could do it if it's not resolved later I guess.

helloo guys. Im having trouble getting my notifications to work prop. the job keep failing me . However when i press the test button it work fine

that seems like a bug in seat. Are you willing to do some database digging to extract this particular notification for debugging?

hm, 391k failed jobs

oof

How's your jobs holding up with people shifting? Mine had a mild panic attack when we moved our bpo cans from impass 🤣

well ... a lot of failures so I don't know 😂

Sounds like my corp members more than my jobs 🤣

hi there, I'm trying to install some packages on the docker install

unable to get image 'redis:7-alpine': permission denied while trying to connect to the Docker daemon socket at unix:///var/run/docker.sock: Get "http://%2Fvar%2Frun%2Fdocker.sock/v1.51/images/redis:7-alpine/json": dial unix /var/run/docker.sock: connect: permission denied

this is the response I get

any idea how to fix this?

nvm im stupid

fixed

I didn't have an affiliation. That fixed it.

Hola. Running seat:5 via Docker, seems my guys are unable to login to Seat since today, getting a "The requested 'esi-characterstats.read.v1' scope is not valid." I read this scope is depcreated, why do I run into this? Did I miss a manual cleanup with some realease, or..?

update to the latest version

.docs.upgrading

@user_614098468218339348, available .docs commands are: .docs.upgrade, .docs.moons, .docs.squads, .docs.donate, .docs.troubleshooting, .docs.requirements, .docs.bugs, .docs.help, .docs.auth, .docs.email, .docs.plugins, .docs.sso, .docs.install.docker, .docs.install.manual, .docs.logs

https://eveseat.github.io/docs/upgrading/general/

thanks. Was wondering, how to use what the Bot posts 🙂

I am running Watchtower, If you just wanted to point me to "run docker compose pull" -- this will not be the issue here rather 😦

yeah the commands that work it will link the docs directly. But there aren't commands for every docs page

I have a onenote notebook for these things...

can be a tip 🙂

doesn't have to be onenote, but having somewhere to put "runbooks" for your own reference can be a good idea

Dont really see what I can do here:

• as stated, I automatically pull latest images

• there are no manual steps for Docker documented (that is all `Blade Installation?!)

I had to run a manual php artisan seat:cache:clear which seems to have resolved my Issues. Thanks! Suppose that is related to your guides stating "run docker compose down && up" for Upgrades -- but since watchtower does handle them rather automatically, I did miss the cleaning of Redis and whatnot..

yeah that's still required iirc

or things get squirrely

Just becareful with using watchtower on criticial containers..

reliably get these errors from one specific toon

It says the state is wrong, but litterally had them stream logging into seat, clicking link character and it erroring

getting a lot of errors recently for Seat\Eveapi\Jobs\Character\Titles

Same

Same...

My errors topped out at half a million before my server croaked

Had to clear cache and restart

Hi all. I've gotten seat up and running for my corp using a vps. However, I keep getting this message whenever anyone accesses the webpage via our domain. Going to look at the cert, its a TRAEFIK DEFAULT CERT, so I imagine that I've setup Traefik cert incorrectly? Anyone have a link to a guide for how to set it up correctly or could talk me through it? Thanks

How to dump the corp's papas for a chosen month into a csv format? Like 2025-06.

Looks like the SSL wasnt setup correctly. Did you setup the .env for "Lets Encryp"? https://eveseat.github.io/docs/installation/docker_installation/#seat-docker-configuration

Are you using Docker or Baremetal?

I doubt they are using traefik in baremetal if they were they wouldn't have a default cert being configured as they'd understand how to deploy traefik on bare host at that point..

Check the compose see if you uncommented the LE lines

Using docker install. Following the documentation, In the .env file I set the traefik email to my own email account. Looking at docker-compose.traefik.yml, I dont see any lines commented out. What was unclear to me is if I needed to create a lets encrypt account(?) or similar in order to get a certificate? I feel like im missing a step here.

You don't need to create any account with Let's Encrypt. Maybe check the folder "acme" if there is any data or issue with permissions.

Also please post the traefik logs

all i did was follow this guide for SSL, works great https://www.digitalocean.com/community/tutorials/how-to-secure-nginx-with-let-s-encrypt-on-ubuntu-20-04

This is not useful or relevant when using the docker deployment method with traefik

in the acme folder I have an acme.json that has account email, registration, and a private key. Underneath it has a keytype (labeled 4096) and a certificates labeled null. That certificates makes me suspicious. I have also followed the instruction in the docs to chmod 600 acme/acme.json, which I imagined would take care of any permission issues?

If its helpful, I used the automated setup script.

Where can I find these? I can get it to print basic logs (docker compose logs) but I can't find anything relating to traefik specific logs

docker compose logs traefik

huh

wait

I thought being in the wrong directory would matter, but i get the same thing in /opt/seat-docker

try docker compose -f docker-compose.yml -f docker-compose.mariadb.yml -f docker-compose.traefik.yml logs traefik in the installation directory

If there is some kind of data in the acme.json file, then i dont think it was a permission issue. Running chmod for the permission wont hurt it.

But lets see what the logs say..

Follow that if you hate yourself with traefik as traefik has certbot intregated effectively

``docker compose -f docker-compose.yml -f docker-compose.mariadb.yml -f docker-compose.traefik.yml ps`` Any traefik containers?

that ran and I got this message very early on. I'm working on getting the full logs off my vps to share, but that's trickier than I thought. I imagine that this would point to what the issue is?

looks like it doesn't appreciate your DNS

that looks normal? unless I don't understand something

Yes, but im not sure why? Letsdebug.net doesnt show any issues for my domain, and I can access the site normally if I just select to skip past the certificate.

I think she needs to try to generate the cert again

she

oh sorry

see also: tired 😐

lol

also used to EVE being a sausage fest, so that breeds the assumption 😛 without having looked at the name at all (or the profile)

Trying to setup discord connector for my SeAT, question that I can't find an answer to in the docs, what is a set?

set=role when talking about the discord connector

It wasn't the last message, but I might pull it down and then back up again if it doesn't resolve on further testing.

This makes me think that its cached on my devices, as im still getting the error. Ill try and test when I get off work on a device that I havent used before.

Would this work if I restarted seat? I'm not sure how to generate the cert again

If you still get the same traefik error after restarting: Do no try multiple times. You can get blocked for causing too many errors

yes, docker compose -f ... down, then up

https://tenor.com/view/dozens-of-us-there-are-dozens-of-us-dozens-arrested-development-arrested-development-dozens-gif-18162954729757007693

Thanks, Ill let is be for a bit and then if I need to try again later

Thanks for all the help

Is there a package to help identify AFK people in corp, i can see that there is simplyunnamed/seat-user-last-login but it says its not on SeAT 5

You're using http validation so you need to make sure a valid A record points directly to the server and have port 80 open.

Yeah and the block is 1 week.. @user_276156199299186688

If still doesn't work it could be firewall or region blocks or something or just opt for dns validation and skip all this

https://go-acme.github.io/lego/dns/ionos/

I believe this is the code to run to get the SSL.

echo "Setup the domain"
while true; do
  while [ -z "$SEAT_DOMAIN" ]
  do
    read -p "SeAT domain (eg: seat.domain.tld): " SEAT_DOMAIN
  done

  read -p "SeAT will be served over https://${SEAT_DOMAIN} - is that correct ? [Y/n]" -n 1 -r
  echo    # (optional) move to a new line
  case $REPLY in
    [yY]|"") sed -i -- 's/SEAT_DOMAIN=seat.domain.tld/SEAT_DOMAIN='"${SEAT_DOMAIN}"'/g' .env
      break ;;
    *) SEAT_DOMAIN=''
       echo "No changes have been applied. Please provide the domain on which SeAT will be served"
  esac
done

echo "Enabling SSL entrypoint"
echo "Please provide a valid e-mail address for Let's Encrypt account creation (this service will handle your SSL certificates) - leave empty to not use SSL"
read -p "e-mail: " ACME_EMAIL
if [ -n "$ACME_EMAIL" ]; then
  sed -i -- 's/TRAEFIK_ACME_EMAIL=you@domain.tld/TRAEFIK_ACME_EMAIL='"${ACME_EMAIL}"'/g' .env
  sed -i -- 's/      #- "traefik.http.routers.seat-web.tls.certResolver=primary"/      - "traefik.http.routers.seat-web.tls.certResolver=primary"/g' docker-compose.yml
else
  echo "No e-mail address has been provided, SSL will not be available"
  echo "SeAT will be reachable on http://${SEAT_DOMAIN} only"
fi

echo "Preparing an acme.json file for Traefik and Let's Encrypt"
mkdir acme
touch acme/acme.json
chmod 600 acme/acme.json

Otherwise @user_276156199299186688, maybe it would be best to blow up VPS and start again from scratch.

The lines should still exist likely commented

no point in going scorched earth just to update some lines if they are commented..

Yeah it's generally not the best approach to just nuke a setup. If you can figure out what is wrong, then fix it and move on. If you can't figure out what is wrong, nuking the setup won't help you learn

If the lines are commented it's just a matter of uncommenting them in docker-compose.traefik.yml and also modifying the env file to have the right variables set.

Question, i'm getting Gateway Timeout's on my Seat instance (docker/traefik), but I'm not seeing anything hitting the logs in storage/logs within Seat, neither does anything hit /var/log on the main server. Turning on debug hasn't shown anything either. Any pointers are welcome lol docker-compose ps Name Command State Ports ----------------------------------------------------------------------------------- seat-dockercache1 docker-entrypoint.sh redis ... Up seat-dockerfront1 /docker-entrypoint.sh web Up 80/tcp, 8080/tcp seat-dockerscheduler1 /docker-entrypoint.sh cron Up 80/tcp, 8080/tcp seat-dockerworker1 /docker-entrypoint.sh worker Up 80/tcp, 8080/tcp nmap localhost PORT STATE SERVICE 22/tcp open ssh 80/tcp open http 443/tcp open https 3306/tcp open mysql

Check seat-dockerfront1 logs.

It's not going show in /var/log and not sure why you are getting a 3306 as being open either.

As I wrote, there is nothing hitting the laravel logs. Will try to add some debug logging in traefik tonight. Workers esi seem to run fine, if I look in the db things still get updated.

Look at the front container logs. Bad gateway usually means seat didn’t even get to laravel, so laravel logs are empty

is this a known issue? my logs are now getting to be 70mb per day because of this error

traefik time="2025-07-17T23:38:25Z" level=info msg="Starting provider *acme.Provider" time="2025-07-17T23:38:25Z" level=info msg="Testing certificate renew..." providerName=primary.acme ACME CA="https://acme-v02.api.letsencrypt.org/directory" time="2025-07-17T23:38:25Z" level=warning msg="Defaulting to first available network (&{\"seat-dockerseat-gateway\" \"172.29.0.2\" '\\x00' \"\" \"718ee2c4e319ef70b23396a8e8a1d7cc9531319765b6610b5139fd2aceb27f9b\"}) for container \"/seat-dockerfront1\"." container=front-seat-docker-4f10d379c7bc2967e810c4f044f2468bce5440e39f94c4660815c7ecc899da0e serviceName=seat-front providerName=docker time="2025-07-17T23:38:26Z" level=warning msg="Defaulting to first available network (&{\"seat-dockerseat-internal\" \"172.27.0.4\" '\\x00' \"\" \"b81619fac16bac3156f909092f968114d9bb9f5001531253e398a7df3cd41fad\"}) for container \"/seat-dockerfront1\"." providerName=docker container=front-seat-docker-4f10d379c7bc2967e810c4f044f2468bce5440e39f94c4660815c7ecc899da0e serviceName=seat-front time="2025-07-17T23:48:25Z" level=warning msg="A new release has been found: 3.4.4. Please consider updating." front AH00558: apache2: Could not reliably determine the server's fully qualified domain name, using 172.29.0.2. Set the 'ServerName' directive globally to suppress this message AH00558: apache2: Could not reliably determine the server's fully qualified domain name, using 172.29.0.2. Set the 'ServerName' directive globally to suppress this message [Thu Jul 17 23:38:37.858377 2025] [mpm_prefork:notice] [pid 1:tid 1] AH00163: Apache/2.4.62 (Debian) PHP/8.2.29 configured -- resuming normal operations [Thu Jul 17 23:38:37.858412 2025] [core:notice] [pid 1:tid 1] AH00094: Command line: 'apache2 -D FOREGROUND' After that nothing

I'm not sure, but it seems that the seat gives the corporation's directors the right to view everything related to the corporation, including the corporation's members' accounts, etc. I can't figure out how to turn this off.

Access management

https://your_seat.com/configuration/access/roles

Character options in role that you want to edit.

Idealy some of your directors should have that access to keep an eye on potential market fuckery etc if your alliance cares about that. Unless you have specialized role for that.

I'm not talking about this tab, because no matter what I change, the directors can still do it, and I don't want them to. this role is either hidden or I'm too stupid to see it.

I currently have one general role for people that only allows them to see killmail, but people with director rights in eve can see much more than the role allows.

That's because its baked into seat. CEOs and directors get full access to their own corps. That's not configurable

That's because its baked into seat. CEOs and directors get full access to their own corps. That's not configurable

oh

kk

ty

@user_301981661761896449 any chance you can accept my 2 pull requests for seat-srp and push a new release?

To get back on this. Things started working again yesterday evening (after yet another down / up -d in debug mode) but I still have zero clue what was wrong during those 3 days of downtime

what sheduler i need to run for updaating the corperation titels ?

It is part of the whole character update process using buckets

Will get to looking at them shortly

friendly poke. the SRP ones are the most important imo

how do you "run an admin login"? only way to log into my instance is with SSO (through the webpage I mean... I can log into the Ubuntu CLI just fine) nvm I found the article https://eveseat.github.io/docs/admin_guides/admin_login/

I'm about halfway through my backlog of prs from while I was away.. Seat won't be too far away

np

Hello. An error occurred while viewing the corporation starbase page. The error occurred today after one POS was removed. How can this be fixed?

The error is probably due to this line in the database

Yeah it's a known issue when a starbase is unanchoring https://github.com/eveseat/seat/issues/877 ill take a look tonight on resolving

is there any news? or can i delete bad rows in db directly to fix problem?

if you know what you are doing, yes you can delete the row. However, seat might add it again on the next ESI pull

if it was unanchoring? doubt it

Hi everyone! Over the past few days, my Seat server has been frequently showing errors like the ones in the screenshot. Has anyone else encountered this issue? I'm running the latest version.

Can you show us the details why a job failed?

is it ? from laravel log

eseye log?

only character and structure error character errors are related to deleted tokens and characters sent to dunheim

Is there a way to remove characters that have been moved to Doomheim? SeAT is running in Docker.

Could my issue be related to that message? Should I try adding that parameter to the config file? Since I’m seeing a lot of errors specifically related to structures. https://discord.com/channels/821361165791133716/1345031288226054245/1345378583333568634

Hi Team. Looking for help with creating operations via the calendar. I get the following error, and the event does not get created. All updates have been performed including calendar and fitting modules. It seems that there is no entry for "doctrineid" in the DB (assuming I've read that correctly) Any insight would be greatly appreciated. SQLSTATE[42S22]: Column not found: 1054 Unknown column 'doctrineid' in 'INSERT INTO' (Connection: mysql, SQL: insert into calendar_operations (title, role_name, importance, staging_sys, staging_sys_id, staging_info, fc, fc_character_id, description_new, start_at, integration_id, doctrine_id, user_id, updated_at, created_at) values (Test, ?, 2, K-6K16, 30004751, Okunda Standing Pew Pew, KayLe, 162266737, Okunda Standing Pew Pew, 2025-08-03 12:05:00, ?, 2, 7, 2025-08-03 12:07:03, 2025-08-03 12:07:03)) {"userId":7,"exception":"[object] (Illuminate\\Database\\QueryException(code: 42S22): SQLSTATE[42S22]: Column not found: 1054 Unknown column 'doctrineid' in 'INSERT INTO' (Connection: mysql, SQL: insert into `calendaroperations (title, rolename, importance, stagingsys, stagingsysid, staginginfo, fc, fccharacterid, descriptionnew, startat, integrationid, doctrineid, userid, updatedat, createdat`) values (Test, ?, 2, K-6K16, 30004751, Okunda Standing Pew Pew, KayLe, 162266737, Okunda Standing

what fitting module are u using?

what calendar and what is the seat version

Seat 5.0.1; hermesdj/seat-calendar 1.0.8; dysath/seat-fitting 5.0.9

can you check this in your db? SHOW COLUMNS FROM calendar_operations LIKE 'doctrine_id';

if you didnt change anything in your setup during installation this is how to get into DB @user_554081948201517056 Just please confirm you are docker seat user. If so: docker compose exec mariadb mariadb --user=seat -p --database=seat it will ask you for a password. this one you will find in your .env file. by default its located in /opt/seat-docker and after getting into it run the command that I asked you for.

This is not the correct version of seat fittings. You should use this one https://github.com/eveseat-plugins/seat-fitting

If doctrineid is missing from the calendaroperations table it also means a db migration did not execute correctly

Yes this is the schema we have active:

I'll update the plugin as you suggest....

hmm this is the current setup

what Jay said ---- If doctrineid is missing from the calendaroperations table it also means a db migration did not execute correctly

Yeah. I checked the migrations in the latest version of fitting plugin and none of those migrations add the column to the calendar ops table

It's the calendar plugin that adds it, not the fittings plugin

https://github.com/hermesdj/seat-calendar/blob/master/src%2Fdatabase%2Fmigrations%2F2024_08_16_110100_alter_operation_table.php

Ok will take a look

I have the uninstalled the calendar plugin, and reinstalled it - still no doctrine_id column? Is there any way to force the migration to run from the commandline, or is that dangerous in terms of wider integrity...

can you run php artisan migrate:list

can you try to run php artisan migrate:status in the front container?

also what does php artisan migrate print?

let me try that....

Could you please drop the command for how i get an interactive terminal in front container?

docker compose -f <your docker-compose files depending on if using traefik or proxy> exec front php artisan migrate

thanks !

and the status one?

so it reports it as already ran when it appears it hasn't

this is the strange thing yes

the schema of the table remains unchanged - no doctrine_id

can i force a migration to rerun? perhaps

On another note, did you do anything like restoring a backup or migrating the database on this instance?

no definitely not

this has been a clean docker install, very stable and adding plugins incrementally

running for some time now, the problem we have now is the first issue

i removed the calendar plugin, restarted to force the unmigration, then reinstalled again and no migration

friendly ping <3. sorry, the srp one is the most impactful that my prod is waiting on

as I understand it, everyone's custom styles for the authentication page have been reset ? I can't understand why they aren't working now

Hi! I’m still experiencing issues with character data updates. For some characters, updates complete within a few hours, but for others it can take 2–3 days. There are ESI status errors, as well as errors related to structures, deleted characters, or characters whose tokens have been removed. I’ve also noticed something strange: even though the SSO profile name is present and set as the primary one, some characters in the lower part of the access list show the following line: Scopes Profile: Unknown (Profile not found) The seat:buckets:update task is scheduled as: /20 * There are a total of 1080 tokens. The service is running on a virtual server with 8 CPU cores and 16 GB of RAM. What would you recommend to improve or fix the issue?

are the characters without the scopes profile the same as the ones that don't update

Two different characters show the same status, but one of them updates correctly while the other doesn’t

The same profile is being used

I've also noticed some other strange behavior: for some newly added characters, not all information gets updated. For example, standings, wallet transactions, and the skill queue are visible — but the skills section, total SP, and wallet balance are empty. More than 24 hours have passed since the character was registered.

are there jobs for the missing data, and do they succeed or fail?

It seems that for some reason, the jobs for those characters weren’t being created. I’ve just added another profile with full access, and now the information is updating quickly for it. I’ll try setting this profile as the default and use it for now. I’ll keep monitoring the situation. The list of ESI scopes is identical between the profiles.

are you using docker?

I will need your help btw

my styles for the authentication page for some reason don't work anymore

before this it worked but after updating seat it no longer works and takes the standard page

docker-compose.override.yml

services:
  front:
    volumes:
      - /opt/seat-docker/custom/custom-layout-mini.css:/var/www/seat/public/custom-layout-mini.css
      - /opt/seat-docker/custom/logo.png:/var/www/seat/public/logo.png

before this file contained a version but I saw in the console that the version designation should be removed but it didn't help

WARN[0000] /opt/seat-docker/docker-compose.override.yml: the attribute `version` is obsolete, it will be ignored, please remove it to avoid potential confusion 

I assume u did up -d with override right?

let my check

i use

docker-compose -f docker-compose.yml -f docker-compose.mariadb.yml -f docker-compose.traefik.yml up -d

and also i can remove -d some times

ok so each time If u modify something you need to add this ... you used default... so it works as it should. instead of this command use docker compose -f docker-compose.yml -f docker-compose.override.yml -f docker-compose.mariadb.yml -f docker-compose.traefik.yml up -d after any update you do from now on to include override file. That includes any css changes etc.

Thank you

yes

It's normal and can be ignored, but if you want to get rid of that message, add a

before the first line version: "3.2" in the docker-compose.yml file.

Newer versions of Docker no longer require specifying the version.

I've already figured it out.

I think it should help you.

try to re-request everything in turn where possible btw it helped me.

there was the same problem

I tried those commands, but they didn’t help. Everything is working fine so far with the new SSO profile — hopefully it stays that way.

Oh yes, the SSO was recently updated in seat.

this may have caused the error.

As a curiosity, the previous default profile had fewer permissions than the one referenced? My instance has always been the opposite where the most permissions were on the default profile, but wondering if that understanding might help with any reproduction.

I have a problem, namely, the market is not loading, the forge region is selected, but there are no orders in the Market Browser

tasks are also not created when requesting a market, although I do not know where the market data comes from.

The market updates once a day. The job is called Orders or something similar

i use esi:update:prices

It helped

they are the same

I encountered the same problem as you last year. when by default it was only with public data and when linking alts, nothing but the public was updated Despite the fact that the main was registered with SSO with full rights

Is there anyway to bulk delete characters from SEAT? When you delete one it always resets back to page 1 and takes forever to go through them all

use the api

(its how I do it)

latest update is getting front-1 | Your requirements could not be resolved to an installable set of packages. front-1 | front-1 | Problem 1 front-1 | - eveseat/eveapi is locked to version 5.0.27 and an update of this package was not requested. front-1 | - eveseat/eveapi 5.0.27 requires maatwebsite/excel ^3.1.45 -> satisfiable by maatwebsite/excel[3.1.66]. front-1 | - maatwebsite/excel 3.1.66 requires phpoffice/phpspreadsheet ^1.29.12 -> satisfiable by phpoffice/phpspreadsheet[1.29.12]. front-1 | - maennchen/zipstream-php 3.2.0 requires php-64bit ^8.3 -> your php-64bit version (8.2.29) does not satisfy that requirement. front-1 | - phpoffice/phpspreadsheet 1.29.12 requires maennchen/zipstream-php ^2.1 || ^3.0 -> satisfiable by maennchen/zipstream-php[3.2.0].

🤔

Hmm odd.. My testing showed it working locally on both 8.2 and 8.3.. Maybe I missed something updating deps. Will have a look shortly

i could be completely wrong, but it looks like the dockerfile forces php8.2 (https://github.com/eveseat/seat-docker/blob/master/Dockerfile#L1) and somewhat recently zipstream-php removed 8.2 and requires 8.3

im not too familiar with the docker process though, ive spent some time trying to debug and no idea

Yeah the docker file does use 8.2. And only mainly because I haven't put in any effort to check for regressions when going up to 8.3 or 8.4. That said a lot of the manual install guys tend to run 8.3 so it's probably safe enough.

is there an easy way for me to change that locally on my docker install? i dont think i can override what the repo says

right now my entire seat installation is offline, just trying to find a way to get it back up

Im about to push an 8.3 image

You can be my test hamster for the 8.3 image. As I am on mobile now and cant check if it boots easily

hamster reporting for duty

@user_301981661761896449 that was it, it booted

seems like its functioning well

Cheers. Let me know if there's any trouble and I'll have a closer look

But should be OK in theory lol

Also in theory slight performance gains are possible. At this point I have no benchmark to test that though

Hello, I can login to the seat, but calendar is 500. Does anyone else see this issue?

.docs.troubleshooting

@user_301981661761896449, https://eveseat.github.io/docs/troubleshooting/

Check what your logs say

thx

When I click the calendar, I got front-1 | 172.18.0.1 - - [13/Aug/2025:14:36:34 +0000] "GET /queue/short-status HTTP/1.1" 200 1213 "https://seat.blabla.cc/home" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/138.0.0.0 Safari/537.36"

this is the live log

it says nothing

I was deploying a new version this afternoon, maybe update seat and check ?

ok will give it a pull and restart it

o7

I just have a few simple questions when someone has a moment. I am not a professional. I've been at it a few days now. 😆

Just ask in here and we will try to help 🙂

The laravel log is where all details should be

So there's an issue with the either the config or the discord client. I'm assuming you have it set up to reach out to discord.

Hey ya'll - for whatever reason the seat 5.0 calendar plugin stopped working for me and i get this error. Where do i look to see what happened?

.docs.troubleshooting

@user_301981661761896449, https://eveseat.github.io/docs/troubleshooting/

Look at the laravel logs

But given the pattern with the messages directly above yours, may be something for @user_164076488294006785 to have a look at.

I see this

maybe it's the same problem?

and another one regith below that one [calendar][discord] Client error: GET https://discord.com/api/guilds/xxxxxxxxxxxxxxxxxxx/channels resulted in a 401 Unauthorized response: Context: { "message": "401: Unauthorized", "code": 0 }

ooops not sure if i should have hidden that....

use the tilde (`) key three time before the text/code to open and again at the end to close it

Ah yes the discord bot need new permissions. I'm going to patch it so it doesn't break There is a new feature where you can set a voice channel for an operation. The event created on discord by the plugin set the channel as the "location" and only people who can see that channel can see the event. Security measure ^^ If you want that to work you need to go to calendar settings and click on save under the discord credentials form. After I fix this error...

should update to seat-calendat@1.1.0

If I update the bot permission, this should solve the issue?

yep

it needs MANAGEEVENTS, CREATEEVENTS, CONNECT, VIEW_CHANNEL

(the last 2 are the new ones)

how to update the discord bot permission?

but I couldn't open the setting tab rn

did you update to 1.1.0 ?

oh, just realized I should do it first

get it fixed

thx

hello. I will try to make this quick. I can log into SeAT, but after I auth the character the website refuses to connect. I assume my callback url is incorrect?

https://www.horizonlogistics.org/auth/eve/callback