do I need to create these folders on the server?

nvm. My config file was in the wrong location. thanks.

\o/ its up! and I was today years old when I learned this is called a "bare metal" install.

is there another plugin for HR besides the cryocaustik/seat-hr? I keep getting a 500 server error.

Can you check logs what is the error?

Settings - > system logs

It was missing an entire table and throwing an ajax error. I fixed it.

haha jeez

Hello, after restoring backup to new server have an loop error:

front-1      |
front-1      |    INFO  Routes cached successfully.
front-1      |
front-1      |
front-1      |    INFO  Running migrations.
front-1      |
front-1      |   2022_01_07_124200_create_job_batches_table ........................ 4ms FAIL
front-1      |
front-1      | In Connection.php line 829:
front-1      |
front-1      |   SQLSTATE[42S01]: Base table or view already exists: 1050 Table 'job_batches
front-1      |   ' already exists (Connection: mysql, SQL: create table `job_batches` (`id`
front-1      |   varchar(255) not null, `name` varchar(255) not null, `total_jobs` int not n
front-1      |   ull, `pending_jobs` int not null, `failed_jobs` int not null, `failed_job_i
front-1      |   ds` text not null, `options` mediumtext null, `cancelled_at` int null, `cre
front-1      |   ated_at` int not null, `finished_at` int null, primary key (`id`)) default
front-1      |   character set utf8mb4 collate 'utf8mb4_unicode_ci')
front-1      |
front-1      |
front-1      | In Connection.php line 587:
front-1      |
front-1      |   SQLSTATE[42S01]: Base table or view already exists: 1050 Table 'job_batches
front-1      |   ' already exists
front-1      |
front-1      |
front-1 exited with code 1

Any ideas to fix this?

Are you restoring a seat 4 backup onto a seat 5 install?

Nope im restored seat5 backup to fresh installation p.s. - have no idea what happened but problem gone 🤷‍♂️

I am using docker within WSL2 to host SEAT, but I am having an issue displaying any info. I am able to login and see the web interface, but it shows nothing for my character info and says I am not apart of a CORP. Is there a way I can check if the data is not making it there?

Does it have to be public facing to work?

Do the jobs for your character get scheduled? If so, do they succeed? If not, what is the error you get?

SeAT must be able to reach ESI, but the web interface doesn't have to be public-facing

They do appear to succeed, but the “data” area is just blank. I know it can reach ESI, because the diagnostic program ran and said everything was fine. But it still just shows nothing.

If the job succeed, the data should be there. Are you sure the jobs succeed for the right characters? And what kind of jobs are succeeding? There are also a few jobs not related to characters.

A screenshot is probably the easiest

the way you ask your question it seems like you have a new install. Did you set an admin email, and did you configure the ESI secrets?

Yes on both

"Seat\Eveapi\Jobs\Assets\Character\Assets Collapse Queue high Pushed 2025-08-16 00:57:32 Completed 2025-08-16 00:57:35 Data Collapse { "queue": "high", "characterid": <OPSSEC>, "token": { "class": "Seat\Eveapi\Models\RefreshToken", "id": <OPSSEC>, "relations": [ ], "connection": "mysql", "collectionClass": null }, "esi": { "instance": { "authentication": null, "fetcher": null, "accesschecker": null, "querystring": [ ], "requestbody": [ ], "version": "/latest", "cachable_verb": [ "get" ] } } } Tags Collapse [ "character", "asset", <OPSSEC> ]"

lol, my instance must have got scard of me talking to you, when I went to bed last night it wasn't working. Right after, I sent that output just now...It's working lol

Thanks!

As a rule of thumb, it can take up to an hour for data of a new character to appear, and in case of a new install, some things like market data can take up to a day (although it can be faster)

Yeah, it must have been trying to get all the data and was stuck doing so before it could get to my characters.

Is there a way to export CORP assets as an excel?

Not that I know. We accept improvements if you want to code it though.

Ok, I am mainly using this tool for CORP asset management and I have a couple of ideas that’ll help!

Is it normal for SeAT to be a bit slow with populating data?

what kind of data?

everything, anything. ESI is set for all....but only corp name and sec status has been populating.

how long uptime do you have on SeAT now?

honestly about an hour

...this time.

meh...looks like my SeAT worker crashed.

blade or docker install? on what OS?

ubuntu 22.04 bare metal.

ya the worke was dying. I set it to auto restart. Its pulling more....

this thing is frustrating lol

so why is it dying?

.docs.troubleshooting

@user_131158012831203328, https://eveseat.github.io/docs/troubleshooting/

I just tossed in a restart command line

but thank you for this

Do toy have it running through supervisor? What are supervisors logs for the process?

Ugh...

Did you follow the install guide?

Is there a way to get the discord connector to not strip a certain role if its give out maually?

Its not great, but putting that role above the bots role will stop it modifying that role

Another option is to manually hand out that role using a hidden squad

Well so I have a different bot who is handing out a certain role so I guess I can just put the role above it

Unless, is there an easy way to have the bot give the squad in seat?

SeAT has an api you can use

Forgot to thank you earlier for the update. Works smoothly!! So thank you!!

hi all I continue to have problems with the seat 🙁 now there are many such errors

Please tell me what I can try to do

maybe these screenshots will give you some clues

You can turn on debug logging, then have a look through the laravel logs.

APP_DEBUG=false this in .env ?

i do it. now it seems like it's worth waiting a few hours

No it should be a line like LOG_LEVEL=debug

@user_164076488294006785 having trouble configuring the reminders for SeAT calendar, I make an operation and it makes the event in discord all good but at my intervals for reminders dont seem to come through, even if I make a webhook and tie it to the tag I assign to the operation, nothing, any ideas?

Scratch that, im a complete tool

join the club, we're all missing 10mm sockets

ok

more errors like this appeared, probably because I deleted several dozen users and characters from Seat?

Can you send me your laravel log please

send in PM

One thing I note is you dont have the maintainence command in your schedule seat:admin:maintenance, I suggest running that manually now, and in the schedule daily

You also have a lot of bad gateway exceptions, does your server have a good internet connection? And where geographically is it located? Some countries in particular do not perform well

Also eve:update:status and esi:update:status should be run every minute.

It may also be worth resetting your queue and caches with the command php artisan se:ca:cl

What was the Issue ? I sometimes have problems with those reminders as well but never seem to find what is the error because it works for me but not for other people

It was me not configuring the notification group correctly....big dumb dumb

Am I dumb or does the connector api path not work?

Im getting a 404

Ok, I’ll try it. The server is at Hetzner, in a data center in Germany. The connection speed there is around 1 gigabit. thx

I did everything from the recommendations above, but I still see many such errors, as in the screenshot. Tonight I’ll try to migrate the VM to another Hetzner data center.

I moved the server to another Hetzner data center, but a lot of these warnings still remain, and the information about character chats also doesn’t fully update.

identify the peculiarity delete the character whose information is not updated reboot the chair, and during the first 1-2 hours register the character so that in the future the information on it is updated normally and if you register after the server has been working for 12+ hours, then only part of the information about the character is updated 🙂 Strange behavior

That does seem odd... The errors about not being able to reach the eve servers though is weird and does imply some form of network problem..

Today, there have been almost no messages about the EVE server being unreachable. However, there are plenty of warnings that the ESI route is unavailable—even though when I check the ESI endpoint from inside the container, I get a normal (healthy) response. I’m still stumped and not sure what more I can do. ☹ Both the warnings and the container test outputs are shown in the screenshot.

For example, when I try to update the corporation history via the interface, I get the error shown on the screenshot — yet a direct request from inside the container returns the data just fine.

I reached out to ChatGPT for help—sent it various warnings and logs for a while. In the end, it suggested adding these lines to the .env file: QUEUEBALANCINGMODE=auto QUEUE_WORKERS=10 And it seems to have helped 🙂 I'm now monitoring the situation.

hmm oof

This makes sense if you are using a Hetzer VPS on a shared hardware. Most data centers overload their hardware for VPS. If in the middle of handling a request, the CPU cycles were taken away from your VPS, I could see this kind of stuff happening. I use to have the similar stuff happen to me on my VPN VPS

Thanks for the information, I’ll try other options in the future — maybe it really will change the situation.

Also, not sure if it applies to your config but some places allow you to assign extra processes to the networking. I was reading that if the network processes mismatch the CPU processes, it can have network drops; due to the imbalance. Not sure if it applies to Hertzner though

yeah... hypervisors let you do that

but not all cloud providers

And yet, I celebrated too early. As for the characters, the information does get updated in time. Some time after restarting SeAT, the data refreshes normally, but after a while I see in the logs that it keeps trying to send requests to the server too frequently, and because of that the job gets dropped. After some more time, errors start appearing, showing that the ESI server is unavailable. But if I run an access command to ESI from inside the container, it processes normally. Apparently, my next step will be either moving to a bare-metal server or switching to another provider different from Hetzner.

Please don't run seat on trash vps

he worked for VPS in Hetzner for 5 years 🙂 and now he encountered such a problem, but in general I understood yes I have about 1100 tokens, if I use a dedicated server, what configuration would you recommend?

and is it possible to leave it in Hetzner?

you can find deals on their server auctions

how big is the VPS btw?

8 cores, 16 gigabytes of RAM, but most of the time the processor is not loaded more than 50%, and up to 2 gigabytes of RAM were used

I think i7-7700 will be enough

that should be enough...

can you change seat:buckets:update to every 10 minutes instead of every two minutes?

*/10 * * * *

is the cron for that

earlier, on advice in this chat, I already changed it to every 20 minutes and now it's set to /20 *

that's every hour

and it still has issues?

that's very weird

yes, the problem remains and yes it is very strange 🙂

I also see that these jobs often go through quite well, is this normal? Notification, Rescent and Contracts

for each character once every 3 minutes

a few more oddities tried disabling all plugins again, errors disappeared 🙂 before that disabling plugins did not get rid of errors I'm still observing 🙂 at the same time, those characters for which the information was not initially updated and is now not fully updated for example, Wallet, assets, skills how often should this data be updated? And for new characters, the update occurs within a few minutes.

Being spammed with this failed job, anyway to fix it?

This helped me https://discord.com/channels/821361165791133716/1345031288226054245/1345378583333568634

I always see then when a new token was added to seat...

Give it 24 hours..

I run it on one of these https://www.hetzner.com/sb/#search=Intel+Core+i5-12500

with about 3x your tokens

er maybe 5x

along with the rest of the alliance services

going from vps to dedicated box was a massive boost in performance

heres a list I made of all the cpus on the auction btw and their gen/release date

given these are being rotated out of their production farm sorta gives you an idea of how old the gear is

yeah, SeAT is hard on systems 😄

It is all a question of scale. My corp is rather small (70 tokens), we run perfectly fine on the cheapest hetzner instance.

yeah

thing is...

small installs are just fine 😛

Running on a NUC 😂

a NUC with even a moderately shitty SSD will outperform a VPS by a lot 😛

Heheh true that

@user_301981661761896449 I just submitted 4 pull requests on seat-srp to fix various bugs and some enhancements

surprisingly disabling plugins really helped, of all the ones, only the seat-connector with warlof/seat-teamspeak and warlof/seat-discord-connector was critical to me I turned it on, no problems so far it looks like either cryocaustik/seat-hr or recursivetree/seat-rat were to blame oh, but there were no problems associated with them in the logs

I'm pretty sure it's not seat-rat. That plugin doesn't touch anything esi job related at all

I'll experiment with them over the weekend, I don't have time for that now 🙂

Break it hard by setting

env
# App Environment
APP_ENV=production

@user_614098468218339348 it looks like your pull request for the corp info issue is throwing errors for new accounts joining seat: Seat\Eveapi\Jobs\AbstractCorporationJob::construct(): Argument

1 ($corporation_id) must be of type int, null given, called in /var/www/seat/vendor/eveseat/eveapi/src/Bus/Character.php on line 116 {"exception":"[object] (TypeError(code: 0): Seat\\Eveapi\\Jobs\\AbstractCorporationJob::

construct(): Argument

1 ($corporation_id) must be of type int, null given, called in /var/www/seat/vendor/eveseat/eveapi/src/Bus/Character.php on line 116 at /var/www/seat/vendor/eveseat/eveapi/src/Jobs/AbstractCorporationJob.php:45)

https://github.com/eveseat/eveapi/pull/442 it looks like this fixes it for my instance

That is interesting because it implies the character has an affiliation model, but the corporation id is null. Which is really weird because corporationid on characteraffiliations is not nullable

it seems to only error on a new user+character joining. its like the job runs before corp id isnt null

maybe a race condition?

i made that change from my pull request and it worked for my new guy

maybe more investigation needed

Yes your PR should fix it. But I still don't get why it happens

https://tenor.com/view/imagination-spongebob-squarepants-dreams-magic-gif-12725683

nvm, I know why. The "affiliation" relation of CharacterInfo returns a default: https://github.com/eveseat/eveapi/blob/b85b684259d153fd90df27f341b1a07810fd9d6d/src/Models/Character/CharacterInfo.php#L146

so the affiliation is not null if it doesn't exist, but it doesn't contain data either

I really hate defaults on relations. This is probably the 10th time they got me like this

https://tenor.com/view/teamwork-dream-work-colleagues-coworkers-gif-3325381463059285352

hi guys is it normal ? nothing charge ?

my bad his took many time... 🥲

this keeps happening to one specific character - the main associated with this toon and this toon have both left the corp and I "deactivated" their "user" profile.

is there a way to solve for this? stack trace attached.

The latest update should fix it. Have you installed it yet?

oh no ...not yet. Will do it right away.

updated it last evening and let it run for a bit....still got this error today. and it is only for this one toon.

Greetings everyone, i am looking for assistance with squad filters. I have a few auto squads that sort people based on game titles. However we have found that a new member corp is using html formated titles for colors. We asked a pilot to add a titles with no html expression to be used for the squad filtering, but SeAT had not picked up any of their titles at all. Does SeAT not look at titles if any of them are html formated? How can we work around this issue? Any pointers welcome, thank you ^^

are alliance contacts populating for anyone else?

5

Then you need to use the commands for seat 5 not 4

Click where it says seat 5

there isn't one

https://eveseat.github.io/docs/admin_guides/docker_admin/

for tail there is

but not in the installing plugins section

just says save and run docker-compose up -d

Well that's on us and should change that. But it's all docker compose now. Adapt the command as you can see the difference between 4 and 5 elsewhere

yea

sooo

what do I do now then lol

why is it throwing 404

You need to run the command properly

docker compose up -d

Once the container boots (which can take a little bit) traefik should pick it up

If it doesn't then you can try doing the full stack as per here https://eveseat.github.io/docs/admin_guides/docker_admin/#configuration-changes

That one worked

restarted the stack

Ahhh

yeah

traefik wasn't starting

ty ❤️

trying to install "https://github.com/zenobio93/seat-discord-connector"

Firstly, run the following command composer require warlof/seat-discord-connector

tells me not to run it as root tho..

you have a docker install. therefore should never be using composer directly

mk

just add warlof/seat-discord-connector to the env file as a plugin

ah ok

then use the same command you used above to bring the sttack up and apply the change

ty

guys, where did i find the page on the documentation, to update the server regarding due the recent esi changes ?

no docker image

nevermind got it thx

Hi, I have the cryptatech/seat-srp module installed however am getting issues when trying to submit SRP requests as the killmail url is different to what the placeholder highlights and even if I change the start of the url to match it is still invalid, im assuming this is due to ESI updates or am I using the wrong SRP addon?

@user_301981661761896449

can confirm not working here either

tried both zKill and Ingame methods

another question.. why am I failing jobs?

Am I trying to pull to often or something?

It is normal that some jobs fail. What is your ratio of failed jobs vs successful jobs?

I'm at work now but it's a lot more successful vs failed

Only 5 failed

Like 80+ successful

SeAT Calander plugin only pings to one discord channel while when i test the notifications on the other channel they seem to work. anyone know what i should do?

unless data fails to populate, that seems fairly normal

Sounds good

its working for me, can you confirm you are running seat-srp version 5.0.6?

here's your exact link as an example

it looks like seat-srp returns "invalid killmail address" if getKillMailQuote fails. check your system logs after submitting a killmail and report back

the problem isnt the url, i suspect you're having issues elsewhere getting killmail info

lel

Yea nope, it ain't workin

Go to settings -> system logs and see what errors you’re getting

There is no error corresponding with the times i've tried to run it

my local time rn is 8am

If you go to a character and look at their killlog, does the kill log load properly?

Yes all the killmails load correctly on character profiles

Is it just that killmail that errors or other links as well?

I've put 3-4 other killmail links into it and none of them work

Do you have any failed jobs related to killmails?

heres the top of my failed jobs list

Can you dm me the killmail link and I will take a look

Its not just a select killmail link, it's all of them aren't working but I'll dm you an example in a sec

Click into those “Killmail” failed jobs. What’s the error?

Click show all and share a bit more

Just to clarify that is not the SRP killmail job, it's actually a notification.

But yes I also need to improve the error reporting

Yes ik, @user_224899784257961984 was wanting to look there incase there was something

this is the one that im getting

Ah yeah I didn’t realize it was Discord\Killmail

Yeah... But it's a crap error.. Because that's what you get if absolutely anything goes wrong

One thing you can try. Open the Web browser dev tools, and inspect the post request response you get when clicking submit.

will try now

dm'd you the api call

302 response

Will look into it tomorrow

ty, would love to get some good use out of this plugin 🙂

can you dm me too plz, i can also help, ive done some patching of seat-srp if crypta gets too busy

@user_259785950119460886 in the srp settings, 1) do you have a pricing provider created and selected? 2) are you using "SRP method" simple or advanced?

do you have a default rule specified

so tldr, the error you're getting is because the request to get the killmail quote fails (https://github.com/eveseat-plugins/seat-srp/blob/master/src/resources/views/request.blade.php#L408). if you are saying it is getting a 302 in that request, then that means its probably hitting this: https://github.com/eveseat-plugins/seat-srp/blob/master/src/Http/Controllers/SrpController.php#L79 which means its failing to get the price of the kill. tracking back getKillmailDetails(), the most obvious offender would be either the pricing provider (maybe your janice api isnt working) or your SRP rules not being able to calculate a price

step 1 is to make sure you have a default rule specified (maybe even re-fill it out and save). step 2 is maybe try a new pricing provider, for example use:

then we can narrow down the specific issue

How easy would it be to setup traefik as its own docker container and connect SeAT and other services to it to allow for more than one service to run on my server

Trivial as per what you were told. Just need to extract traefik out and keep one traefik then add a file provider or docker labels and go from there.

Trivial.

I just upgraded my seat server with

sudo docker compose -f
docker-compose.yml -f
docker-compose.mariadb.yml -f
docker-compose.traefik.yml pull

After this my server shows this message

seat-docker-front-1 | https://repo.packagist.org could not be fully loaded (The "https://repo.packagist.org/p2/cryocaustik/seat-hr.json" file could not be downloadeP/2 502 )), package information was loaded from the local cache and may be out of date
seat-docker-front-1 |
seat-docker-front-1 | In CurlDownloader.php line 664:
seat-docker-front-1 |
seat-docker-front-1 |   The "https://repo.packagist.org/p2/cryocaustik/seat-hr.json" file could not
seat-docker-front-1 |    be downloaded (HTTP/2 502 )
seat-docker-front-1 |
seat-docker-front-1 |

Sooo I think I messed up and I don't wanna start over...

I installed SeAT and then logged in with admin 1st and then logged in with my character so now I think its using the admin account for my character?

any ideas?

nvm got it

or not..

welp I restarted.. and I was right...

I logged in as admin 1st and it made me the

1 user

I wasn't showing up in the users list

SRP works btw

If one of my users biomassed an alt and now the alt is flagging as no valid token hence discord connector has stripped their roles, how can I delete just that character?

Is the only option to delete the whole user and relink their alts?

Did they biomass their main? Or just an alt?

It should just invalidate the esi for that one alt. If it's their main, that's a different story

Just an alt but we run our discord connector on strict so if one alt is invalid they get removed

you should be able to delete that particular character?

where and how? haha

Does that answer your question?

Would you be able to share how you fixed this? I have run up a new install of Seat (bare metal). All fine. Installed seat-hr and getting the same error.

I will have to look through my notes. I don't remember off hand.

I got the 500 server error specifically on an Ajax request. In my seat was missing the log directory. I was trying to check the logs, but couldn't find them. After fixing that I discovered that I was missing a table and need to manually migrate it from the vendor.

the missing table was seathrcorporations

sudo -u www-data php /var/www/seat/artisan migrate --path=vendor/cryocaustik/seat-hr/database/migrations

then verify

sudo mysql -e "SHOW TABLES LIKE 'seathr%'" -D seat

You should see "seathrcorporations" in the output

If that fails or isn't created there are a few other options.

basically migrate again or reinstall the plugin

then...

sudo -u www-data php /var/www/seat/artisan config:clear sudo -u www-data php /var/www/seat/artisan cache:clear sudo -u www-data php /var/www/seat/artisan view:clear

to clear caches and retest

My thought is that php artisan migrate wasn't run for what ever reason.

Thanks so much mate. I will investigate

Hi all I ran into a really strange issue again: when registering in TS, the character enters the desired nickname, clicks confirm, and then gets a response saying it couldn’t find them on the server — and it shows a completely different nickname. I have no idea why this is happening 🙂 In the logs I can see that, yes, it indeed failed to find the correct nickname. I can’t figure out what else to check or what to try next 🙂

This issue only happens with one specific user.

I saw this in the logs: indeed, the connector with TS doesn’t have these IDs — my list ends at 710. I wonder where they could have come from, and whether that’s why something went wron

okay... don't think I've seen this one before

`PDOException: SQLSTATE[23000]: Integrity constraint violation: 1452 Cannot add or update a child row: a foreign key constraint fails (`seat`.`corporation_titles`, CONSTRAINT `corporation_titles_corporation_id_foreign` FOREIGN KEY (`corporation_id`) REFERENCES `corporation_infos` (`corporation_id`) ON DELETE CASCADE ON UPDATE) in /var/www/seat/vendor/laravel/framework/src/Illuminate/Database/MySqlConnection.php:45

Is it possible to add custom link with main character id in it? like

 https://example.com/id={user_id}

btw guys can you update plugins selection in documetation? https://eveseat.github.io/docs/community_packages/ replace warlof/seat-discord-connector with : https://github.com/zenobio93/seat-discord-connector as it is updated...

are skyhook notifications working? every other alert seems to work, but just recently had a skyhook get reinforced and got no alert for it. every other notification seems to work.

What do you need this for?

Do I understand correctly that you saw some new error that you hadn't encountered before and simply reported it to the chat?)

Basically

i have my own website where i tax my corp mmebers for moon mining, it would be easy that way to send them directry to their profile on my second website instead giving them list of all ursers and let them go inside other members profiles that way.

im now more into use the discord connector feature where i ll grab discord id from mysql and then use it with my multi bot to pm users., but still link inside SeaT would be helping :S

Hi guys. Anyone got the solution for this problem?

so re: seat-wanderer-access-sync when I run docker compose exec front sh -c 'php artisan wanderer:sync' it just hangs forever and never completes, and I dont seem to get getting any users added to access lists

that sounds weird. wanderer:sync is only supposed to add a job to the queue

it did

but turns out

I just wasnt patient enough

after ~20 minutes the command just finished

Hmm. It shouldn't take that long to do it though...

have a lot of users :\

Yes but just adding the job to the queue is independent of how many users you have. The runtime of the job, perhaps (especially when it runs the first time)

gotcha

re-running it is very fast

well

15secs give or take

Question; is it possible to just have a login button with publicData on the main landing page, and have another one with more scopes tucked away somewhere else?

Settings - - - > single sign-on Create two or more profiles. Add buttons in customise login page section Example

Full SSO Sign-on (Members & Recruits)[[MembersFull]]

Public SSO Sign-on (Friends & Guests Only)[[Guests]]

I figured that part out yes, but that would have both logins appear on the main login screen. We dont want to ask too many scopes from our 'regular' members, but Id like to offer a login with more scopes for those that want the added functionality. Id like that login button to be somewhere else than the main login screen to prevent clutter and possible confusion

@user_614098468218339348 is this perhaps something that can be done on your seat-info module ?

I mean you can copy the link that starts the authentication process from the login page and add it to an article. But to see the seat-info page, they already need to be registered on seat

This worked. Copied the link from the login page, added an article with the login with eve image tied to that that link and deleted that login from the main login page

created seat instance from docker (with help) and got security cert working, cert has expired and the person who helped me is away from game. can anyone help me get my server secure again?

security cert = TLS certificate? That should be automated on the standard seat docker install. Perhaps check the traefik logs

im not very good with linux

if your using default it should auto renew on its own

Being dense - but help me out... does someone have a complete custom-layout.css? Wanting to play with the colours and style, but don't have anything already complete to play with.

Will an attacked Customs office not trigger the Structures Attacked ping? Looking now, does it have to be "attacked Customs Office" and for structures to test I would need to smack my astra?

Give it a love tap

Previously, I was advised to add the following entry to the scheduler: seat:admin:maintenance 0 0 * Now I’m getting a lot of warnings that the connector cannot match a character with the corresponding data in the connector. How can I make sure that the connector data is also cleaned up? Over the last 5+ years, a huge number of such users have accumulated in the connector, and it’s likely this is causing the DriverApplyPolicies job to fail. Queue: high | Attempts: 2 | Tags: connector, teamspeak

hi all )

I’m using the Docker version.

All ports on the TeamSpeak side are definitely open. After clearing the cache and restarting SeAT, the connector assigns permissions in TeamSpeak correctly, but after a few days it stops doing so.

hey there, i have problems with the discord connetor setup. It seems that it is a role problem. I have an alliance Role, that should affect all Members in the alliance. But this does not work correctly. How can i set this up correctly ?

it works for my corporation yes, but it seems the other corporations within the alliance does not affected

I have a newly installed instance of SeAT and I get a lot of "Authentication data invalid/empty". All Tokens are fresh and green.

Is there any place where we can see some errors? Maybe the bot’s role isn’t high enough in the Discord hierarchy? Like — the alliance role is higher than the bot’s, and the bot’s role is higher than your corporation’s role?

Is this happening when you try to add a user? Is the callback URL definitely correct?

The Callback Url is working because I can logon

Exception Seat\Eseye\Exceptions\InvalidContainerDataException: Authentication data invalid/empty in /var/www/seat/vendor/eveseat/eseye/src/Eseye.php:208 Stack trace:

0 /var/www/seat/vendor/eveseat/eveapi/src/Services/EseyeClient.php(109): Seat\Eseye\Eseye->setAuthentication()

1 /var/www/seat/vendor/eveseat/eveapi/src/InteractsWithToken.php(41): Seat\Eveapi\Services\EseyeClient->setAuthentication()

2 /var/www/seat/vendor/eveseat/eveapi/src/Jobs/EsiBase.php(310): Seat\Eveapi\Jobs\EsiBase->configureTokenForEsiClient()

3 /var/www/seat/vendor/eveseat/eveapi/src/Jobs/Killmails/Character/Recent.php(93): Seat\Eveapi\Jobs\EsiBase->retrieve()

Logon / Adding new toons is working. It's the jobs that fail

Is this a fresh installation, and the database was restored from a backup?

No, all new

good afternoon everyone. I have a new seat install that I am trying to get going. After the SSO I get a 500 Server error. and in the logs, it appears I am getting a "401 unauthorized" from the client...hopefully i got the log attached properly for ya'll

also, please bear with a bit if i ask silly questions. I had a friend who is a sys admin standup the vm on my hardware, and my linux command line is rusty at best.

have you completed this? https://eveseat.github.io/docs/configuration/esi_configuration/

yes, i have completed this and verified all the info is correct in the env file

Hi guys o/ can seat be run with a postgres db instead of mariadb ?

It's a goal for the future. But at the moment not quite.

There's still a couple of places it won't work due to raw sql etc

Ah too bad. My mariadb has some stability issues while my postgres is running fine, so I was hoping to migrate everything on pg

The other aspect I've not sorted out is how to migrate a mariadb install to pg, for when it is compatible

I've found quite a few articles about that but have not tested it myself

Yeah I've found a few methods, but not one that worked seamlessly for me. There was always some weirdness

Good evening folks, I am still having the same problem as I had over the weekend. My admin login link from the server works, I have verified all my port forwarding rules in the firewall, and verified all info from the EvE dev app site is correct in my .env file. My issue is that after the SSO login, the characters recieve a 500 server error, and in the logs, I am seeing a 401 unatuhorized. For the life of me I have not been able to figure out why.

Updating to add: I tried again tonight and now have an eseye log as well indicating 502 bad gateway.... contents as follows: Error 2025-10-06 11:11:05 local [http 502, bad gateway] get -> https://esi.evetech.net/v1/status/?datasource=tranquility [t/e: 0.17s/99] Log index 4. Click the button to copy link to this log entry. 4 [http 502, bad gateway] get -> https://esi.evetech.net/v1/status/?datasource=tranquility [t/e: 0.17s/99]

Is this docker or manual?

bare metal, sortof...in a fresh built VM

and after more searching, i may have come across the issue with the callback not reflecting the correct file structure...im working on that now to retest, and report back with my results

Update. finally got what I thought was wrong with my callback address sorted out. Now the SSO pilots get a 404 error instead of a 500, and the log is still generating the 401 unauthorized

Hmm

Post your vhost set up?

Running a vm, fresh install, stood up specifially to run seat, sits behind a reverse proxy vm and a firewall. all the port forwarding is correct. I used specs from seat documentation to setup the server with resources to manage 500 characters, with the intent of only having roughly 50 for now. If you need more specific info, please feel free to ask

For the manual install you need a Web server. We have docs for nginx. Can you share your config file for that

Also whatever reverse proxy you have what it's config is

i will work on getting all that info. the initial setup was followed by my friend who is a sys admin...so some info i have to get from him

my command line is rusty, if you point me towards a file though, i can do my best to get a screenshot at the least of it

server { listen 80; listen [::]:80;

If you are hosting this instance on a domain, set that

name here.

server_name seat.yourdomain.com;

SeAT public directory. This is the only directory that

should be exposed by the webserver. If one has to expose

the parent directory then things like the .env file will

be available for anyone to download.

root /sites/seat/public; index index.php; location / { tryfiles $uri $uri/ /index.php?$querystring; }

PHP-FPM configuration.

location ~ \.php$ { tryfiles $uri /index.php =404; fastcgipass unix:/run/php/php-fpm.sock; fastcgiparam SCRIPTFILENAME $documentroot$fastcgiscriptname; include fastcgiparams; }

Even though .htaccess rules mean nothing in the nginx

world, prevent those from being downloaded anyways.

location ~ /\.ht { deny all; }

In case someone messes up, prevent .env files from

being downloaded as well.

location ~ /\.env { deny all; } }

as i look at this, im starting to realize maybe a critical step was missed here

but i may be wrong, so if someone can guide me either way, thats fantastic

I have run up bare metal seat server successfully - a number of times - in the past few weeks on Ubuntu using mariadb, nginx and traefik. I may be able to help if you are using this or similar config.

If you are running bare metal, you need to follow the guide, or be competent enough to know where you can deviate...

I had my friend who is a sys admin setup the initial vm and everything for me. He followed the documentation and only deviated in a few places. However, he just had a baby and is currently unavailable to help me diagnose what going wrong here. The server is installed on Debian 6.12.43-1, nginx is in use, I am unsure of traefik but have reached out to him to confirm. I will update with more info as I get it, or insights as I play around trying to sort this out. Edit to add: I have confirmed that Traefik was not used, but nginx proxy manager took its place.

a question i have...am i missing a line here redirecting to my domain, or is that not an issue cause the dns redirects to this VM internally

If you are hosting this instance on a domain, set that

name here.

server_name seat.yourdomain.com;

In my experience the server name is only used for internal redirects and is not super critical. To me it sounds like either an issue with the app registration or with the app id / secret stored in the seat config file. This plus the ability for the seat instance being accessible externally I believe are the main moving parts for SSO

The redirect is on the app registration side and as also stored in seat config file

The only real things I change in the seat .env file are App URL - external DNS name DB Password - set earlier EVE Client ID - from the app registration EVE Client secret - app registration EVE Callback URL - external DNS name/auth/eve/callback - also set in app registration

Note - i use traefik purely to get a free SSL certificate so I can utilise SSL. I imagine you are just using HTTP

My advice is probably check or redo the app registration - make sure callback url is correct in both sides, copy the ID and secret into your .env file, make sure it is all externally accessible. Reboot

Hello 🙂 My corp titles aren't updating. I only see the ones my character has assigned to them. How do I pull through the full corp titles?

So I swear everything is correct to my eyes, it all matches up on both ends. However, because my buddy installed seat to /sites/seat, do i need to adjust my callaback for that. I saw the note in the install documentation and am a little confused. If i place /sites/seat in the callback, I just get a 404 from the sso and the internal admin interface notes it as a bad callback and suggests something else.

If the documentation was followed this won't happen. Seat is the default website so you dont need sites/seat in the url

Callback url should be as described above

ok, I am creating a new app in the dev website and gonna updatethe server with that info...at this point im at a total loss. Wondering if maybe the most recent debian release is the problem with this build

I went over everything again, and now I truly feel like an idiot. After checking 50 times, there was in fact an error in the .env secret for the dev app....I appreciate everyones help with this, and please forgive me idiocy

OK glad my advice helped 😅

learning by making mistakes 🙂

now i get to learn more by installing plugins and seeing what i break..lololol

but hey, my command line is polished up, and ya'lls advice got me to me problem, even if it was something small and stupid....its always the small stupid stuff that gets me

A word of warning i have blown up 1 or 2 installs with plug ins 😂

oh good, corssing my fingers, my corp has a list of 6 we want, so heres hoping i wont be back asking for help with what i broke

make a dev instance...

Sorry to bump... Hoping this is another easy one?

1. Do the relevant jobs get scheduled 2. If yes, do they succeed? 3. If they fail, what is the error

4. Are the scopes correct

5. I‘m not sure if it is one of the director only endpoints, but if yes, do you have the ceo on seat

1. Which are the relevant scheduled jobs?

When i try to add esi:update:corporations

instead of trying to schedule new job find where is the issue...

you have steps in here - https://discord.com/channels/821361165791133716/821361546608508938/1426583375732281364

this is not how u run this job

check logs for issues, if you dont have any logs from this job check if its even running, if not check if something else is not updating, check sso as well.

if you will run this job without fixing your issue it might fix the issue for now but the issue will still be present in your seat.

I don't see any errors there, but honestly this is all new to me - apologies. I'm going to have the corp director sync and see if that's what was required to pull through all the roles

you need to have CEO or full director for most of the corporation stuff related

understood, thank you @user_405130542674083842

ok last one - I promise. How do I safely reduce the scope of the ESI? I installed with all scopes ticked, and have now removed all the scope items I thought I didn't need but I'm now getting errors thrown when I try to login {"error":"invalidscope","errordescription":"The requested 'esi-assets.read_assets.v1' scope is not valid."}

Ideally I'd like to get down to: publicData esi-mail.organizemail.v1 esi-mail.readmail.v1 esi-mail.sendmail.v1 esi-skills.readskills.v1 esi-skills.readskillqueue.v1 esi-corporations.readcorporationmembership.v1 esi-characters.readloyalty.v1 esi-characters.readmedals.v1 esi-characters.readstandings.v1 esi-characters.readcorporationroles.v1 esi-corporations.trackmembers.v1 esi-corporations.readdivisions.v1 esi-corporations.readtitles.v1 esi-corporations.readstandings.v1 esi-corporations.readmedals.v1 esi-characters.readtitles.v1 esi-alliances.readcontacts.v1 esi-characters.readfwstats.v1 esi-corporations.readfw_stats.v1

no worries, you can ask as much as you like. It just that doing support is not really fun and we're all doing this as a hobby. Sometimes we might not be as responsive or a bit harsh because of that, especially if it is something basic, well-known under the experts, or if it requires a lot of effort from our side (be it typing a long guide or looking up the name of a job). It is not always a you-issue. You have to set the scopes in the ESI application on the eve developer portal. Then you also have to set it under the SSO settings in seat. Have you done both?

The job is called Titles. I just confirmed it, it can only run for directors of the corp. You can see job on the horizon dashboard (truck icon in the top right on admin accounts).

understood yeah, it's very much appreciated!! I had done it in the ESI application, but not in the SSO settings. Hopefully I can figure it out from there - thank you @user_614098468218339348 ❤️

it worked perfectly, thank you!

I have an issue with characters being sync'd a little weird. They show as having a non-player corp, but in their Employment History it shows the corp correctly. It wouldn't be a problem, but I'm using the corp to set Discord titles using seat-discord-connector

What are the logs from this char? any failed jobs? Is his esi token valid? With what scopes he logged into your seat?

u could run php artisan esi:update:corporations [CHAR ID] for a quick fix but it wont solve your issue.

Apologies, how do i see the logs from this char? I see no failed jobs in horizon esi token is fresh, he just auth'd.

it's to be noted, I don't have esi:update:corporations in my schedule anywhere, and I can't add it

it's happened by itself in the last minute, annoyingly.... it seems my problem is just with the discord bot not adding the roles

That job is being done by one of the containers. Scopes looks OK. Just run this command to fix it for now but you probably have errors in logs from this character. Check esi tokens on it if they are OK and ask this guy to login to seat again. Sometimes that solves this issue. I had it random like 2-3 times on different chars in my seat.

Then check discord connector errors and check if the bot can manage that role that u want to be assigned.

appreciate the support Matt! It seems the bot can now remove roles but not add them. Nothing in logs, the bot's role is higher than all the roles it needs to manage. I have created a new bot, tried it in another discord, removed the module then reinstalled it... nothing works I did notice that everytime I uninstalled the discord module (restarted the docker images) then reinstalled it - it already had the bot details in there... I do wonder if my issue is somewhere I can't see. Would be good to know how to completely wipe out that module and all of its saved config so I can start again.

thats because u remove the plugin but not its tables.

so when u install bot again it already had all data to work with

check your schedule and see if u have seat-connector:sync:sets and seat-connector:apply:policies

also go to connector itself and try those:

see if that will work

if not check logs

u might have some errors in connector logs so check those if in laravel logs u dont see anything

check if u see something from Warlof\Seat\Connector...

thanks @user_405130542674083842 will look into this! That said I think I found the issue with the connector. If there are any Discord sets (roles) that it doesn't have permission to add/remove, then it just doesn't do anything for that user. I wanted this to be the case by design (for another bot to assign a different role). Looking into other ways to do it now instead

thank you so much for your support

I'm a little confused and not sure if i do something wrong. Goal: Create a Recruiter Role which has access to User/Characters that are applying to a corp Where I'm stuck: As soon as I give that role " Grant access to Character Sheet" access it can see every character within SeAT. I've tried the hr plugin but it seems like it needs the same permission so that the recruiters got access to all characters again. What am I missing?

So the manual way to manage it, is in the permission you can set a filter so it only applies to the filters you set

Saw that, but filter applies only on character base, im now lookint into squad sync which can be an idea i guess

The way I would recommend to automate it, is to have a look at the seat-squad-sync package (plug as its mine)

ah ok 🙂

still confused why seat doesn't have this kinda build in

Your plugin works nice, thanks for that! Quick question, would it be possible to add everyone to that group if his "main" character is not part of corp a/b or alliance?

Seat doesn't have a distinction between main char and alts, so not really I am afraid

Hmm, so there seems no way for a autogroup/squad since that would force users with neutral alts into that squad also. is there a way and idea to skip the step that a user needs to join the squad?

Hello! I have just finished self hosting my first seat. Has ben like one hour since i have login with my characters for test. I have my main set as admin. But i cannot see any data, info about any of them... like assets, skills, wallet and stuff. Is there a issue or the server need to pull that data?

Have you set the scopes required for that?

Yes. I found the issue now.. i didint start the sync service

Im having a issue where SEAT says it has 3million+ notifications to read thru, then fails each "job" and generates errors for each one, fills my hdd and crashes my server with error logs

https://i.imgur.com/fSrSB7E.png

https://i.imgur.com/4omOoos.png

https://i.imgur.com/WdoCoCf.png

https://i.imgur.com/kNxmK53.png

yesterdays errorlog was 4,2gb, so far in 20mins this one has 66k+ of the same error

seat was working fine until updated to the latest version, this all happened after that

hmmm... wtf:

PDOException: SQLSTATE[22007]: Invalid datetime format: 1292 Incorrect datetime value: '2016-03-27 02:29:00' for column `seat`.`corporation_medals`.`created_at` at row 1 in /var/www/seat/vendor/laravel/framework/src/Illuminate/Database/MySqlConnection.php:45

Any one here running seat on synology nas with cloudflare tunnel?

Do the other jobs succeed?

some jobs do succeed yes

That looks like a correct datetime format? Or at least I could insert it into my testing DB?

i took my seat down to clear the horizon log: https://i.imgur.com/eeQNQWc.png

gonna start it up and see whats up now

when they fail, what is the runtime on the jobs?

between 0.01s and 0.05s

as a mitigation, you could probably remove esi:update:notifications from the schedule. But you'll have to remove it after every restart of seat

its completing jobs: https://i.imgur.com/tGnIemi.png

but still failing the notification ones: https://i.imgur.com/5KomuiT.png

Is your server located in the US or Europe?

US

ok, because then the notification jobs fail before reaching out to ESI servers looking at the timing

In that case, could you 1. set the logging level to debug in the .env 2. clear the log file 3. let a notification job fail 4. revert the logging level to the old value 5. send me the logs Running on debug log for a long time can generate gigabytes of logs, so make sure to turn it of again once you need it no longer.

I’d also like to check if these characters’ tokens are valid.

Any idea why Discord connector DriverApplyPolicies job just hangs forever in pending jobs and doesn't actually do anything?

All other jobs are running.

nvm fixed with override

i put on debug mode, it stopped generating as many errors, so i let it run all jobs and now its finally finished with em, looks to be working fine now lol

as vladimir_vladykin suggested i looked for invalid tokes and deleted those, may have been a part of the issue

https://i.imgur.com/pavGGkk.png

https://i.imgur.com/A4IP3RJ.png

i dont really understand what caused this whole thing, but hopefully its now over lol

I have set SEATDOMAIN=seat.maculaalliance.eu And also: EVECLIENTID=test EVECLIENTSECRET=test EVECALLBACK_URL=http://test.eu/auth/eve/callback But im not shure if its ok

it should be ``EVECALLBACKURL=${APP_URL}/auth/eve/callback``

Ok, so that it will take the info from SEAT_DOMAIN=seat.maculaalliance.eu

yes

that is correct

Thank yoU!

Because there is another command in the ENV will get your domain ``Seatdomain=test.test.com`` to ``APPURL=https://${SEAT_DOMAIN}``

Its the 5 time i try do config correctly... i want to host it on my synology nas using cloudflare tunnel I have make it work one time but somehow it was not using https, only http. Now im in a loop

soo.. first thing you need to make sure the cloudflare tunnel is on the same docker network as seat

and use this setting in cloudflare tunnel

@user_229678072176050179

docker run -d --name cloudflare-seat --restart unless-stopped --network seat-docker_seat-frontend cloudflare/cloudflared:latest tunnel run --token <your token from cloudflare>

@user_229678072176050179 here is the command i use to connect cloudflare tunnel with seat network.

And is this a good compose for tunnel: ```version: '3.8' services: cloudflared: image: cloudflare/cloudflared:latest containername: cloudflare-tunnel restart: always command: tunnel run environment:

• TUNNEL

TOKEN=test volumes:

• ./config:/etc/cloudflared

networks:

• seat

networks: seat: external: true name: seat_network ``

that can work too

but the network name is ``seat-docker_seat-frontend``

Manage to get online. But my issue now is that in /volume1/docker/seat-docker/.env i have everything set up. But when i try to login i get error "invalidrequest" errordescription "The redirect URL does not match any of the configured values for this client." And this because of the /var/www/seat/.env I have check with nano /var/www/seat/.env and and its seems that some info are missing: EVECLIENTID=null EVECLIENTSECRET=null EVECALLBACKURL=https://seat.local/auth/eve/callback APPNAME=SeAT APPENV=local APPKEY= APPDEBUG=false APP_URL=http://localhost

thats because your ESI URL doesn't match your local URL if you are running custom domains even local ones they need to match

is the last output the .env you edited?

you need to fix the ``APPURL=`` to ``APPURL=https://${SEAT_DOMAIN}`` and leave it..

The ``EVECALLBACKURL=`` isnt right.. but it should work.. however, best to change it back to ``EVECALLBACKURL=${APP_URL}/auth/eve/callback``