You don't necessarily need a domain, but it will need few tweaks - starting with traefik container usage Self-hosting is not recommended for a prod environment though - starting with security questions

domains are pretty cheap depending on what company you buy from

It's more a question of used tld. Some of them are especially expensive due to the low amount of domains registered on them. As an exemple, org, net and com are far less cheaper than space, cloud or io

.space is normally what I see for eve

.io is pricy for me

Manually added them to squad but now seat is removing them from squads... ugh

what is your critera for the squad?

``Alliance is <Alliance in Question>``

It's been working fine for a month or more now just started to stop assigning people to the squad

token state ?

also any way to see your squad filters ?

Hey folks, in the context of notification groups, what exactly should the notifications::alerts.disabled_token alert do?

I assume it's to notify you if someone biomasses their character? or they got banned or token is invalid or revoked?

Yeah… it doesn’t seem to do anything for me when a token becomes invalid which is what I’m trying to get an alert to fire on… so I’m not sure.

Supposed to

Aww so CCP indeed issues some bad tokens

and better but still bad

Wow

It already appends in the past Almost reason of my issue

I suspect them to had deploy some new server in their pool which has not timezone or npc properly set

Is there a way to install another branch than the default(e.g. a dev branch) on a docker install?

Just change the tag in your compose to the dev branch tag

Is there a method of having Seat-connector ignore certain roles??

put those roles above the bot in role list

anything above bot won't be touched

Yes, but on docker, plugins sre usually installed over the .env file

Again change your tag in the compose to run the dev branch

You asking a plugin or are you asking for docker if docker change the tag.. if plugin not fully sure.

Plugin

Oh then no idea

You can append the branch/tag in the .env of each plugin you want. @user_614098468218339348

As long as the tag exists in packagist you can target it directly I.e creator/pluginname:dev or creator/pluginname:4.1

thanks, I'll try it out

hello everyone, I installed seat, but my user data is not loaded, the icon is not updated and the server status is displayed as Unknown. what could be the problem ?

F12 any errors?

yes, just this one

any errors in the server logs?

And is this an existing install or new one?

how fresh is the install? it might take a few minutes for it to load the data for the first time

installed 10 minutes ago or even more where can i see the logs on the server ?)

How'd you install it?

manually installed on unbuntu 20.04

check if the workers are running: your-site.com/horizon/dashboard

freaked out, reinstall everything)

using manual installation it was not possible to install, but deployed in docker everything works fine )

docker is mostly following the same as manual install so it just highlight that you made a mistake while reading the documentation but what matter is you've been able to reach your objective, so...

docker is better in the end as you don't butcher anything host wise and if you want to remove it you don't have to spend an hour removing all traces of something. But that's just my opinion.

Docker (and others like it) are generally better not in only managing install, updates and eol cleanup but in also security if properly setup as it limits exposure to the host itself.

Though the sad thing I see is people often (new) to containers will put the (--privileged) or ``privileged: true`` on every container as they seen one container use it...

Rip, yeah I ran across that once on another install I assisted and immediately ripped that off.

If you can't tell me why you need it or why you can't use CAP_ADD for it if you really need it then there's no point in having it.

I have this error

@user_562000345547669513 tried to clear your eseye cache ? php artisan seat:cache:clear will handle this for your. It occurres time to time but we never find a valid reason except serializing issue. What's used php version ?

thats work, thnx, PHP 7.4.29

Hi, hoping someone can help. I keep getting

DataTables warning: table id=dataTableBuilder - Ajax error. For more information about this error, please see http://datatables.net/tn/7

When trying to load assets for any character

latest entry to the log is...

[2022-06-22 21:08:25] local.ERROR: ReflectionFunction::__construct() expects parameter 1 to be string, object given {"userId":20,"exception":"[object] (Symfony\\Component\\Debug\\Exception\\FatalThrowableError(code: 0): ReflectionFunction::__construct() expects parameter 1 to be string, object given at /var/www/seat/vendor/yajra/laravel-datatables-oracle/src/Utilities/Helper.php:67)
[stacktrace]
#0 /var/www/seat/vendor/yajra/laravel-datatables-oracle/src/Utilities/Helper.php(67): ReflectionFunction->__construct(Object(Seat\\Web\\Http\\DataTables\\Character\\Intel\\Assets\\Columns\\Station))
#1 /var/www/seat/vendor/yajra/laravel-datatables-oracle/src/Processors/DataProcessor.php(137): Yajra\\DataTables\\Utilities\\Helper::compileContent(Object(Seat\\Web\\Http\\DataTables\\Character\\Intel\\Assets\\Columns\\Station), Array, Object(Seat\\Eveapi\\Models\\Assets\\CharacterAsset))
#2 /var/www/seat/vendor/yajra/laravel-datatables-oracle/src/Processors/DataProcessor.php(106): Yajra\\DataTables\\Processors\\DataProcessor->addColumns(Array, Object(Seat\\Eveapi\\Models\\Assets\\CharacterAsset))

...

One sec

You will need to do this https://discord.com/channels/821361165791133716/821361546608508938/953875615398903918

Likely you've also ran into the same issue others ran into with indexing on tables.

will give it a go ty

lessons learned: don't fuck with route binds in plugins, it creates a very confusing mess

😅

didnt work

Is there anyway to have SeAT connect back to discord with the integrations tab? Aka notifying a specific channel when someone applied to corp? I tried to do it this way but I assumed it wouldnt work

Append /slack

In the webhook url? or in a config somewhere?

webhook url

so it'll look similar like this https://discord.com/api/webhooks/discordserverid/yourwebhookid/slack

Awesome thanks 🥳

Does the Paps section of a character's sheet tie into the Calendar or naw?

It's a calendar feat

I want the notifications to trigger when someone applies to corp (ingame). Im not sure how to set it up correctly.

Will this trigger for anybody? Or do I need to add something to the affiliations section?

Assume if not set on affiliations it's for anyone and everyone

I would imagine you would want to set it to a single toon that can receive notification of application otherwise it 'could' search ALL toons and ping for each character that received this.

Ohhhh I see, Ok that makes more sense thanks

Hey there i am new here but i am having issues getting this going i have tried docker desktop as well as ubuntu on a raspberry pi 8gb ram

i ran the automated docker setup on ubuntu it finished in an odd way but when trying to start the containers i get this error

That looks like your docker compose file is invalid

Did you also install docker compose? It’s not included with the docker package, hard to tell but I think it’s not there based on the error.

Docker-compose is installed, it gave the running path of it. It is stating on line 1 of docker-compose.yml is being taken as a command, which is incorrect as that should be the version controller line.

Could be the new docker-compose, ``docker compose`` or the "best one" ``docker-compose``

that would be why the instructions say once that auto script is done that you go straight to monitoring the stack let me try doing it again and do the compose thing

i remember docker-compose

This was the actual correct answer, you should review the compose file.

It is also worth checking what version of docker and docker-compose you have installed... Older versions of Ubuntu have repositories containing incompatible versions. As to the raspberry pi, seat at the moment doesn't build arm containers meaning you would need to build them yourself to run it on the rpi.

Ah I see ok so I’d be better off running the docker desktop then. I’m not the best at building arm containers lol.

By docker desktop, you mean on your personal PC? Are you intending to share this install with other people?

docker desktop is just really bleh in the end, I'd take just docker engine over the docker desktop version (docker desktop is for windows/mac and includes a gui to make containers with and such) I'd prefer (docker engine which is just the cli and no gui what linux hosts that have docker have)

In Connection.php line 669:

                                                                               

  SQLSTATE[HY000] [1045] Access denied for user 'seat'@'172.29.0.5' (using pa  

  ssword: YES) (SQL: select * from information_schema.tables where table_sche  

  ma = seat and table_name = migrations and table_type = 'BASE TABLE')         

                                                                               


In Exception.php line 18:

                                                                               

  SQLSTATE[HY000] [1045] Access denied for user 'seat'@'172.29.0.5' (using pa  

  ssword: YES)                                                                 

                                                                               


In PDOConnection.php line 40:

                                                                               

  SQLSTATE[HY000] [1045] Access denied for user 'seat'@'172.29.0.5' (using pa  

  ssword: YES)

this is error i get on docker desktop after setup this is coming from the web container

just cheap and its a server like all other games i have running on it

If you're running containers off windows or mac just know containers are limited.

yeah indeed i have a domain and we want seat but i do not know enough about it to get it up and going

i wish there was a way to make it near one click easy haha

Well if you need help just ask

yeah here i am lol

genuinely don't know, not super experienced with seat

I would've been having a panic attack or something if I was told someone deleted stuff in the Scheduler

o lol

ight i set it to restart the queue at 00 every night now

Cause it happened once... and people are often not thrilled about reauthing their tokens

and put back the bucket update/list commands(

+------------------------------+
| location_flag                |
+------------------------------+
| AssetSafety                  |
| AutoFit                      |
| Cargo                        |
| CorpDeliveries               |
| CorpSAG1                     |
| CorpSAG2                     |
| CorpSAG3                     |
| CorpSAG4                     |
| CorpSAG5                     |
| CorpSAG6                     |
| CorpSAG7                     |
| DroneBay                     |
| FighterBay                   |
| FighterTube0                 |
| FighterTube1                 |
| FighterTube2                 |
| FighterTube3                 |
| FighterTube4                 |
| FleetHangar                  |
| FrigateEscapeBay             |
| Hangar                       |
| HiddenModifiers              |
| HiSlot0                      |
| HiSlot1                      |
| HiSlot2                      |
| HiSlot3                      |
| HiSlot4                      |
| HiSlot5                      |
| HiSlot6                      |
| HiSlot7                      |
| Impounded                    |
| Locked                       |
| LoSlot0                      |
| LoSlot1                      |
| LoSlot2                      |
| LoSlot3                      |
| LoSlot4                      |
| LoSlot5                      |
| LoSlot6                      |
| LoSlot7                      |
| MedSlot0                     |
| MedSlot1                     |
| MedSlot2                     |
| MedSlot3                     |
| MedSlot4                     |
| MedSlot5                     |
| MedSlot6                     |
| MedSlot7                     |
| OfficeFolder                 |
| QuantumCoreRoom              |
| RigSlot0                     |
| RigSlot1                     |
| RigSlot2                     |
| SecondaryStorage             |
| ServiceSlot0                 |
| ServiceSlot1                 |
| ServiceSlot2                 |
| ServiceSlot3                 |
| ShipHangar                   |
| SpecializedAmmoHold          |
| SpecializedCommandCenterHold |
| SpecializedFuelBay           |
| SpecializedOreHold           |
| StructureFuel                |
| SubSystemBay                 |
| SubSystemSlot0               |
| SubSystemSlot1               |
| SubSystemSlot2               |
| SubSystemSlot3               |
| Unlocked                     |
+------------------------------+

There is no known current issue. To actually help you we need to see the logs of what happened.

OK.... You really don't need to or want to clear your caches for config, route or views. They do not need to be regularly cleared unless you are developing packages etc.

Hi guys, is there a way to give access for a person to see the "Settings -> User" list seperately? or is giving admin the only way to show them this list?

pretty sure thats admin only

like 80%

I dont think theres a read only version of that page

yeah we dont have a granular permission like that atm

fuck yea, got one right

I see an option under SeAT Settings, Maintenance, to clean up stale data. I've set this to true. I'm testing retention of data. How often does it clean up stale data? Is there a way I can manually trigger this?

Have a look in your scheduler for the maintenance job. That will tell you how often it will run automatically. Likewise if use artisan to run the maintenance command then that job will happen immediately.

speaking of the maintenance job, what constitues a invalid user?

Ah ha! Found both seat:admin:maintenance and how to use artisan! Thank you! 🙂

docker or bare metal install?

I noticed the maintenance command doesn't remove mail from Characters, Mail Timeline. Is there another process that does that? Does that currently get cleaned?

Docker. This is what I ran to have it execute immidiatly: docker compose exec seat-web php artisan seat:admin:maintenance

docker-compose exec -T seat-web php artisan seat:admin:maintenance

Is there a trick to make the discord connector ignore a role?

How long does SeAT retain mail?

until the begining of time of the db

unless someone wipes the user out

I've wiped the user out, but the mail is still there. :/

Ah fun can you open it and view it ?

When I log in as admin, I go to Characters, Mail Timeline, all e-mails are already open, so I can read it. I don't need to open them.

Well it makes sense actually that it's not tied to character data as it relates to other info and such

Is mail tied to something else?

i'm getting an issue where when i try to mysqlcheck -A -o ALL my tables return with either

note     : Table does not support optimize, doing recreate + analyze instead
error    : Table 'users' already exists
status   : Operation failed

or

note     : Table does not support optimize, doing recreate + analyze instead
error    : Index for table 'corporation_role_histories' is corrupt; try to repair it
status   : Operation failed

i m getting an issue where when i try to

Does anyone know a way to bring back a "new and fresh" DB for MariaDB? I wiped out the MariaDB and I cannot recovery from Backup. Is it possible just to make a new container for MariaDB while retaining the other containers as they are?

I mean you can.. But you lose all your data

I believe the command docker-compose down -v mariadb will do what you are looking for. But I would stop the other containers first.

But this loses you everything. Tokens! The lot is gone.

The first thing I would be trying to find out is what caused the db to be corrupted in the first place. And see if you have a backup from before then

with the thread I made above it seems noone knows how to help, also looked up on other sources and I end up to be in this circle of "cannot repair, oh repair it" state. the question is, I do not know if whether MariaDB is derping or my Backups are corrupted.

it all happend one day because they told me seat was giving off a 500 error, so i just did a usual, docker down docker up with reboot

then the page never loaded and i check logged and it had a bunch of maria db issues, i tried to do the "mysqlcheck -A -o" fix to clean and optimize it up but it just never improved from there.

cannot back up because the db index is corrupted try to fix. cannot repair because its innodb, then innodb either gives off, cannot optimize (apparently it is the method for fixing InnoDB's indexes) please fix... and its just a circle or InnoDB just yells at me for some error that is unfixable with my current knowledge of SQL

Yeah it's not an error I have seen before. What you could try and do is inspect your backups to make sure they are intact. You could try restore them to another mariadb instance and see if that works? Other than that I am not sure what recourse there is here. I am also not familiar enough to deep dive into mariadb to recover a broken database.

If the backups are intact, you should be able to blow away the database completely and then restore the backup and be back up and running. Without a working backup I'm not sure what else you can do...

my problem is i dont know how to send data over linux. i tried an email like mailutil or mutt. or any advice on this part on how to send 1-2 gigs file? idk if its possible to load it onto a google drive from commandline :/

I like to use rsync, but it really depends what you mean by sharing data. Downloading it to your local machine? Putting it on google drive to make it accessible for multiple peoples? A backup on another computer?

i just need to move the backup copies from one machine to another, the size of the file is about 2 gb. we have another test instance already set up on another remote server so we can try to use that.

Sounds like a job for rsync

Hey iam new to the admin business of seat. And i have a new Corp Memeber who has created seat Profile and added alla acounts and chars, but i cant get information about his chars in Seat. Token status is yellow. And it says chack Jobs. But all seems fine in the little Job indicator on top right. What can i do to chek why Chars arent loading. Didnt find the right documentation about it. Thanks in advance

nvm restartet the whole seat and worked

How do you setup seat to auto assign roles to new members? for instance i want anybody who joins my seat to see the connector tab so they can join discord. I dont want to have to assign the roles to members manualy

create a squad with auto assign with filter as example anyone but you and add role to it seat connector view that you will have to create as well.

So after upgrading from 4. to 5. it appears that a lot of jobs are failing with: Illuminate\Queue\MaxAttemptsExceededException: Seat\Eveapi\Jobs\Contracts\Character\Contracts has been attempted too many times. in /var/www/seat/vendor/laravel/framework/src/Illuminate/Queue/MaxAttemptsExceededException.php:24 Also the render time of a page is more than 10 secs which seems insane

Is your CPU load constantly on >95%?

if so, try to reduce workers, if not increase them

Will reduce workers, thanks for the heads up 🙂

 Illuminate\Queue\MaxAttemptsExceededException: Seat\Eveapi\Jobs\Skills\Character\Skills has been attempted too many times. in /var/www/seat/vendor/laravel/framework/src/Illuminate/Queue/MaxAttemptsExceededException.php:24

way to fix or do i just wait for awhile

Getting this error after a recent upgrade of services on docker 5.x

apache2: Syntax error on line 146 of /etc/apache2/apache2.conf: Could not open configuration file /etc/apache2/mods-enabled/mpm_event.load: Permission denied

No config changes

Frontend is just cycling ad-infinitum

AFAIK I'm not managing the apache config or sub modules in any meaningful way

Looks like there's an issue with that module sometime after seat:5.0.33 - pinning front to this version stabilizes apache2

https://discord.com/channels/821361165791133716/821361546608508938/1233852629914877963

Just release a new build which disable that event module

Sweet, thank you!

I too need to do some database tuning it seems. Had almost 31k failed jobs for “attempted too many times.” Currently have 32 workers on a 6 core 16GB host. Workers seem fine until mariadb jumps off a cliff

i had too much workers and reduced them, jpm went up and no failing jobs anymore

What’d you end up at as a sweet spot?

14 Workers for four dedicated Milan 7003 processors

dont know your CPU, but core performance itself must be considered, too. E.G. 6 cores i7-8700 vs. 6 cores i7-11850

server load isn't consistent, too. So observing a 5 minute period doesn't mean your amount of workers are fine. Also reducing worker count takes some time until the queue catches up (for me it were several hours)

I see, I was at 16 workers before increasing to 32 tonight. Letting it run for a day to see how things go.

I manage a pretty consistent 30% cpu usage with 16 workers despite all the “attempted too many times” errors in the failed jobs.

if the low load is consistent, i would increase the amount of workers^^

but doubling doesn't mean doubled CPU load, cause Mariadb has to do work, too. So maybe it's more a quadrupling? ^^

maybe "divide and rule" should be used to find a sweet spot for the amount of workers

Yep, or the age old “fuck around and find out”

I’m on 5.0.33 which is 22 days old which as I understand doesn’t have some of the scheduled job optimizations.

5.0.37 was just released so I’ll switch to that in the morning

ohh, there were several bug fixes which caused failing jobs^^

Whats the easiest way to reduce numbers for docker? I'm having a absolute brainfart for some reason

So just to check on this, ran docker-compose exec front php artisan se:ca:cl to clear the queu and it seems fine now which is odd

i've the suspicion that exceptions, which are hitting CPU hard, are an effective downward spiral in causing even more failing jobs, since every exception causes a high cpu load

yup the failed ones are causing more strain causing it to just stop lol

Just be careful, each worker takes a minimum of 128MB of memory to intiate as well. So minimum you'd need 4GB memory to run all workers at once.

whats the command to increase workers ?

its in the .env file QUEUE_WORKERS parameter, I think the default is 4

is there any real benefit of increasing number of workers?

more then 10 I mean

It's strange that my seat can work when I'm using, but shows 500 when new pilot tring to connect in.

You will want more workers as you gain more tokens that are processed. Too few and it'll be a bottleneck where it cannot process all jobs before the next hour.

If your jobs are completing with ample time before the next batch then 10 workers is fine. Adjust as needed otherwise

enable debug mode to catch the error or look at the logs. 500 errors always leave something behind in the logs

Doesn't exist in my .env file

You have to add it.

Do you have enough workers? Are you getting a lot of errors? What is your jobs per minute?

It should. There is no update command etc, squad membership is calculated each time a character or squad has a change that effects eligibility.

That doesn't seem to be the case, if you change a filter it takes quite a few minutes, up to an hour, for the members to update.

I mean... It is the case because that's the only code path that will affect the membership of a squad.. It can take some time though as it will then have to go through your full user list to check eligibility again. For some filters that is a very expensive operation.

Yeah. If you have enough tokens to need it, and enough processing power to facilitate it.

Typically it's ram.. There is a runaway case because laravel stores the failed jobs in redis to start with. So more failed jobs consumes more ram. This leaves less ram available for processing which slows down jobs and can lead to more failures etc. This problem has always existed.. It's just more prevalent given we are running into the issues with failed jobs lately.

I just completely down&up the entire stack, which reset the errors. But before, I had always over 100k failed jobs, and a queue of over 10k, and it taking over 1h for character-related jobs to go through

That's what it looks like right now

Interestingly, before that down&up, the mass-failing jobs were PlanetDetail ones. Now it's all Citadel ones.

And they all fail with Illuminate\Queue\MaxAttemptsExceededException: Seat\Eveapi\Jobs\Universe\Structures\Citadel has been attempted too many times. in /var/www/seat/vendor/laravel/framework/src/Illuminate/Queue/MaxAttemptsExceededException.php:24

Citadel is normal and nothing we can do about it.... They will get better over time.. But reset each down and up you do.

Yeah that exception just says jobs are going too slowly...

I set it to have 12 queue workers with auto balancing

so I really doubt it's lacking for resources

Well I mean 500 jobs per minute is not a lot.. But I think that's because your queue is empty I imagine..

Yeah, I just cleared it out an hour ago, it was much much more before, but I don't recall the exact number

Don't conflate workers with resources.

But yes that's plenty of CPU and ram

I wonder if mariadb can't keep up somehow

It's the single biggest CPU hog

What sort of disk is this running on? Mariadb should be making good use of available ram while it has it.

It's a Proxmox VM on top of a raid1 on NVMe SSDs

Are the disks passed through or virtualised?

It's virtualized, via a zvol

but performance is more than fine

Right now everything looks fine, there's a lot of failed Citadel jobs, but you say that's normal?

Define more than fine.... Seat uses a very large amount of disk io and will burn through disks reasonably quickly

Right now, the read/write load is topping out at ~2M/s, according to iotop

Yes. Unfortunately citadel jobs fail a lot because of the way esi is designed. We have a mapping in cache that will prevent follow up errors, but if you restart and clear your cache that will reset that mapping.

And it's capable of writing over 4G/s sustained

But what about the iops. Bandwidth is less relevant for a db

Yes they are related but not necessarily linearly.

From how the system behaves, it's generally very clear that the storage is not stressed at all as well. iostat also confirms that there is no high load

Cool. It would be one to watch though as perf on disk is a big performance driver for seat jobs.

The Other thing you can try is because you have so much ram, placing your eseye cache in ram.

Is that just a setting in .env? Cause yeah, the RAM is kinda unused right now

Yeah it's configured from the env file. It's a new feature that is not for general use (as most people have much smaller servers) but should help for bigger installs. Though maybe just keep an eye on ram usage over time.

In the docs, it looks like I set ESEYECACHEDRIVER=redis, and the defaults should be fine for the other options?

Should be yeah

I would love to hear feedback on how it goes for you

I guess I just restart the whole stack again for that to take effect?

Or workers only?

Really only the workers and front services need it

You shouldn't need to restart redis or mariadb or traefik

No traefik in my case

compose up -d figured it out on its own

Yeah it will

Right now it doesn't look very different. Will give it a few days and see how jobs behave

The main issue, all the PlanetDetail jobs, seems to have resolved by just clearing out the cache

.env file is fine

something is messed up somewhere. and im not sure

That is normal yes

I figured

Any time you see a 5XX error there will be logs detailing the issue.

.docs.troubleshooting

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

Would these be the contributing issues?

esi logs

The full stacktrace from this one would be helpful. And as text not an image please as I won't be able to read it easily on mobile.

sure thing, sorry

didnt want to blow the channel up with text

All good, usually it's long enough discord will make it into a tct file anyway

Here's the logs in their entirety from the last attempt at rebooting.

Well most of them.

Are these the logs you specified?

It looks like it cannot connect to redis. Is the cache container running?

yes

root@srv515318:~

cd /opt/seat-docker

docker compose logs --tail 10 -f WARN[0000] /opt/seat-docker/docker-compose.yml: version is obsolete scheduler-1 | scheduler-1 | scheduler-1 | INFO No scheduled commands are ready to run. scheduler-1 | scheduler-1 | scheduler-1 | INFO No scheduled commands are ready to run. scheduler-1 | scheduler-1 | scheduler-1 | INFO No scheduled commands are ready to run. scheduler-1 | worker-1 | worker-1 | worker-1 | INFO Configuration cached successfully. worker-1 | worker-1 | worker-1 | INFO Routes cached successfully. worker-1 | worker-1 | worker-1 | INFO Horizon started successfully. worker-1 | cache-1 | 1:M 01 May 2024 07:03:26.056 100 changes in 300 seconds. Saving... cache-1 | 1:M 01 May 2024 07:03:26.057 Background saving started by pid 27 cache-1 | 27:C 01 May 2024 07:03:26.062 DB saved on disk cache-1 | 27:C 01 May 2024 07:03:26.063 Fork CoW for RDB: current 0 MB, peak 0 MB, average 0 MB cache-1 | 1:M 01 May 2024 07:03:26.157 Background saving terminated with success cache-1 | 1:M 01 May 2024 07:08:27.076 100 changes in 300 seconds. Saving... cache-1 | 1:M 01 May 2024 07:08:27.079 Background saving started by pid 28 cache-1 | 28:C 01 May 2024 07:08:27.084 DB saved on disk cache-1 | 28:C 01 May 2024 07:08:27.084 Fork CoW for RDB: current 0 MB, peak 0 MB, average 0 MB cache-1 | 1:M 01 May 2024 07:08:27.179 Background saving terminated with success front-1 | 172.23.0.3 - - [01/May/2024:07:09:43 +0000] "GET /queue/short-status HTTP/1.1" 503 256 "https://auth.feastingfrenzy.space/home" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/123.0.0.0 Safari/537.36"

Hmm can you bring the whole stack down and up using the docker compose command, not a host reboot

yes

its down

and its up

Cache log spat this out 1:C 01 May 2024 07:14:39.594 # WARNING Memory overcommit must be enabled! Without it, a background save or replication may fail under low memory condition. Being disabled, it can also cause failures without low memory condition, see https://github.com/jemalloc/jemalloc/issues/1328. To fix this issue add 'vm.overcommit_memory = 1' to /etc/sysctl.conf and then reboot or run the command 'sysctl vm.overcommit_memory=1' for this to take effect.

cache-1 | 1:C 01 May 2024 07:14:39.594 oO0OoO0OoO0Oo Redis is starting oO0OoO0OoO0Oo cache-1 | 1:C 01 May 2024 07:14:39.594 Redis version=7.2.4, bits=64, commit=00000000, modified=0, pid=1, just started cache-1 | 1:C 01 May 2024 07:14:39.594

Warning: no config file specified, using the default config. In order to specify a config file use redis-server /path/to/redis.conf

cache-1 | 1:M 01 May 2024 07:14:39.595 monotonic clock: POSIX clock_gettime cache-1 | 1:M 01 May 2024 07:14:39.596 Running mode=standalone, port=6379. cache-1 | 1:M 01 May 2024 07:14:39.596 Server initialized cache-1 | 1:M 01 May 2024 07:14:39.596 Ready to accept connections tcp

DB seems to be starting just fine

Yeah that's semi normal I believe. Any better luck on the 503?

No luck

Maybe the DB is corrupt?

Hoping this doesnt come down to a full reinstall because I've already made my corp register twice lol

should i just leave it for a while and see if it magically starts to work?

Firewall?

no, at least nothing that would have changed between the time it was working and stopped working

CPU load ?

Host packages update ? Make sure you're up-to-date

ran update, restarted, no luck

Just going to do a fresh install.

leave that hammer down, what did you fix ?

what do you mean?

what I mean - you've a problem, you reinstall - what did you fix ?

what's your host OS ?

run iptables -L to show your firewall rules

also, is host able to ping a hostname, like ping esi.evetech.net ?

root@srv515318:/opt/seat-docker

iptables -L

Chain INPUT (policy ACCEPT) target prot opt source destination Chain FORWARD (policy DROP) target prot opt source destination DOCKER-USER all -- anywhere anywhere DOCKER-ISOLATION-STAGE-1 all -- anywhere anywhere ACCEPT all -- anywhere anywhere ctstate RELATED,ESTABLISHED DOCKER all -- anywhere anywhere ACCEPT all -- anywhere anywhere ACCEPT all -- anywhere anywhere ACCEPT all -- anywhere anywhere ctstate RELATED,ESTABLISHED DOCKER all -- anywhere anywhere ACCEPT all -- anywhere anywhere ACCEPT all -- anywhere anywhere ACCEPT all -- anywhere anywhere ACCEPT all -- anywhere anywhere ctstate RELATED,ESTABLISHED DOCKER all -- anywhere anywhere ACCEPT all -- anywhere anywhere ACCEPT all -- anywhere anywhere Chain OUTPUT (policy ACCEPT) target prot opt source destination Chain DOCKER (3 references) target prot opt source destination ACCEPT tcp -- anywhere 172.28.0.3 tcp dpt:https ACCEPT tcp -- anywhere 172.28.0.3 tcp dpt:http Chain DOCKER-ISOLATION-STAGE-1 (1 references) target prot opt source destination DOCKER-ISOLATION-STAGE-2 all -- anywhere anywhere DOCKER-ISOLATION-STAGE-2 all -- anywhere anywhere DROP all -- !172.27.0.0/16 anywhere DROP all -- anywhere !172.27.0.0/16 DOCKER-ISOLATION-STAGE-2 all -- anywhere anywhere RETURN all -- anywhere anywhere Chain DOCKER-ISOLATION-STAGE-2 (3 references) target prot opt source destination DROP all -- anywhere anywhere DROP all -- anywhere anywhere DROP all -- anywhere anywhere RETURN all -- anywhere anywhere Chain DOCKER-USER (1 references) target prot opt source destination RETURN all -- anywhere anywhere root@srv515318:/opt/seat-docker#

Ping not working apparently

ping to google is fine but not to that esi site

0.0.0.10 ? I don't see how it can work indead 😄

show your network config using ifconfig

Dont be snooping my IP lol

your IP, what do you mean ?

just joking

show your current dns ?

systemd-resolve --status

ok I guess its formatting with markdown

Current DNS Server: 217.21.79.10 DNS Servers: 217.21.79.10 8.8.4.4 1.1.1.1

try those

nslookup esi.evetech.net 217.21.79.10
nslookup esi.evetech.net 8.8.4.4
nslookup esi.evetech.net 1.1.1.1

should have an IP front of each of them, even if it might differ

all seem to work

and doing ping esi.evetech.net is returning you 0.0.0.10 ?

no

not anymore

so definitely might have some DNS issue

ping is blocked by ESI, so this is now ok

hmm..

running SeAT 5 ?

yes

using traefik or proxy config ?

traefik

not sure you can run ping with seat image... let's try it

docker compose -f docker-compose.yml -f docker-compose.mariadb.yml -f docker-compose.traefik.yml exec front sh -c 'ping cache'

nope

what's the output ?

just blankl

then, not blank 😛

typed it wrong first time lol

might think about adding it in base image...

could try to execute a query to the db?

you've an issue with cache, isn't it ?

@user_76781846418956288

can you update your SeAT stack ?

sure

done

root@srv515318:/opt/seat-docker

docker compose -f docker-compose.yml -f docker-compose.mariadb.yml -f docker-compose.traefik.yml exec front sh -c 'ping cache'

WARN[0000] /opt/seat-docker/docker-compose.yml: version is obsolete WARN[0000] /opt/seat-docker/docker-compose.mariadb.yml: version is obsolete WARN[0000] /opt/seat-docker/docker-compose.traefik.yml: version is obsolete PING cache (192.168.48.2) 56(84) bytes of data. 64 bytes from seat-docker-cache-1.seat-dockerseat-internal (192.168.48.2): icmpseq=1 ttl=64 time=0.139 ms 64 bytes from seat-docker-cache-1.seat-dockerseat-internal (192.168.48.2): icmpseq=2 ttl=64 time=0.117 ms 64 bytes from seat-docker-cache-1.seat-dockerseat-internal (192.168.48.2): icmpseq=3 ttl=64 time=0.091 ms 64 bytes from seat-docker-cache-1.seat-dockerseat-internal (192.168.48.2): icmpseq=4 ttl=64 time=0.088 ms 64 bytes from seat-docker-cache-1.seat-dockerseat-internal (192.168.48.2): icmpseq=5 ttl=64 time=0.092 ms 64 bytes from seat-docker-cache-1.seat-dockerseat-internal (192.168.48.2): icmpseq=6 ttl=64 time=0.112 ms 64 bytes from seat-docker-cache-1.seat-dockerseat-internal (192.168.48.2): icmpseq=7 ttl=64 time=0.091 ms 64 bytes from seat-docker-cache-1.seat-dockerseat-internal (192.168.48.2): icmpseq=8 ttl=64 time=0.107 ms

ping works now

Now redo the same but replacing cache with esi.evetech.net

root@srv515318:/opt/seat-docker

docker compose -f docker-compose.yml -f docker-compose.mariadb.yml -f docker-compose.traefik.yml exec front sh -c 'ping esi.evetech.net'

WARN[0000] /opt/seat-docker/docker-compose.yml: version is obsolete WARN[0000] /opt/seat-docker/docker-compose.mariadb.yml: version is obsolete WARN[0000] /opt/seat-docker/docker-compose.traefik.yml: version is obsolete PING esi.evetech.net (34.249.87.194) 56(84) bytes of data.

Sounds good Checking your job queue ?

Using queue dashboard I mean

Site is still unavailable

https://auth.feastingfrenzy.space/home

Is it time to throw the hammer yet? lol

No

It's just time to read log, diagnose and fix issues

So, what logs say ?

Can you confirm front can ping cache ?

Also can you dump your .env minus eve app and db credentials ?

Need to go to bed. Will pick back up tomorrow

I'm convinced the DB is corrupted. Tried a new instance and it worked fine, imported the backup of the db and got the old error

This also happened during setup front-1 | Copying directory [vendor/laravel/framework/src/Illuminate/Mail/resources/views] to [resources/views/vendor/mail] DONE front-1 | Copying file [vendor/laravel/tinker/config/tinker.php] to [config/tinker.php] DONE front-1 | front-1 | front-1 | INFO Configuration cached successfully. front-1 | front-1 | front-1 | INFO Routes cached successfully. front-1 | front-1 | front-1 | In Connection.php line 829: front-1 | front-1 | SQLSTATE[HY000] [1045] Access denied for user 'seat'@'172.22.0.4' (using pa front-1 | ssword: YES) (Connection: mysql, SQL: select tablename as name, (datale front-1 | ngth + indexlength) as size, tablecomment as comment, engine as engi front-1 | ne, tablecollation as collation from informationschema.tables where ta front-1 | bleschema = 'seat' and tabletype in ('BASE TABLE', 'SYSTEM VERSIONED') or front-1 | der by tablename) front-1 | front-1 | front-1 | In Connector.php line 65: front-1 | front-1 | SQLSTATE[HY000] [1045] Access denied for user 'seat'@'172.22.0.4' (using pa front-1 | ssword: YES) front-1 | front-1 | front-1 exited with code 0 front-1 | PONG front-1 | starting web service front-1 | Processing plugins from SEATPLUGINS front-1 | Installing plugins: cryocaustik/seat-hr:v2.x-dev front-1 | ./composer.json has been updated front-1 | Running composer update cryocaustik/seat-hr front-1 | Loading composer repositories with package information front-1 | 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. front-1 | Updating dependencies

this error is happening and random points whenever I try to reindex the table, i cannot seem to get the reindexing to finish before of this. mysqlcheck: Got error: 2013: Lost connection to server during query when executing 'OPTIMIZE TABLE ... '

Anything interesting in the docker logs for the mariadb container?

it didn't print any unique errors of any sort. i tried googling this but they all mention some sort of error. i checked the logs and no error other than the 2013 as a hint.

Sorry im a little confused with the documentation regarding about Styling. https://eveseat.github.io/docs/styling/ It says i need to create a "docker-compse.override.yml" and the customer-layout so it can be attached as a volume, however, that override.yml isnt in the docker run command below... or am i supose to make a copy of the "docker-compose.yml" file and rename it as "override" and the custom-layout and replace the normal docker-compse.yml to override in the docker run command???

the .override.yml will merge into the docker-compose.yml

25G seat-docker_mariadb-data oof 😄

any chance to reduce it?

I can't get the styling working for love nor money at the moment.

Same, cant find the logs why the custom-layout are not working...

The documentation sample yml appears to have the mounts commented out (

) so that might be the reason it’s not working for you.

Assuming you copied verbatim and didn’t remove them per the instruction that is

I followed the instructions precisely but it still doesn't work. Last time I had it working was when I had a bare metal install.

docker-compose.override.yml should look like this

version: "3.2"

services:

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

Yep. It does.

in opt/seat-docket you have created the folder custom and added the files, right?

Yep

all you have to do now is run

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

- 2 commands

Correct

run this in opt/seat-docker

that was it for me

WARN[0000] /opt/seat-docker/docker-compose.yml: version is obsolete WARN[0000] /opt/seat-docker/docker-compose.override.yml: version is obsolete

Yea that’s fine

I get that too

Now do a hard refresh and check inspect element on you app check head and see if you have the custom layout in there

Yea... i did all of that.. transfer the layout via filezilla first, moved them to the custom folder and then ran those command.. but.. im not sure why but all of my containers got renamed. E.g seat-docker-front-1 to seat-dockerfront1....

Hmm… have no ideea, maybe someone here can help

redone the command again and it went back to normal "seat-docker-front-1".. but away, it still didnt load the custom-layout.

this might be a dumb question, but want about permissions? mine looks like this....

What do u mean it didn't loaded?

Check the website

Probably it loaded but traefik is blocking images so it displays default ones.

Can u inspect your website in web browser?

it loaded, also im using cloudflare to bypass traefik

I had to either change traefik or downloaded files on a drive and add volumes in override

Ok

Anyway usually it's traefik fault so in order for custom layout to work you have to edit traefik. But the easiest way to check what is the issue is to check website.

> WARNING: Found orphan containers (seat-docker-traefik-1, seat-docker-mariadb-1) for this project. If you removed or renamed this service in your compose file, you can run this command with the --remove-orphans flag to clean it up. When i ran "docker-compose down" i got that message.... why?

Did u do any changes?

Also for layout u dont need to do docker compose down

Just reload all docker-compose files

The only changes was adding override.yml to the run command. I didn't know you didn't have to run docker compose down.. I'll try it again after I make changes..

@user_517842699928469535 @user_574274925775421440 I have got it to work. Check the "custom" folder permissions. I think that was my issue.

Mine is working too now although I can't get images to load.

Might be down to hosting though.

Could do with some more CSS examples.

Having checked the elements, I don't seem to see it loading the background image.

But I do see the custom style.

Same

Im asuming it says unknown because SeAT hasnt loaded the information ?

disk space was full fixed it and now horizon is just a blank page, what do?

Sorry, I'm new here and I'm having trouble updating the seat.

front-1 | front-1 | INFO Routes cached successfully. front-1 | front-1 | front-1 | INFO Running migrations. front-1 | front-1 | 2020_03_26_110237_convert_to_seat_400 ............................. 1ms FAIL front-1 | front-1 | In Connection.php line 829: front-1 | front-1 | SQLSTATE[42S22]: Column not found: 1054 Unknown column 'group_id' in 'field front-1 | list' (Connection: mysql, SQL: select distinct groupid from seatconne front-1 | ctorusers order by groupid asc limit 1000 offset 0) front-1 | front-1 | front-1 | In Connection.php line 423: front-1 | front-1 | SQLSTATE[42S22]: Column not found: 1054 Unknown column 'group_id' in 'field front-1 | list' front-1 | front-1 |

I used docker for the deployment.

are you updating from seat 3?

seat 4

I even downloaded the sql content to check it out, and he shows a groupid that I don't see in seatconnector_users

I searched for group_id again, and found that plugins such as seat_srp, seat_billing, etc. have this data.

It may not have anything to do with that, I'm a newbie and don't really understand the correlations.🥲

Resolved. The way I did it was to re-pull seat-docker, install the plugin, and finally re-import mariadb.

Once I do the update though, the problem reoccurs. I then have to delete seat-docker again and then pull so repeat the operation.

So, i had a user register as blue initially but then switch to recruit (different pulls, 1 full, 1 public only)

Is there a way to get the link character button to default to whichever the most recently selected option is?

Cus even after registering his main as full, the next toons were back to only public until i deleted his user

Looks like this is preventing images from loading... login:1 Refused to load the image 'https://www.imageurl.com/image.png' because it violates the following Content Security Policy directive: "img-src 'self' data: https://images.evetech.net https://img.shields.io".

https://sitecoretweaks.wordpress.com/2020/03/03/refused-to-load-the-image-url-because-it-violates-the-following-content-security-policy-directive-img-src-self-data-sitecore/

sigh..... maybe it would be easier to load the image into the custom folder and load it from the override............

https://github.com/eveseat/seat-docker/blob/f057040806aaf6838d7073a84aea14cb45ad8c7a/docker-compose.traefik.yml#L53

anyone has a hint regarding a blank page with horizon?

Cannot use cryocaustik/seat-hr's latest version v1.0.4 as it requires php ^7.3 which is not satisfied by your platform.isn't that something inside the docker image?

still a blank page in debug mode 😦

It's just the top part "version: 3.2" They phased it out as people thought they can just define it like a versioning.. but if you set it to be like "version: 1" you are telling compose to use the features of the compose spec version 1.. which is very broken and we're on 3.9 right now.. so now they changed if to be if you omit version it'll default to 3.9 and it's also configured to default to 3.9 for the most part to prevent "Dumb" issue reports.

/horizon HTTP/1.1" 200 1207 "https://seat.balanced-unity.de/home" "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:125.0) Gecko/20100101 Firefox/125.0"no errors, just a blank page

Using a 4 or 5 stack ?

Docker 5 with latest versions

Then you can't install this hr package version SeAT 5 is requesting PHP 8 and the image is currently shipped with 8.2 The package version you're using is allowing any PHP 7 version greather or equal to 7.3

deactivated the plugin, now i get a 404 😦 is that a problem: worker-1 | 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.?

nvm, had to do a up -d on all containers not only the basic ones

but horizon is still blank, is it advied to backup db/instance and do a fresh install?

docker compose down and docker volume prune did the trick, just dont down mariadb as it would get pruned, too (and do backups :d)

when i get seat installed in docker and running im getting an error I managed to clear this error with a fresh install. My next problem is my website isn't showing up and it just times out. I'm using a cloudlflare domain that I own. I entered that web address in the setup and have the A records pointing to my server that i have running.

Is this extensive RAM usage due to throwing exceptions or just the failed jobs itself?

@user_301981661761896449 typo?^^

https://discord.com/channels/821361165791133716/821361546608508938/1198017066616246314

When using cloudflare tunnel, link to "seat-docker-front-1:8080"

also make sure both cloudflare tunnel and the seat-docker-front-1 are on the same docker network.

how do i do that link

im sorry im a new person aat this

If you are using cloudflare tunnel with docker, the command i used was this: "docker run -d --name cloudflare-seat --restart unless-stopped --network seat-docker_seat-frontend cloudflare/cloudflared:latest tunnel --no-autoupdate run --token"

And that above you enter in cloudflare tunnel.

now my site doesnt time out it just instantly denies me so 1 problem solved

says no such host

btw.. did you copy your token from cloudflare? Only cloudflare will give that once you setup the tunnel..

i hit the copy in the docker setup and pasted it

from cloudflare?!... next you will need to add that tunnel to seat's network..

I cheated by using portainer

@user_135376836723998720 sorry.. i think i used the wrong command for the docker

lol

docker run -d --name cloudflare-seat --restart unless-stopped --network seat-docker_seat-frontend cloudflare/cloudflared:latest tunnel --no-autoupdate run --token <this is where cloudflare will give you your token to connect to their services>

remove your cloudflare tunnel and try again.

so following asrik's directions i got the tunnel setup and running but for some reason i still cant access seat. here is the error i get now

is the DNS entry existing? (nslookup seat.weformbl0p.com)

For the discord connector to work does the user name in discord have to match the user name in seat

finally got it running using a full manual install however characters can be linked but jobs will queue not run...what do i need to do (or show devs) to find out why jobs are not running?

went back and did a packages update (had two that needed updated) and that broke horizon haha. I can still see that jobs are not running but i get a 500 server error when going to horizon

No

So a 500 error implies there will be logs as to the fault. Check those first

.docs.troubleshooting

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

bleh shitainer ahaha

When using tunnels when you actually configure the path it creates the dns entry to point to the argo url for you

my cron miss some schedule task, where i can find some error log? my seat version is 4.x, use docker

[2024-05-05 21:53:27] local.ERROR: Undefined variable $assetsAreCurrent (View: /var/www/seat/vendor/laravel/horizon/resources/views/layout.blade.php) {"userId":2,"exception":"[object] (Illuminate\\View\\ViewException(code: 0): Undefined variable $assetsAreCurrent (View: /var/www/seat/vendor/laravel/horizon/reson/resources/views/layout.blade.php) at /var/www/seat/storage/framework/views/abc473b97b113419365880aae003be6f.php:146)

here is the error

i also ran the diag script from the seat docs and that checked out everything is good there

checked the site and now jobs are running however i can not view the /horizon page i get this instead

I am asking a question hoping for an answer that I can pass along to my guy who technically runs my Seat. We updated Seat some time ago and ever since we did, the SRP module only allows integers, and 1= 100%, so if I set SRP to 80%, I am paying 80x SRP. As the person in charge of alliance finances, this hurts my soul. Anybody know if there is some backend thing that might be messed up or am I just stupid?

Either that or cyno ships are now worth 25b a pop 😄

Assuming that an integer of 1 is 100% SRP, we may deduce from that statement 0.8 is 80% SRP. I am guessing that is what was entered?

.8 is spat back as being dissallowed

as .8 is not an integer

I can grab a screeny of the error it tosses me

Yeah, I am not very familiar with SeAT.

But, you would think that is what they would want to enter, but it is only taking integer types.

The question is which module you are using, i.e. plugin.

I will ask and see what I can find out as I do not know offhand

Yeah, so far by looking at the SeAT website I see two different packages for SRP.

Literally months of this being ap roblem, had SS and had other people test it, every time it came back with wrong amount or not being configurable

I ask a question here and go back to get SS of the issues

it has fixed itself

This discord is magic and I intend to ask all my lifes questions here, next up, I keep working not having a million dollars in my checking acccount, can some one tell me how this is easily fixed?