Supabase Old Docker Composer

[Adarsh-Kumar-Gupta]

Hello Guys,

I just wanted to know, why we're still using an old version of docker composer file for supabase, we've latest releases in supabase and there docker compose file also get updated last month, but we're still using an old version of docker composer, can someone please update this or help me to use latest docker compose version? I tried manually from my side and I got lot's of errors.

"Container supabase-vector Error dependency vector failed to start dependency failed to start: container supabase-vector is unhealthy Error: ❌ Docker command failed Error occurred ❌, check the logs for details."

[ismaildasci]

the supabase-vector unhealthy error is usually because of missing config or volume permissions.

try these:

1. check vector logs:

docker logs supabase-vector

2. common fixes:

• make sure all required env vars are set in .env
• check if volumes have correct permissions:

sudo chown -R 1000:1000 ./volumes/

3. if using latest supabase docker-compose:
   they changed some service names and configs. check their repo for migration guide:
   https://github.com/supabase/supabase/tree/master/docker

4. quick workaround - disable vector if you dont need it:
   comment out vector service in docker-compose and remove depends_on references

5. manual update steps:

# backup your data first!
docker-compose down
git pull  # or download new docker-compose
docker-compose pull
docker-compose up -d

what version were you on before and what version are you trying to update to? the error logs from vector container would help debug

[Adarsh-Kumar-Gupta]

I did fix by adding mount volume, now getting an another error

"
2026-01-17T15:47:48.096617Z INFO vector::app: Internal log rate limit configured. internal_log_rate_secs=10
2026-01-17T15:47:48.097859Z INFO vector::app: Log level is enabled. level="vector=info,codec=info,vrl=info,file_source=info,tower_limit=trace,rdkafka=info,buffers=info,lapin=info,kube=info"
2026-01-17T15:47:48.098179Z INFO vector::app: Loading configs. paths=["/etc/vector/vector.yml"]
2026-01-17T15:47:48.099604Z ERROR vector::cli: Configuration error. error=Missing env var required in config. name = "LOGFLARE_PUBLIC_ACCESS_TOKEN", error = "LOGFLARE_PUBLIC_ACCESS_TOKEN is required"
2026-01-17T15:47:48.099706Z ERROR vector::cli: Configuration error. error=Missing env var required in config. name = "LOGFLARE_PUBLIC_ACCESS_TOKEN", error = "LOGFLARE_PUBLIC_ACCESS_TOKEN is required"
2026-01-17T15:47:48.099746Z ERROR vector::cli: Configuration error. error=Missing env var required in config. name = "LOGFLARE_PUBLIC_ACCESS_TOKEN", error = "LOGFLARE_PUBLIC_ACCESS_TOKEN is required"
2026-01-17T15:47:48.099799Z ERROR vector::cli: Configuration error. error=Missing env var required in config. name = "LOGFLARE_PUBLIC_ACCESS_TOKEN", error = "LOGFLARE_PUBLIC_ACCESS_TOKEN is required"
2026-01-17T15:47:48.099840Z ERROR vector::cli: Configuration error. error=Missing env var required in config. name = "LOGFLARE_PUBLIC_ACCESS_TOKEN", error = "LOGFLARE_PUBLIC_ACCESS_TOKEN is required"
2026-01-17T15:47:48.099880Z ERROR vector::cli: Configuration error. error=Missing env var required in config. name = "LOGFLARE_PUBLIC_ACCESS_TOKEN", error = "LOGFLARE_PUBLIC_ACCESS_TOKEN is required"
2026-01-17T15:47:48.099930Z ERROR vector::cli: Configuration error. error=Missing env var required in config. name = "LOGFLARE_PUBLIC_ACCESS_TOKEN", error = "LOGFLARE_PUBLIC_ACCESS_TOKEN is required"

I updated the "vector.yml" but there was an env variable, which I already added in enviroment variable but it was not taking that env and give this error

[Adarsh-Kumar-Gupta]

Let me know if you have any idea

[ismaildasci]

oh nice you fixed with volume 👍

so this error means env variable is not passing to container. adding to .env file is not enough, you need to define it in docker-compose.yml too.

add this to vector service:

environment:
  - LOGFLARE_PUBLIC_ACCESS_TOKEN=${LOGFLARE_PUBLIC_ACCESS_TOKEN}

or if you have env_file: .env it should work but sometimes it bugs.

if you dont use logflare anyway just put dummy value:

LOGFLARE_PUBLIC_ACCESS_TOKEN=test123

then run:

docker-compose up -d --force-recreate vector

if still not working share your vector part from docker-compose.yml i can check

[Adarsh-Kumar-Gupta]

Okay that's fixed now another error comes :(

"Container supabase-analytics Error dependency analytics failed to start
Container supabase-analytics Error dependency analytics failed to start
Container supabase-analytics Error dependency analytics failed to start
Container supabase-analytics Error dependency analytics failed to start
Container supabase-analytics Error dependency analytics failed to start
Container supabase-analytics Error dependency analytics failed to start
Container supabase-analytics Error dependency analytics failed to start
Container supabase-analytics Error dependency analytics failed to start
dependency failed to start: container supabase-analytics is unhealthy
Error: ❌ Docker command failed
Error occurred ❌, check the logs for details."

[ismaildasci]

haha welcome to supabase self-hosting 😅 always one more error

ok now check analytics logs:

docker logs supabase-analytics

and share here what it says

but tbh if you doing local dev you probably dont need analytics and vector services. they are for logging/monitoring stuff.

you can just comment them out in docker-compose.yml:

# analytics:
#   image: supabase/logflare
#   ...

# vector:
#   image: timberio/vector
#   ...

and remove depends_on: analytics from other services

this way you get clean supabase running without all the monitoring stuff. you can enable them later when you actually need

what are you trying to build btw? maybe i can help better if i know your use case

[Adarsh-Kumar-Gupta]

I undertand you point but I really check logs for the database, self-host feels very hard for someone new🥹

Here the logs for "supabase-analysis"

`16:24:16.947 [error] ** (Protocol.UndefinedError) protocol String.Chars not implemented for {:error, %DBConnection.ConnectionError{message: "connection not available and request was dropped from queue after 12074ms. This means requests are coming in and your connection pool cannot serve them fast enough. You can address this by:\n\n 1. Ensuring your database is available and that you can connect to it\n 2. Tracking down slow queries and making sure they are running fast enough\n 3. Increasing the pool_size (although this increases resource consumption)\n 4. Allowing requests to wait longer by increasing :queue_target and :queue_interval\n\nSee DBConnection.start_link/2 for more information\n", severity: :error, reason: :queue_timeout}} of type Tuple. This protocol is implemented for the following type(s): Atom, BitString, Ch.Query, ConfigCat.User, Date, DateTime, Decimal, Explorer.Duration, Float, Integer, List, Money, NaiveDateTime, OpenApiSpex.Cast.Error, Phoenix.LiveComponent.CID, Postgrex.Copy, Postgrex.Query, Tesla.Client, Tesla.Env, Time, URI, Version, Version.Requirement
(elixir 1.17.3) lib/string/chars.ex:3: String.Chars.impl_for!/1
(elixir 1.17.3) lib/string/chars.ex:22: String.Chars.to_string/1
(logflare 1.27.0) lib/logflare/repo.ex:28: Logflare.Repo.get_uptime/0
(logflare 1.27.0) lib/logflare_web/controllers/health_check_controller.ex:11: LogflareWeb.HealthCheckController.check/2
(logflare 1.27.0) lib/logflare_web/controllers/health_check_controller.ex:1: LogflareWeb.HealthCheckController.action/2
(logflare 1.27.0) lib/logflare_web/controllers/health_check_controller.ex:1: LogflareWeb.HealthCheckController.phoenix_controller_pipeline/2
(phoenix 1.7.21) lib/phoenix/router.ex:484: Phoenix.Router.call/5
(logflare 1.27.0) lib/logflare_web/endpoint.ex:1: LogflareWeb.Endpoint.plug_builder_call/2

16:24:16.949 [warning] Going Down - :shutdown - Elixir.Logflare.Sources.Source.Supervisor - last state:

16:24:16.949 [warning] [Elixir.Logflare.Sources.Counters] terminating - shutdown

16:24:16.962 [notice] Application logflare exited: Logflare.Application.start(:normal, []) returned an error: shutdown: failed to start child: Logflare.SystemMetricsSup
** (EXIT) shutdown: failed to start child: Logflare.SystemMetrics.AllLogsLogged
** (EXIT) an exception was raised:
** (DBConnection.ConnectionError) connection not available and request was dropped from queue after 11755ms. This means requests are coming in and your connection pool cannot serve them fast enough. You can address this by:

1. Ensuring your database is available and that you can connect to it
2. Tracking down slow queries and making sure they are running fast enough
3. Increasing the pool_size (although this increases resource consumption)
4. Allowing requests to wait longer by increasing :queue_target and :queue_interval

See DBConnection.start_link/2 for more information

            (ecto_sql 3.13.2) lib/ecto/adapters/sql.ex:1098: Ecto.Adapters.SQL.raise_sql_call_error/1
            (ecto_sql 3.13.2) lib/ecto/adapters/sql.ex:996: Ecto.Adapters.SQL.execute/6
            (ecto 3.13.5) lib/ecto/repo/queryable.ex:241: Ecto.Repo.Queryable.execute/4
            (ecto 3.13.5) lib/ecto/repo/queryable.ex:19: Ecto.Repo.Queryable.all/3
            (ecto 3.13.5) lib/ecto/repo/queryable.ex:163: Ecto.Repo.Queryable.one/3
            (logflare 1.27.0) lib/logflare/system_metrics/all_logs_logged/all_logs_logged.ex:20: Logflare.SystemMetrics.AllLogsLogged.init/1
            (stdlib 6.2.2.2) gen_server.erl:2229: :gen_server.init_it/2
            (stdlib 6.2.2.2) gen_server.erl:2184: :gen_server.init_it/6

{exit,terminating,[{application_controller,call,2,[{file,"application_controller.erl"},{line,511}]},{application,enqueue_or_start,6,[{file,"application.erl"},{line,380}]},{application,ensure_all_started,3,[{file,"application.erl"},{line,359}]},{elixir,start_cli,0,[{file,"src/elixir.erl"},{line,195}]},{init,start_it,1,[]},{init,start_em,1,[]},{init,do_boot,3,[]}]}
Kernel pid terminated (application_controller) ("{application_start_failure,logflare,{{shutdown,{failed_to_start_child,'Elixir.Logflare.SystemMetricsSup',{shutdown,{failed_to_start_child,'Elixir.Logflare.SystemMetrics.AllLogsLogged',{#{message => <<"connection not available and request was dropped from queue after 11755ms. This means requests are coming in and your connection pool cannot serve them fast enough. You can address this by:\n\n 1. Ensuring your database is available and that you can connect to it\n 2. Tracking down slow queries and making sure they are running fast enough\n 3. Increasing the pool_size (although this increases resource consumption)\n 4. Allowing requests to wait longer by increasing :queue_target and :queue_interval\n\nSee DBConnection.start_link/2 for more information\n">>,reason => queue_timeout,'struct' => 'Elixir.DBConnection.ConnectionError','exception' => true,severity => error},[{'Elixir.Ecto.Adapters.SQL',raise_sql_call_error,1,[{file,"lib/ecto/adapters/sql.

Crash dump is being written to: erl_crash.dump...done
LOGFLARE_NODE_HOST is: 127.0.0.1

16:24:21.538 [info] Starting migration

16:24:22.096 [error] Postgrex.Protocol (#PID<0.151.0>) failed to connect: ** (Postgrex.Error) FATAL 3D000 (invalid_catalog_name) database "_supabase" does not exist

16:24:22.096 [error] Postgrex.Protocol (#PID<0.150.0>) failed to connect: ** (Postgrex.Error) FATAL 3D000 (invalid_catalog_name) database "_supabase" does not exist

16:24:23.815 [error] Postgrex.Protocol (#PID<0.150.0>) failed to connect: ** (Postgrex.Error) FATAL 3D000 (invalid_catalog_name) database "_supabase" does not exist`

[ismaildasci]

yeah self-hosting supabase is pain honestly 😅

ok i see the problem. look at this line:

database "_supabase" does not exist

analytics service trying to connect to _supabase database but it doesnt exist yet.

you need to create it manually. connect to your postgres and run:

CREATE DATABASE _supabase;

or via docker:

docker exec -it supabase-db psql -U postgres -c "CREATE DATABASE _supabase;"

then restart analytics:

docker-compose restart analytics

---

but honestly bro, if you just want to learn/develop with supabase, just use their cloud free tier. self-hosting is for production stuff and needs lot of devops knowledge.

all these analytics, vector, logflare services are just for monitoring. core supabase (db, auth, storage, realtime) works without them.

what are you trying to do exactly? maybe cloud version is enough for you?

[Adarsh-Kumar-Gupta]

Yes I undertand your concern, I was working on a project and I really don't want backend stuff, and currently I don't have enough data I can surely use cloud supabase but may be in future I'll deal with more data, so self host will be better option I belive.

So, I run this command but still no result

Here the logs
"16:44:10.481 [notice] Application logflare exited: Logflare.Application.start(:normal, []) returned an error: shutdown: failed to start child: Logflare.SystemMetricsSup
** (EXIT) shutdown: failed to start child: Logflare.SystemMetrics.AllLogsLogged
** (EXIT) an exception was raised:
** (DBConnection.ConnectionError) connection not available and request was dropped from queue after 11803ms. This means requests are coming in and your connection pool cannot serve them fast enough. You can address this by:

1. Ensuring your database is available and that you can connect to it
2. Tracking down slow queries and making sure they are running fast enough
3. Increasing the pool_size (although this increases resource consumption)
4. Allowing requests to wait longer by increasing :queue_target and :queue_interval

See DBConnection.start_link/2 for more information

            (ecto_sql 3.13.2) lib/ecto/adapters/sql.ex:1098: Ecto.Adapters.SQL.raise_sql_call_error/1
            (ecto_sql 3.13.2) lib/ecto/adapters/sql.ex:996: Ecto.Adapters.SQL.execute/6
            (ecto 3.13.5) lib/ecto/repo/queryable.ex:241: Ecto.Repo.Queryable.execute/4
            (ecto 3.13.5) lib/ecto/repo/queryable.ex:19: Ecto.Repo.Queryable.all/3
            (ecto 3.13.5) lib/ecto/repo/queryable.ex:163: Ecto.Repo.Queryable.one/3
            (logflare 1.27.0) lib/logflare/system_metrics/all_logs_logged/all_logs_logged.ex:20: Logflare.SystemMetrics.AllLogsLogged.init/1
            (stdlib 6.2.2.2) gen_server.erl:2229: :gen_server.init_it/2
            (stdlib 6.2.2.2) gen_server.erl:2184: :gen_server.init_it/6

{exit,terminating,[{application_controller,call,2,[{file,"application_controller.erl"},{line,511}]},{application,enqueue_or_start,6,[{file,"application.erl"},{line,380}]},{application,ensure_all_started,3,[{file,"application.erl"},{line,359}]},{elixir,start_cli,0,[{file,"src/elixir.erl"},{line,195}]},{init,start_it,1,[]},{init,start_em,1,[]},{init,do_boot,3,[]}]}
Kernel pid terminated (application_controller) ("{application_start_failure,logflare,{{shutdown,{failed_to_start_child,'Elixir.Logflare.SystemMetricsSup',{shutdown,{failed_to_start_child,'Elixir.Logflare.SystemMetrics.AllLogsLogged',{#{message => <<"connection not available and request was dropped from queue after 11803ms. This means requests are coming in and your connection pool cannot serve them fast enough. You can address this by:\n\n 1. Ensuring your database is available and that you can connect to it\n 2. Tracking down slow queries and making sure they are running fast enough\n 3. Increasing the pool_size (although this increases resource consumption)\n 4. Allowing requests to wait longer by increasing :queue_target and :queue_interval\n\nSee DBConnection.start_link/2 for more information\n">>,reason => queue_timeout,'struct' => 'Elixir.DBConnection.ConnectionError','exception' => true,severity => error},[{'Elixir.Ecto.Adapters.SQL',raise_sql_call_error,1,[{file,"lib/ecto/adapters/sql.

Crash dump is being written to: erl_crash.dump...done
LOGFLARE_NODE_HOST is: 127.0.0.1

16:44:15.526 [info] Starting migration

16:44:15.930 [error] Postgrex.Protocol (#PID<0.150.0>) failed to connect: ** (Postgrex.Error) FATAL 3D000 (invalid_catalog_name) database "_supabase" does not exist

16:44:15.930 [error] Postgrex.Protocol (#PID<0.151.0>) failed to connect: ** (Postgrex.Error) FATAL 3D000 (invalid_catalog_name) database "_supabase" does not exist

16:44:17.433 [error] Postgrex.Protocol (#PID<0.150.0>) failed to connect: ** (Postgrex.Error) FATAL 3D000 (invalid_catalog_name) database "_supabase" does not exist

16:44:18.057 [error] Postgrex.Protocol (#PID<0.151.0>) failed to connect: ** (Postgrex.Error) FATAL 3D000 (invalid_catalog_name) database "_supabase" does not exist

16:44:20.805 [error] Postgrex.Protocol (#PID<0.150.0>) failed to connect: ** (Postgrex.Error) FATAL 3D000 (invalid_catalog_name) database "_supabase" does not exist

16:44:22.397 [error] Postgrex.Protocol (#PID<0.151.0>) failed to connect: ** (Postgrex.Error) FATAL 3D000 (invalid_catalog_name) database "_supabase" does not exist"

[ismaildasci]

still same error bro - database "_supabase" does not exist

did you create the database? run this:

docker exec -it supabase-db psql -U postgres -c "CREATE DATABASE _supabase;"

if it says container not found, check your db container name:

docker ps | grep postgres

might be supabase-postgres or supabase-db or something else

after creating database, restart everything:

docker-compose down
docker-compose up -d

---

btw looking at all these errors... honestly just disable analytics and vector services. they are optional and causing all these headaches.

in your docker-compose.yml comment out:

• analytics service
• vector service
• and remove any depends_on: analytics or depends_on: vector from other services

you will have working supabase without logging/monitoring stuff. you can always enable later when you actually need them

are you following official supabase self-hosting guide? https://supabase.com/docs/guides/self-hosting/docker

which guide/repo are you using?

[Adarsh-Kumar-Gupta]

Yes I follow the documentation, but it's for direct self hosting using docker, not ushing any third party docker providers like dockploy

I run this command

root@supabase:/etc/dokploy/compose/supabase-zluskb/code# docker exec -it supabase-db psql -U postgres -c "CREATE DATABASE _supabase;"
CREATE DATABASE
root@supabase:/etc/dokploy/compose/supabase-zluskb/code#

[ismaildasci]

ohh you using dokploy, that changes things. dokploy has its own networking and volume setup, thats probably why things are breaking

ok so database is created now. run this to check:

docker exec -it supabase-db psql -U postgres -c "\l"

you should see _supabase in the list

then full restart:

docker-compose down
docker-compose up -d

if still not working, the problem might be timing - analytics starts before db is fully ready. check if db container is healthy:

docker ps | grep supabase

---

but real talk bro - dokploy + supabase self-host is advanced combo. you're fighting two systems at once 😅

my honest advice:

1. for learning/small project → just use supabase cloud, its free up to 500mb
2. if you really want self-host → use plain docker-compose on a vps, not through dokploy. follow official guide exactly: https://supabase.com/docs/guides/self-hosting/docker
3. if you must use dokploy → disable analytics and vector services, they're not worth the headache for now

when your project grows and you actually need self-hosting, you'll have more devops experience to handle these issues

what does your project do? maybe i can suggest simpler setup

[Adarsh-Kumar-Gupta]

This project is an for realtime canvas design, an basic project but more advance features, like multi-user can do edits in a single canvas like an figma, not an complete figma, it just 15% of figma.

I think I need to use supabase directly, insted of using dockploy, or else dockploy need to update there supabase template to latest verison with complete test and run, cause supabase template is too old there and that template only works perfectly.

I'm not an expert at devops and docker, I might not be the correct person who solve this problem, I request the team if anyone can do fix and provide the latest version of supabase template, that would be awesome.

Thanks for giving your valuable time and great help @ismaildasci , I really appricates your patient and debuging skill, keep doing great work and make community always active and happy 😄