zarkoff

Well yes, you're right I need to check how to add a user in postgres... I don't know the syntax or command line.

Yes, it has fixed my problem. (the: psql -U immich -d immich -c "DROP EXTENSION IF EXISTS vectors CASCADE;" in Console). A big thank you for your help, it help me to understand the problem and then find the solution.

@Nodiaque Ok but in Immich log, I see: [Nest] 613 - 07/11/2026, 3:57:47 PM LOG [Api:WebsocketRepository] Initialized websocket server /app/immich/server/dist/services/database.service.js:93 throw new Error(messages.invalidDowngrade({ name, extension, availableVersion, installedVersion })); ^ Error: The database currently has VectorChord 0.5.3 activated, but the Postgres instance only has 0.3.0 available. This most likely means the extension was downgraded. If VectorChord 0.5.3 is compatible with Immich, please ensure the Postgres instance has this available. at /app/immich/server/dist/services/database.service.js:93:23 at process.processTicksAndRejections (node:internal/process/task_queues:104:5) at async /app/immich/server/dist/repositories/database.repository.js:392:27 at async /app/immich/server/node_modules/.pnpm/[email protected]/node_modules/kysely/dist/cjs/kysely.js:541:20 at async DefaultConnectionProvider.provideConnection (/app/immich/server/node_modules/.pnpm/[email protected]/node_modules/kysely/dist/cjs/driver/default-connection-provider.js:12:20) at async /app/immich/server/dist/repositories/database.repository.js:389:13 So 0.5.3 should work no ? But if I use the version : ghcr.io/immich-app/postgres:16-vectorchord0.5.3-pgvector0.8.1 Then the Immich log says (still version 2.7.5): [Nest] 531 - 07/11/2026, 4:12:47 PM WARN [Api:DatabaseService] Could not run vector reindexing checks. If the extension was updated, please restart the Postgres instance. If you are upgrading directly from a version below 1.107.2, please upgrade to 1.107.2 first. node:internal/process/promises:394 triggerUncaughtException(err, true /* fromPromise */); ^ PostgresError: could not access file "$libdir/vectors": No such file or directory at ErrorResponse (file:///app/immich/server/node_modules/.pnpm/[email protected]/node_modules/postgres/src/connection.js:815:30) at handle (file:///app/immich/server/node_modules/.pnpm/[email protected]/node_modules/postgres/src/connection.js:489:6) at Socket.data (file:///app/immich/server/node_modules/.pnpm/[email protected]/node_modules/postgres/src/connection.js:324:9) at Socket.emit (node:events:508:28) at addChunk (node:internal/streams/readable:563:12) at readableAddChunkPushByteMode (node:internal/streams/readable:514:3) at Readable.push (node:internal/streams/readable:394:5) at TCP.onStreamRead (node:internal/stream_base_commons:189:23) { severity_local: 'ERROR', severity: 'ERROR', code: '58P01', file: 'dfmgr.c', line: '208', routine: 'internal_load_library' }So I don't know what to do...* Edit: I think I find the way to do it (I keep previous message to understand to what error the below fix is referring to). in Postgres container: 1. keep the tensorchord/pgvecto-rs:pg16-v0.3.0, 2. in container console, type psql -U immich -d immich -c "DROP EXTENSION IF EXISTS vectors CASCADE;" (Note: If you changed the default database user or database name from "immich", update the -U and -d flags accordingly). Update with new ghcr.io/immich-app/postgres:16-vectorchord0.5.3-pgvector0.8.1 Run Immich it should be ok. Original post: Hope it helps.

@Nodiaque Thank you for your help. I could restore version 2.7.5 and get it working again, but migration is problematic. Attempting to upgrade from tensorchord/pgvecto-rs:pg16-v0.3.0 to ghcr.io/immich-app/postgres:16-vectorchord0.5.3-pgvector0.8.1 breaks my installation: Postgres logs show errors, and Immich enters an infinite loop, as if the upgrade never triggers. Since every upgrade attempt fails, I’ve decided to create a fresh instance and restore from my backups. My Share and Backup files are intact, so I’m confident this will work—I just need time to follow your guide and set it up properly.

Hi, I've a problem with the new version 3. Immich didn''t work anymore: I switched from Redis to Immich_Valkey I switched from Postgres_SQL from tensorchord/pgvecto-rs:pg16-v0.3.0 to ghcr.io/immich-app/postgres:18-vectorchord0.5.3-pgvector0.8.1 After that, Immich started but like it was a new setup I clicked on "restore from a backup" or something like this (my Immich is in French), load a back-up in version v2.7.5 but it fails: Error: /usr/lib/postgresql/18/bin/psql non-zero exit code (3) and 1 other object (see server log for list) ERROR: extension "vectors" is not available HINT: The extension must first be installed on the system where PostgreSQL is running.But I don't know what I have to do. I tried alternative Postgres_SQL (rorar's Repository, bmartino1's Repository) with same result. I read on forum that I have to edit an .env file but I don't see where. Do you have an idea ?

Sorry, I didn’t have time to check before. I deleted an old MariaDB container, removed the syncin container, and tried again. It worked a little further, but I got this message in the MariaDB container and I’m not sure what it means: 2026-02-13 21:59:35 3 [Warning] Aborted connection 3 to db: 'syncindb' user: 'syncinuser' host: '192.168.1.56' (Got an error reading communication packets) I think I might have to give up. Thank you for your help. EDIT 15/02/2026: I finally got it working by following this tutorial, but I used your existing container instead of creating a new one from the template. Here’s what I did differently this time: Deployed the MariaDB container first, using these extra parameters: --user=99:100 --memory=4G --memory-swap=-1 (with a dedicated IP on br0 and keeping the default port) Deployed a Redis container—I’m not sure if it was necessary, but I did it anyway. (with a dedicated IP on br0 and keeping the default port) Uploaded the environment.yaml file before deploying the Sync-in container. Deployed the Sync-in container (with a dedicated IP on br0 and keeping the default port). I did not install OnlyOffice, as it is optional. Now it works! I’m still not sure what the game changer was: the extra MariaDB parameters, adding Redis for caching, or simply uploading environment.yaml before deploying the Sync-in container.

Thank you for taking the time to help me. Here is the docker command (I just changed the HOST_HOSTANME) /usr/local/emhttp/plugins/dynamix.docker.manager/scripts/docker create --name='Sync-in' --net='br0' --ip='192.168.1.76' --pids-limit 2048 -e TZ="Europe/Paris" -e HOST_OS="Unraid" -e HOST_HOSTNAME="MyLab" -e HOST_CONTAINERNAME="Sync-in" -e 'TCP_PORT_8080'='8080' -e 'INIT_ADMIN'='true' -e 'PUID'='099' -e 'PGID'='100' -e 'CA_TS_FALLBACK_DIR'='/app/environment/environment.yaml' -l net.unraid.docker.managed=dockerman -l net.unraid.docker.webui='http://[IP]:[PORT:8080]/' -l net.unraid.docker.icon='' -v '/mnt/user/appdata/sync_in/environment.yaml':'/app/environment/environment.yaml':'rw' -v '/mnt/user/appdata/dynamix/data':'/app/data':'rw' -v '/mnt/user/appdata/sync_in/releases':'/app/static/releases':'ro' 'syncin/server:1' But I did some additional test and: if I start WITHTOUT starting mariadb container, it starts with error of course: Error: unable to migrate database schema ! Error: connect ECONNREFUSED 192.168.1.76:3306 at createConnectionPromise (/app/node_modules/mysql2/promise.js:19:31) at connectToMySQL (/app/node_modules/drizzle-kit/bin.cjs:81027:47) at process.processTicksAndRejections (node:internal/process/task_queues:103:5) at async Object.handler (/app/node_modules/drizzle-kit/bin.cjs:93700:39) at async run (/app/node_modules/drizzle-kit/bin.cjs:93117:7) { code: 'ECONNREFUSED', errno: -111, sqlState: undefined } Error: unable to migrate database schema ! Reading config file '/app/server/infrastructure/database/configuration.js' if I start with mariadb container, it crashes immediately. I tried with official mariadb countainer or linuxserver'one, it's the same. But what's strange, is that it doesn't return an error when trying to connect, it just crashes. Thank you

I've a dedicated mariadb for sync-in, I checked db_name, db_user, and db_password is here the yaml file, all secret are generated via openssl rand -hex 32 command environment-for forum.yaml Here is the message, it closes immediately. log file of the docker says nothing, so I don't know what's wrong...

Hi, thanks for your help. I deleted the file and replaced it with the YAML file instead. Now, Unraid returns an "Execution Error: Bad parameter." I’m not sure if the issue is with what I put in my YAML file or something else, since the log doesn’t show anything. Maybe it’s the MariaDB version or something similar. Here’s what I put in my YAML file—did I miss anything? Does it work for you? mysql: url: mysql://user:[email protected]:3365/dbname auth: encryptionKey: *** generated with openssl rand -hex 32 *** token: access: secret: *** generated with openssl rand -hex 32 *** refresh: secret: *** generated with openssl rand -hex 32 *** applications: files: dataPath: /app/data collabora: enabled: false onlyoffice: enabled: false secret: *** generated with openssl rand -hex 32 ***

Hi, I’m trying to install Sync-In (maintained by @grtgbln , of course). First, thank you for the XML. According to the requirements, I’ve created a MariaDB database and an environment.yaml file. I placed this file in /mnt/user/appdata/sync_in/environment.yaml, but Sync-In can’t find it (error: not found). I didn’t change anything in the XML file except the port, which was already in use by another container. I tried moving the file to a different folder and changing its permissions, but I still get the same error: environment.yaml is not found. Just to clarify, I downloaded the yaml file from GitHub (https://github.com/Sync-in/server/releases) and edited it to add the MariaDB endpoint and encryption keys. Do you have any idea how I could fix this?

Ah super, it works which is strange because I thought OpenDNS was blocked in France (where I am). Thank you so much Edit: well in fact they are... I put other DNS than OpenDNS and it's ok. And I think this is was linked to a problem with AWS affecting a lot of web services... including Orange, my ISP.

Hi, Currently, almost all of my Docker apps are showing as "Not available." I’m running version 7.1.4. I’ve tried the following troubleshooting steps, but the issue persists: Forced an update Edited the app template Restarted my server When I went to the "Apps" tab, a pop-up message briefly appeared—something about "fix..."—but it disappeared before I could read or click on it. Do you know what I can do or where I should look to resolve this issue? Or could this be a temporary problem?

Hi, Sorry I did not notice your answer. I've changed Owner to nobody and access permission and now it works. Thank you.

I've this: /mnt/user/appdata/dumbpad

Hi, I can't get DumbPad to work. Is there any additional setup to do (create en env. or something) ? It looks like a problem of permission. Thank you. LOG