latest upgrade: container already in use and "user not found "

[dirkpetersen]

I am just doing an update the latest version:

First I get this

Error response from daemon: Conflict. The container name "/chat-meilisearch" is already in use by container "054dfcfabe9d8343367abc8cb4e560f47bb72cde08a044a061a59958a6015b6c". You have to remove (or rename) that container to be able to reuse that name.            55.6s

and after deleting the old chat-meilisearch container I get this, user not found

LibreChat-API     | 2026-03-05 03:45:36 error: Passport Local Strategy - User Not Found
LibreChat-API     | 2026-03-05 03:45:36 error: [Login] [Login failed] [Username: xxxxx@yyy.edu] [Request-IP: 172.18.0.7]
LibreChat-NGINX   | 128.193.8.81 - - [05/Mar/2026:03:45:36 +0000] "POST /api/auth/login HTTP/2.0" 404 55 "https://chat.hanswurst.de/login" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/145.0.0.0 Safari/537.36" "-"

any ideas what this could be ?

[danny-avila]

These are two separate issues:

1. MeiliSearch container conflict

This happens when the old container still exists. Running docker compose down before pulling the new version and running docker compose up should
prevent this.

2. "User Not Found" after update

This suggests your MongoDB data may not be accessible anymore. To help narrow down the cause, could you provide:

• Your docker-compose.yml (or the relevant mongo/api sections)
• What version you're upgrading from and to
• How you performed the update (what commands you ran, in what order)
• Specifically, did you run docker compose down -v at any point? (the -v flag removes volumes, which would wipe your database)

In the meantime, you can check if your user data still exists by running:

  docker exec -it <your-mongo-container> mongosh

Then:

  use LibreChat;
  db.users.findOne({ email: "xxxxx@yyy.edu" });

If the user document is there, the issue is likely a connection/config mismatch.

[dirkpetersen]

Thanks Danny,

below is my deploy-compose yaml file, the db.users.findOne query returns 'nul' for each user i try but I did not not use the -v option . I am not sure which version i was running prior upgrade but it ran into this

services:
  api:
    # build:
    #   context: .
    #   dockerfile: Dockerfile.multi
    #   target: api-build
    image: ghcr.io/danny-avila/librechat-dev-api:latest
    container_name: LibreChat-API
    ports:
      - 3080:3080
    depends_on:
      - mongodb
      - rag_api
    restart: always
    extra_hosts:
    - "host.docker.internal:host-gateway"
    env_file:
      - .env
    environment:
      - HOST=0.0.0.0
      - NODE_ENV=production
      - MONGO_URI=mongodb://mongodb:27017/LibreChat
      - MEILI_HOST=http://meilisearch:7700
      - RAG_PORT=${RAG_PORT:-8000}
      - RAG_API_URL=http://rag_api:${RAG_PORT:-8000}
    volumes:
      - type: bind
        source: ./librechat.yaml
        target: /app/librechat.yaml
          #      - type: bind
          #        source: ./api/data/auth.json
          #        target: /app/api/data/auth.json
      - ./images:/app/client/public/images
      - ./logs:/app/api/logs

  client:
    image: nginx:1.27.0-alpine
    container_name: LibreChat-NGINX
    ports:
      - 80:80
      - 443:443
    depends_on:
      - api
    restart: always
    volumes:
      - ./client/nginx-ourchat.conf:/etc/nginx/conf.d/default.conf
      - /etc/letsencrypt:/etc/letsencrypt
      - ./client/ssl:/etc/librechat/ssl
  mongodb:
    container_name: chat-mongodb
    ports:  # Uncomment this to access mongodb from outside docker, not safe in deployment
      - 27018:27017
    image: mongo
    restart: always
    volumes:
      - ./data-node:/data/db
    command: mongod --noauth
  meilisearch:
    container_name: chat-meilisearch
    image: getmeili/meilisearch:v1.7.3
    restart: always
    # ports: # Uncomment this to access meilisearch from outside docker
    #   - 7700:7700 # if exposing these ports, make sure your master key is not the default value
    env_file:
      - .env
    environment:
      - MEILI_HOST=http://meilisearch:7700
      - MEILI_NO_ANALYTICS=true
    volumes:
      - ./meili_data_v1.7:/meili_data
  vectordb:
    image: ankane/pgvector:latest
    environment:
      POSTGRES_DB: xxxxx
      POSTGRES_USER: xxxx
      POSTGRES_PASSWORD: xxxxxx
    restart: always
    volumes:
      - pgdata2:/var/lib/postgresql/data
  rag_api:
    image: ghcr.io/danny-avila/librechat-rag-api-dev-lite:latest
    # image: ghcr.io/danny-avila/librechat-rag-api-dev:latest
    environment:
      - DB_HOST=vectordb
      - RAG_PORT=${RAG_PORT:-8000}
    restart: always
    depends_on:
      - vectordb
    env_file:
      - .env

volumes:
  pgdata2:

i have been running this update script nightly for the last 6 months:

cat ~/bin/update-librechat.sh
#!/bin/bash

### essentially running this
#  cd ~/LibreChat
#  docker compose -f deploy-compose-ourchat.yml down
#  git pull
#  docker compose -f deploy-compose-ourchat.yml pull
#  docker compose -f deploy-compose-ourchat.yml up -d
#  #systemctl --user restart rag-api-beta.service
#  docker image prune -a --force

# Configuration
COMPOSE_FILE="deploy-compose-ourchat.yml"
PROJECT_DIR="$HOME/LibreChat"
LOG_DIR="$HOME/.logs"
LOG_FILE="$LOG_DIR/librechat-update.log"

# Colors for output
RED='\033[0;31m'
GREEN='\033[0;32m'
YELLOW='\033[1;33m'
NC='\033[0m'

# Logging function
log() {
    echo "[$(date '+%Y-%m-%d %H:%M:%S')] $1" | tee -a "$LOG_FILE"
}

# Error handler
error_exit() {
    echo -e "${RED}[ERROR] $1${NC}" | tee -a "$LOG_FILE"
    exit 1
}

# Success message
success() {
    echo -e "${GREEN}[SUCCESS] $1${NC}" | tee -a "$LOG_FILE"
}

# Warning message
warning() {
    echo -e "${YELLOW}[WARNING] $1${NC}" | tee -a "$LOG_FILE"
}

# Check that script is NOT running as root
if [ "$EUID" -eq 0 ]; then
    error_exit "This script should NOT be run as root. Run as a regular user in the docker group."
fi

# Check that user is in docker group
if ! groups | grep -q '\bdocker\b'; then
    error_exit "Current user '$USER' is not in the docker group. Run: sudo usermod -aG docker $USER && newgrp docker"
fi

# Create log directory if it doesn't exist
mkdir -p "$LOG_DIR" || error_exit "Failed to create log directory: $LOG_DIR"

# Main execution
log "=== Starting LibreChat update ==="
log "Running as user: $USER (UID: $UID)"

# Change to project directory
log "Changing to project directory: $PROJECT_DIR"
cd "$PROJECT_DIR" || error_exit "Failed to change to directory $PROJECT_DIR"

# Step 1: Stop containers
log "Stopping Docker containers..."
docker compose -f "$COMPOSE_FILE" down || error_exit "Failed to stop Docker containers"
success "Containers stopped"

# Step 2: Pull latest code
log "Pulling latest code from Git..."
git pull || error_exit "Git pull failed"
success "Code updated"

# Step 3: Pull Docker images
log "Pulling latest Docker images..."
docker compose -f "$COMPOSE_FILE" pull || error_exit "Failed to pull Docker images"
success "Docker images updated"

# Step 4: Start containers
log "Starting Docker containers..."
docker compose -f "$COMPOSE_FILE" up -d || error_exit "Failed to start Docker containers"
success "Containers started"

# Wait for containers to initialize
sleep 5

# Verify containers are running
log "Verifying containers are running..."
if docker compose -f "$COMPOSE_FILE" ps | grep -q "Up"; then
    success "Containers are running"
else
    #echo
    error_exit "Containers failed to start properly"
fi

# Step 5: Prune ALL unused Docker images
log "Pruning ALL unused Docker images..."
PRUNE_OUTPUT=$(docker image prune -a --force 2>&1)
if [ $? -eq 0 ]; then
    success "All unused Docker images pruned"
    log "$PRUNE_OUTPUT"
else
    warning "Docker image prune had issues: $PRUNE_OUTPUT"
fi

# Show disk space usage
log "Docker disk usage:"
docker system df | tee -a "$LOG_FILE"

log "=== LibreChat update completed successfully ==="
echo ""
echo -e "${GREEN}✓ Update complete!${NC}"
echo "View logs: tail -f $LOG_FILE"

I run this nightly to purge messages that are reaching an expiration threshold

#! /usr/bin/env python3

import sys, datetime
try:
    import pymongo
except:
    print('pymongo missing, to install run:\n python3 -m pip install --upgrade pymongo')
    sys.exit(1)

"""
 # Some organizations want to purge older messages
 # to stay compliant with their regulatory framework
"""

DAYSAGO=90
PORT='27018'

client = pymongo.MongoClient(f"mongodb://127.0.0.1:{PORT}")
# print(client.server_info())
db = client['LibreChat']
messages = db['messages']
files = db['files']

# Calculate the date X days ago from today
time_ago = datetime.datetime.now(datetime.timezone.utc) - datetime.timedelta(days=DAYSAGO)

# Delete all messages and files where 'createdAt' is older than X days
messages_result = messages.delete_many({"createdAt": {"$lt": time_ago}})
files_result = files.delete_many({"createdAt": {"$lt": time_ago}})

# Output the result of the deletion
print(f"Deleted {messages_result.deleted_count} messages.")
print(f"Deleted {files_result.deleted_count} files.")

[danny-avila]

I think this is related to the registry URL for the container image changing, looking into it. Thanks for the additional details!

[dirkpetersen]

I found there was a block corruption in the disk, I am changing to another VM