Is it possible to run Open-Canvas on Linux fully-local using a local LangGraph CLI instead of LangGraph Studio cloud? #181
Comments
|
I shared an approach for running Open-Canvas on Windows. This method might also work on Linux, so feel free to check it out—it could be useful! However, it still requires a LangSmith API key to push trace information to the cloud, unfortunately. |
|
@lanesky Thank you. This is the log tail, it's on a Arch Linux LXC container on Proxmox VE 8.2. These are the installation commands, starting from a clean ArchLinux-base template with the # Install prerequisites
sudo pacman -S git wget less python python-pip docker
sudo systemctl enable --now docker
sudo systemctl status docker
# Clone repo
cd && mkdir Github && cd Github
git clone https://github.com/langchain-ai/open-canvas/
# Update repo
cd && cd Github/open-canvas
git branch
git switch staging
git branch
git pull
# Install LangGraph-CLI
python -m venv .venv
source .venv/bin/activate
pip install --upgrade pip
pip install -U langgraph-cli
langgraph dockerfile -c langgraph.json Dockerfile
docker build -t open-canvas .
# Install & run Redis: https://hub.docker.com/_/redis/
sudo docker run --name some-redis -d redis redis-server --save 60 1 --loglevel warning
# Install & run Postgres: https://hub.docker.com/_/postgres
sudo docker run --name some-postgres -e POSTGRES_PASSWORD=mysecretpassword -d postgres
# Install & run Open-Canvas
touch .env
# Production: export LANGGRAPH_CLOUD_LICENSE_KEY="***************"
docker run \
--env-file .env \
-p 57318:8000 \
-e REDIS_URI="redis://host.docker.internal:6379/0" \
-e DATABASE_URI="postgresql://myuser:[email protected]:5432/mydatabase" \
-e LANGSMITH_API_KEY="lsv2_pt_*******************" \
open-canvasTo eventually replicate exactly, this is my test LXC housekeeping before the above installation: # On the PVE server, "Create CT" using archlinux-base template
# RAM=4GB+, CPU=1+, DISK=16GB+
# LXC housekeeping, login as root:
pacman-key --init && pacman-key --populate
pacman -Sy archlinux-keyring --noconfirm && pacman -Su --noconfirm
sync && reboot
# Add admin account
pacman -S sudo --noconfirm # In Debian: apt install sudo
useradd --create-home admin
passwd admin
usermod -aG wheel admin # In Debian: use sudo instead of wheel
visudo
# Uncomment to allow members of group wheel to execute any command
%wheel ALL=(ALL:ALL) ALL # In Debian: it's %sudo instead of %wheel
# Enable sshd
nano /etc/ssh/sshd_config
PasswordAuthentication yes
systemctl enable --now sshd
# Get the IP address
ip a | grep 'inet '
echo $HOSTNAME
# Access via ssh from another computer
$ ssh admin@<hostname_or_ip>
sudo nano /etc/pacman.conf
# Misc options
#UseSyslog
Color
ILoveCandy
#NoProgressBar
#CheckSpace
VerbosePkgLists
ParallelDownloads = 5
sudo poweroff
# Backup now the fresh installation,
# then start the LXC. |
|
@PieBru Welcome! Based on the error you posted, it looks like this is likely caused by a name resolution issue — meaning the connection to PostgreSQL failed because the hostname couldn’t be resolved. Looking at the install commands you posted: I think the names for Redis and PostgreSQL in your environment variables should be |
|
Brother, help, I used your method to start docker on linux, and an error occurred when starting open-canvas image. I used langsmith's personal key root@pekpheds01309:/deps/open-canvas# docker logs 60e67404cdd4 |
|
It seems that the API revision you’re running is different from mine. The API revision that worked fine in my environment was Here is the log from my environment." |
|
Brother, I have changed it to langchain/langgraphjs-api:20-4da016a and re-mirrored it, but the same error still occurs. podman start -a open-canvas_langgraph-redis_1 |
|
Sorry to hear that. What is the "https://apigw-beta.test.xxx.com/api/langsmith/auth?langgraph-api=true" in your log? Is it an api stub? |
|
The container is deployed in the company's intranet. Accessing langchain requires a proxy. This address is accessible to the Auth authentication interface. What a strange problem. Brother, is there any other way?
|
|
I have reviewed the related source code (you can also access it by dumping async def migrate() -> None:
async with connect() as conn, conn.cursor() as cur:
try:
results = await cur.execute(
"select version from schema_migrations order by version desc limit 1",
prepare=False,
)
current_version = (await results.fetchone())["version"]
except UndefinedTable:
await cur.execute(
"""
CREATE TABLE schema_migrations (
version bigint primary key,
dirty boolean not null
)
""",
prepare=False,
)
current_version = -1
for migration_path in sorted(os.listdir(config.MIGRATIONS_PATH)):
version = int(migration_path.split("_")[0])
if version <= current_version:
continue
with open(os.path.join(config.MIGRATIONS_PATH, migration_path)) as f:
await cur.execute(f.read(), prepare=False)
await cur.execute(
"INSERT INTO schema_migrations (version, dirty) VALUES (%s, %s)",
(version, False),
)
await logger.ainfo("Applied database migration", version=version) |
|
Hello brother, I have another new question. The langchain/langgraphjs-api service has been started successfully (port is 8123:8000), and the ok interface is returned normally, but when I adjust the front end, I find that 404 is found. Is there something wrong with my opencanvas configuration?
So, what should be configured for the hosts of these interfaces /api/threads/search /api/threads /api/assistants etc./api/*? The log of langchain/langgraphjs-api service is also printed, but 404 2024-11-17T13:47:10.137338Z [info ] POST /api/threads/search 404 0ms [langgraph_api.server] api_revision=4da016a api_variant=local latency_ms=0 method=POST path=/api/threads/search path_params={} proto=1.1 query_string= req_header={'host': '10.68.2xx:8123', 'connection': 'keep-alive', 'content-length': '68', 'user-agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36', 'content-type': 'application/json', 'accept': '/', 'origin': 'http://localhost:3000', 'referer': 'http://localhost:3000/', 'accept-encoding': 'gzip, deflate', 'accept-language': 'zh-CN,zh;q=0.9'} res_header={'content-length': '9', 'content-type': 'text/plain; charset=utf-8'} route=None status=404 |
|
@dashan1127 |
|
Brother, I haven't worked on this recently, and started working on it again today. The front-end service can be started, but when I was doing Q&A, I got an error when I increased the model size. The log showed MODEL_NOT_FOUND. Can you help me figure out what's going on? 2024-11-28T13:28:41.233678Z [info ] HTTP Request: POST http://graph/agent/streamEvents "HTTP/1.1 200 OK" [httpx] api_revision=4da016a api_variant=local Also, I'm using AZURE_MODELS and AZURE_KEY |
Said all in the title.
Thank you.
The text was updated successfully, but these errors were encountered: