cannot connect hister

[widowild1]

Hello,
Hister is pretty slow; I've crawled 22,000 pages so far. Hister is running in a Docker container under LXC.

My base_url points to my domain name (data/config.yml)

I'm not crawling right now.
The database is in SQLite and is 4.2M in size.

I also have a question: when I crawl, Hyster becomes inaccessible.

[asciimoo]

My base_url points to my domain name (data/config.yml)

Do you use a reverse proxy?

I also have a question: when I crawl, Hyster becomes inaccessible.

It should not become inaccessible. How do you run the crawler? is it an independent command while the server is running?

[widowild1]

Do you use a reverse proxy?

Yes, cloudflare tunnel, I use Cloudflared in a separate LXC-Docker container.
https://community-scripts.org/scripts/cloudflared

I ran the command in Portainer and added a rule through the Hister interface; it's probably a problem with writing to the database.

Other than indexing datasets (https://hister.org/datasets), I haven't had any issues without running a search.

When I quickly refresh the Hister homepage, I see the same message for a few seconds, and then it disappears.
https://pastebin.com/KbDUYJRX

For Hister, I'm allocating the following:
4 Cores
8 GB
512 MB swap
60 GB hard drive

[asciimoo]

Yes, cloudflare tunnel, I use Cloudflared in a separate LXC-Docker container.

Probably it does not forward the websocket connection properly.

the following configuration is required for nginx to make websocket forwarding work:

# Configuration necessary for the /search endpoint.
proxy_set_header   Upgrade    $http_upgrade;
proxy_set_header   Connection $hister_connection; # This variable is defined in a `map` below.
proxy_read_timeout 86400;

map $uri $hister_connection {
	/search upgrade;
	default close;
}

Similar configuration is required for CF. Full nginx config: https://hister.org/docs/server-setup#nginx .

I ran the command in Portainer and added a rule through the Hister interface; it's probably a problem with writing to the database.

I'm not familiar with this setup. Make sure the server is running when you execute the crawl/import commands and specify your server URL for the crawl/import commands with the -u or --server-url flags.

In case of no success, try to increase the log level to debug. (Hister options can be set using environment variables instead of command line flags: https://hister.org/docs/configuration#environment-variables )

[widowild1]

Actually, I don't know what to do with Nginx. I've been browsing and searching through the Cloudflare documentation and came across this article
https://developers.cloudflare.com/network/websockets/

The WebSocket option is indeed enabled.
but I'm continuing my research

[asciimoo]

Actually, I don't know what to do with Nginx.

You don't have to do anything with nginx, I just gave an example configuration where reverse proxying is solved in nginx. Probably similar header forwarding is required in other software as well.

You can check if there is an error with the websocket connection using your browser's network inspector panel (press F12).