Browser Not Working

[cuneglas]

Hi there, First off all, thanks for this project. I've been using your online version and now I was trying to make the jump to a local setup.

All went well until I tried to record a website. When I open a website the browser window keeps showing me

Unable to connect Firefox can’t establish a connection to the server at localhost:8092.

I've tried with different browsers and same error persists.

What should I do to resolve this problem?

Cheers

[ikreymer]

Apologies for late response.. Do you see any errors in the docker-compose logs, eg. docker-compose logs app especially? Have you changed any default settings?

[cuneglas]

No problem.

I've tried to make a clean install to see if the error persisted, but now I can't even launch webrecorder.

I'm using a Synology DS916+, I followed your guide, but had to make some changes to complete the installation.

I installed webrecorder in a custom folder, so I've used git clone https://github.com/webrecorder/webrecorder . to install in the current folder.

When I did run the sudo docker-compose up -d command I got an error saying:

ERROR: for dat Cannot start service dat: Bind mount failed: '/volume2/docker/webrecorder/data/dat/secret_keys' does not exists

I've created the missing folder. This error stopped but I've got a new one.

ERROR: for frontend UnixHTTPConnectionPool(host='localhost', port=None): Read timed out. (read timeout=60)

To resolve this error this, I did sudo COMPOSE_HTTP_TIMEOUT=3600 docker-compose up -d and the installation was completed with no errors.

Right after the installation I received a notification saying that the containers app_1, dat_1, nginx_1, recorder_1, warcserver_1 had stopped unexpectedly, I went to see logs and the information from each container is displayed below:

dat_1

The rootDir argument was not supplied and is required
Usage: run [options]
Options:
  -V, --version            output the version number
  -p, --port [port]        The port the api server is to bind to (default: 3000)
  -h, --host [host]        The host address the server is listen on (default: "0.0.0.0")
  -s, --swarm-port [port]  The port the swarm is to bind to (default: 3282)
  -r, --rootDir <dir>      The root directory that contains the contents to be shared via dat
  -l --log                 should logging be enabled for both the api server and swarm manager
  -h, --help               output usage information

nginx_1:

2019/04/23 13:40:23 [emerg] 1#1: host not found in upstream "app:8081" in /etc/nginx/nginx.conf:45
nginx: [emerg] host not found in upstream "app:8081" in /etc/nginx/nginx.conf:45

app_1 / recorder_1 / warcserver_1 /

groupadd: GID '100' already exists
groupmod: group 'archivist' does not exist

What should I do?

Thanks for your help and all the hard work.

Cheers

[ikreymer]

Sorry about that, a bad dat-share image was pushed by accident. It should be fixed now, run docker pull webrecorder/dat-share and then try restarting compose with docker-compose up -d as before, hopefully that should start all the containers.. (The GID error may be ignorable)

[cuneglas]

The same errors continue to appear and the same containers continue to stop unexpectedly.

[cuneglas]

Hi, any news?

[Per0x]

I have the same problem on the clean install, but with solr in addition.