search

hcengineering / huly-selfhost

Huly for Self Hosting
Eclipse Public License 2.0
396 stars 50 forks source link

Unable to create/join workspace with v0.6.265+ and Traefik #31

Open Trishar263 opened 3 weeks ago

Trishar263 commented 3 weeks ago

Hello, I’m currently using Huly with Traefik using #19.

It works perfectly well with v0.6.264 but with v0.6.265+ I get an insecure connection with ws when I try to create/join/connect to a workspace. Something I can fix in services.account.enviroment with

The problem is that I get an infinite loading when I try to join/connect to a workspace, and I get stuck at 30% when I try to create one. Apparently a few people had this error some weeks ago:

Do you know how I could resolve it or why it is happening?

fendle commented 3 weeks ago

I had the same issue with the newest version and went back to the version of the docker compose file: v0.6.280

Trishar263 commented 2 weeks ago

Hello,

Little update, I tried again, with v0.6.295 this time . I don’t have the insecure connection error anymore, but I still get stuck when I create a workspace, this time at 0 %. And it’s the same result with both ws or wss.

lexiv0re commented 2 weeks ago

Hello @Trishar263, for v0.6.295 you need to configure an additional service. Please check the updated configuration https://github.com/hcengineering/huly-selfhost/blob/main/template.compose.yaml#L62.

Trishar263 commented 2 weeks ago

Hello, sorry I wasn’t clear, I tried with v0.6.295, after making all necessary changes to the configuration including the service workspace.*

lexiv0re commented 2 weeks ago

Could you provide logs from your workspace container?

Trishar263 commented 2 weeks ago

Hi, I changed a few things meanwhile, so with wss I’m not always stuck at 0 % anymore, and I can create workspace but sometimes it gets stuck at 0 %, often at 100 % and a few times it quickly creates it and I can join it after that, even if I’m unable to invite anyone because once again there is an infinite loading.

Here are the logs for workspace

{"level":"info","message":"####################################################################################################################","timestamp":"2024-09-18T06:53:15.863Z"} {"level":"info","message":"########################SplitLogger workspace initialized: 2024-09-18T06:53:15.864Z###########################","timestamp":"2024-09-18T06:53:15.864Z"} Starting workspace service in region: DEFAULT for operation: all for version: 0.6.294 with brandings: {}

But I don’t have any error in logs or web console.

Trishar263 commented 2 weeks ago

I just have this if I wait long enough (5-10mins) while it is loading infinitely workspace-1 | {"err":{"message":"fetch failed","stack":"TypeError: fetch failed\n at node:internal/deps/undici/undici:13178:13\n at async updateWorkspaceInfo (/usr/src/app/bundle.js:252:440524)\n at async lZt._createWorkspace (/usr/src/app/bundle.js:628:141904)\n at async lZt.doWorkspaceOperation (/usr/src/app/bundle.js:628:143361)\n at async /usr/src/app/bundle.js:628:141014\n at async lZt.exec (/usr/src/app/bundle.js:628:141175)"},"level":"error","message":"error","timestamp":"2024-09-18T08:27:27.678Z","workspace":"w-try-nameofmyworkspace-66ea8741-ddd4d496e4-b414ac","workspaceName":"nameofmyworkspace"}