Hello @webmaster-srpl I was able to set up the Facebook Messenger with chatwoot according to the official chatwoot documentation. I did not need to configure anything with /webhooks/facebook so I am curious how you are trying to set up the Facebook Messenger in chatwoot. [image: 1769179599246-600be0a5-9028-4453-8be5-5d9cb87c9bd3-image-resized.png] [image: 1769179549380-fc97ccc6-92f9-41ec-8b35-96091f4edbc9-image-resized.png]

Hello @luisegundo Have you looked at the official Chatwoot documentation for this? https://developers.chatwoot.com/self-hosted/configuration/features/integrations/instagram-channel-setup If so, have you tried following these steps, and if failing, please describe in detail the steps you have taken to reproduce your issue.

@andreasdueren thank you Andreas, it works perfectly. I think it's different code with the tutorial that come from here https://developers.chatwoot.com/self-hosted/configuration/multi-factor-authentication: # Active Record Encryption keys (required for MFA/2FA functionality) # Replace with the actual keys from rails db:encryption:init output ACTIVE_RECORD_ENCRYPTION_PRIMARY_KEY=EGY8WhulUOXixybod7ZWwMIL68R9o5kC ACTIVE_RECORD_ENCRYPTION_DETERMINISTIC_KEY=aPA5XyALhf75NNnMzaspW7akTfZp0lPY ACTIVE_RECORD_ENCRYPTION_KEY_DERIVATION_SALT=xEY0dt6TZcAMg52K7O84wYzkjvbA62Hz When I tried your code, now it appears on my dashboard: # Active Record Encryption keys (required for MFA/2FA functionality) # Replace with the actual keys from rails db:encryption:init output export ACTIVE_RECORD_ENCRYPTION_PRIMARY_KEY=string export ACTIVE_RECORD_ENCRYPTION_DETERMINISTIC_KEY=string export ACTIVE_RECORD_ENCRYPTION_KEY_DERIVATION_SALT=string Now, I can see the 2FA option. Thank you...

@ajax said in OAuth Issue: Incorrect client_id still used despite environment update: The issue is that this client_id refers to an old OAuth credential, even though I have updated the correct GOOGLE_CLIENT_ID in /app/data/env.sh with a completely different ID. Did you put the export keyword in front of the lines ? Like export GOOGLE_CLIENT_ID=xxx . You shouldn't have to do anything other than restarting .

@ajax White labeling seems to be possible but you'l have to purchase a license for that. You can find it under the super admin settings: /super_admin/settings

@moi01 you have to add it to /app/data/env.sh . See https://docs.cloudron.io/packages/chatwoot/#custom-config . Also, in the env.sh you have to add export GOOGLE_OAUTH_CLIENT_ID=xxx (export is important)

@v_shnu thanks for the fix. Merged and published the new package.

Seems the storage bloat bug was "fixed" by cutting all attachments that are >= 15.

This is set during app startup into a transient env file https://git.cloudron.io/packages/chatwoot-app/-/blob/master/start.sh?ref_type=heads#L22

This worked! Thank you

8.3.1 seems to have fixed this issue! Thanks team Cloudron!

@andreasdueren oh good idea. I have been using mistral so far. should check out claude

From a quick root, postgres requires pgvector which Cloudron's postgres already has.

And do you have any news? Any plan to support Evolution API on Cloudron? Would be very nice, because it is not just "some random piece of software", it is a piece that increases greatly the use of many Cloudron apps, including Typebot, Chatwoot and others.

@vishalw3villa you have to ask this question in the upstream Chatwoot forum - https://github.com/orgs/chatwoot/discussions . This is the Cloudron forum for Chatwoot package.

@girish said in New user emails dont work: @xboxhaxorz cloudron only supports STARTTLS for relay. So, you have to use port 587 and not port 465 . Great, its all working now Thanks

@fbartels said in Move Chatwoot: How did the move go? Well it has not happened yet. The client decided wait until other issues are resolved. But it is still on the roadmap. I will post when it comes back around.

Yes, 95$ month sorry. They does 30% if you qualify for their "startup" discount. They also have a free plan, which is free btw. And a pro plan for 25$ per month, unlimited number of agents. I've noticed they says the issue is fixed on Github, and there's another user says it's still experiencing the issue.