Hi @kkadapa! You can create a .env file and drop the same environment variables that are present in the docker-compose.yml file into it. We recommend using Docker Compose or Swarm to deploy everything, but this will work as well.
Hello again @avaidyam.
Now that we have system running on AWS, accessing dashboard we can create only surveys.
Any other acitivities we try to create, we receive an error.
Is there something we can check?
Bye
Andrea
mindLAMP 2 does not currently support any other activity types. We’ll have an update very soon that enables them.
Thank you,
when you say very soon, are you speaking about days or weeks?
Bye
Andrea
It will be weeks. If you want to use mindLAMP1, that is fully functional today but version 2 is in active development as you can see by the rapid progress!
@Andrea_Colajacomo Specifically, a minor update will be released within 2 weeks that enables the cognitive tests for use in mindLAMP v1. This update will require you to upgrade the database as well. The mindLAMP v2 release with cognitive tests will be several weeks out.
Hello All,
Sorry to bother you on a long labor day weekend. I followed all the steps outlined in this document https://www.notion.so/Deploying-the-LAMP-Platform-e281278b00c1475caee5ccf3ffaafba0 (skipped Netdata and Traefik) but receiving the following error running the LampServer on the Portainer (in Azure VM Sandbox env) as shown in the image below.
Could someone kindly help?
1 Like
Hi @kkadapa,
As far as I know, the server image you are trying to access is the old image and no longer supported. So, it might have been removed.
have you tried this one: docker.pkg.github.com/bidmcdigitalpsychiatry/lamp-server/lamp-server:notification-scheduling
1 Like
Thanks, @ninaes! I updated the stack with the given URL but that didn’t help. Apparently, when I’m trying to pull the docker image directly on to my local machine from Github pkg I ran into no basic auth creds error (please see below), I believe this might be the reason Portainer is unable to pull the image.
1 Like
oh yes, I remember that one too. Got it fixed by creating a package read-only token for github to authenticate. It’s weird it did not tell you that in the first place 
1 Like
Thanks again @ninaes!
I authenticated to Github with a personal token, got the image to my local, and then manually uploaded it to Portainer. I got the same error in the logs as mentioned above by Andrea (pls see below-attached screenshot) and then checked out the specific commit code as mentioned by Aditya https://github.com/BIDMCDigitalPsychiatry/LAMP-server/tree/9d1901af68c53567dc8d8fa8f46231e807b2cb65 built a docker image and uploaded to Portainer. Updated the stack as pointed above by Aditya
legacydb:
image: lampdevsqlsrv:latest
volumes:
- /home/azureuser/data/:/var/opt/mssql
both the servers are up and running but getting the same error FAILED TO CONNECT TO DB. Here is my latest lamp.yml https://gist.github.com/kkadapa/a362180cc174d1a7d045a7c11e656c12
I tried accessing this link mentioned by Prof Torous https://www.notion.so/Cloud-Deployment-Infrastructure-ebf62b0a24614d0c9b22a5817fe3e803#b987a61cc9db4cc9881dc7b925f8f04a but unfortunately, I can’t seem to access even after creating an account.
1 Like
no worries @kkadapa
you need to manually set up the database by running the following commands:
[first, enter the container running the database: $docker exec -it CONTAINERID /bin/sh and copy the db_setup.sql there], then the follwoing:
PATH="/opt/mssql/bin/:/opt/mssql-tools/bin/:${PATH}"
sqlservr --accept-eula & sleep 10
sqlcmd -S localhost -U SA -P $SA_PASSWORD -q “ALTER LOGIN [sa] WITH NAME = [$SA_USERNAME]”
sqlcmd -S localhost -U $SA_USERNAME -P $SA_PASSWORD -i /db_setup.sql
Also, I found that there is a race condition, i.e., api-server starts before database-server. You would just need to restart your server once or twice!
1 Like
As a quick note – please use the master branch or latest image, actually. The notification-scheduling branch or others are not ready for use. 
Hello @avaidyam
using the master branch we receive the original error
/usr/src/app/node_modules/nano/lib/nano.js:62
assert.ok(/^https?:/.test(cfg.url), ‘url is not valid’)
Bye
Hi @Andrea_Colajacomo and @ninaes, I’ve created a new branch specifically to mark the stable release that continues to use SQL Server. This branch will not be updated; migrating the data from SQL Server to CouchDB needs to be done manually right now. Once the master (production) branch is updated with the new CouchDB version I will reach out to both of you regarding migration.
Hello. We have the same issue but this link does not allow access to my account. What should I do?
We were doing everything as was said in the deployment instructions, yet we find ourselves being unable to run lampmind.
We have an url issue I believe. So I want to ask if it is parsed correctly. In lamp.yml example there is " ‘Host(example.uri.com)’ " or something along these lines. should the format remain the same? I mean: ‘Host([url here])’ ? Or should we try to find problem somewhere else?
Hi @sakac,
Yes, the format will be the same. You will replace the example with your names.
Hope this helps!
Rebecca
Okay, me again.
We have been trying to start the lamp_server service with a big help from our server administrator. Still have not moved past the “bad url” in tests.
I would like to request page owner, whoever it may concern, if he/she could give me access to the notion link where the details are discussed.
Or alternativelly I could provide more information about our problem and steps we took to the phase we are right now.
Also thanks to @avaidyam and @rbilden for help, but it turned out to not be a problem I thought we had.