https://medplum.com logo
Join Discord
Powered by
Change Support Email Address - AWS Parameter Store...
# support
c
another in a string of self-hosting starting up issues: how do I appropriately propagate a changed email address in Parameter Store so that my self-hosted AWS project will send emails from the right address? I changed the email address in Parameter Store /supportEmail but the emails aren't going through yet. Thank you.
I think I'm gonna have to tear down the whole stack and redeploy
r
@casualgardener - is there a reason for self-hosting? Using the hosted version might offer some time savings in the short term? Curious!
d
I wonder if restarting the fargate task might help? It was a different parameter but it got it workin' for me with the allowedOrigins one.
c
Not sure of hosted Medplum's extensibility vs. self-hosted, but now that you mention that may be a good option
thanks, i'll give this a try
r
Hi @casualgardener after you make a change to the AWS paramater store, you'll need to bounce the API servers for the change to take effect. Easiest way to do this in a 0-downtime manner is to run the 
aws update-server
command: https://www.medplum.com/docs/self-hosting/install-on-aws#upgrade-the-server
To @reshma 's point though, if you'd like to share the kind of extensibility you're looking for, we can advise on whether the hosted service would be a good option for you
c
thank you!!
y
@rahul1 We actually just ran into this issue as well. We updated our parameter store and when we bounced the API servers with 
npx medplum aws update-server staging
our new urls did not update. For context, we added the oauth2 to these endpoints in our .well-known file. The authorizationUrl was added before:
Copy code
"authorization_endpoint": "https://api.kips.health/oauth2/authorize",
  "token_endpoint": "https://api.kips.health//token",
  "userinfo_endpoint": "https://api.kips.health//userinfo",
  "jwks_uri": "https://api.kips.health//.well-known/jwks.json",
Okay, got it to work But looks like these params didn't get written to the 
.well-known
file on the workflow 1. Create params in AWS 2. cdk deploy local config changes 3. 
aws update-server
However, they did get written to the file when we deployed/updated these CFN resources below (logs taken from my deployment). Seems like the parameter store changes don't actually initiate the deployment necessary to get these params into the necessary places.
Copy code
MedplumStaging | 0/6 | 9:09:16 PM | UPDATE_IN_PROGRESS   | AWS::CloudFront::ResponseHeadersPolicy          | FrontEnd/ResponseHeadersPolicy (FrontEndResponseHeadersPolicyA0E80FF4) 
MedplumStaging | 0/6 | 9:09:16 PM | UPDATE_IN_PROGRESS   | AWS::CDK::Metadata                              | CDKMetadata/Default (CDKMetadata) 
MedplumStaging | 1/6 | 9:09:17 PM | UPDATE_COMPLETE      | AWS::CDK::Metadata                              | CDKMetadata/Default (CDKMetadata) 
MedplumStaging | 1/6 | 9:09:18 PM | UPDATE_IN_PROGRESS   | AWS::ECS::TaskDefinition                        | BackEnd/TaskDefinition (BackEndTaskDefinitionAE6A5DEC) Requested update requires the creation of a new physical resource; hence creating one.
MedplumStaging | 2/6 | 9:09:18 PM | UPDATE_COMPLETE      | AWS::CloudFront::ResponseHeadersPolicy          | FrontEnd/ResponseHeadersPolicy (FrontEndResponseHeadersPolicyA0E80FF4) 
MedplumStaging | 2/6 | 9:09:20 PM | UPDATE_IN_PROGRESS   | AWS::ECS::TaskDefinition                        | BackEnd/TaskDefinition (BackEndTaskDefinitionAE6A5DEC) Resource creation Initiated
MedplumStaging | 3/6 | 9:09:20 PM | UPDATE_COMPLETE      | AWS::ECS::TaskDefinition                        | BackEnd/TaskDefinition (BackEndTaskDefinitionAE6A5DEC) 
MedplumStaging | 3/6 | 9:09:22 PM | UPDATE_IN_PROGRESS   | AWS::ECS::Service                               | BackEnd/FargateService/Service (BackEndFargateServiceD3B260C0) 
3/6 Currently in progress: MedplumStaging, BackEndFargateServiceD3B260C0
MedplumStaging | 4/6 | 9:12:44 PM | UPDATE_COMPLETE      | AWS::ECS::Service                               | BackEnd/FargateService/Service (BackEndFargateServiceD3B260C0) 
MedplumStaging | 5/6 | 9:12:45 PM | UPDATE_COMPLETE_CLEA | AWS::CloudFormation::Stack                      | MedplumStaging 
MedplumStaging | 5/6 | 9:12:46 PM | DELETE_IN_PROGRESS   | AWS::ECS::TaskDefinition                        | BackEnd/TaskDefinition (BackEndTaskDefinitionAE6A5DEC) 
MedplumStaging | 4/6 | 9:12:47 PM | DELETE_COMPLETE      | AWS::ECS::TaskDefinition                        | BackEnd/TaskDefinition (BackEndTaskDefinitionAE6A5DEC) 
MedplumStaging | 5/6 | 9:12:48 PM | UPDATE_COMPLETE      | AWS::CloudFormation::Stack                      | MedplumStaging
c
Thanks for the update @yurtric, will try this out as I still haven't got it working
r
HI @casualgardener , I just wanted to check in to see how your deployment is going?
c
Hi Rahul, I didn't manage to get it working and we've focused our dev on another part of our stack. We'll probably begin another self-hosting attempt towards June
179 Views
PreviousNext
Powered by