in this case : https://github.com/terraform-providers/terraform-provider-aws/blob/master/website/docs/r/iam_policy.html.markdown

oh

apparently we have to pass a dict from python

Anyone know how to create an iam.RolePolicy that includes an Output from another resource? The following fails because vpc.id is an 'output', and policy requires a 'str'

policy=
        {
            "Version": "2012-10-17",
            "Statement": [{
                "Action": [
                    "logs:PutLogEvents",
                    "logs:CreateLogStream",
                    "logs:DescribeLogGroups",
                    "logs:DescribeLogStreams"
                ],
                "Effect": "Allow",
                "Resource": 'arn:aws:logs:us-east-1:xxxxxxxxx:log-group:/aws/vpc/' + vpc.id + '/flows/*'
            }]
        })

does someone use stackReference with local backend ?

it look like all my stack override ~/.pulumi/stacks/dev.json file

i’am trying to use this : https://github.com/pulumi/examples/tree/master/aws-py-stackreference

What design patterns is everybody using for passing things like AWS region to dynamic providers? I'm using boto3 to stand up some resources not covered by the TF provider and need to make sure they're using the right region. Things I've considered: • Set the environment variable boto3 pulls from to set default region (

AWS_REGION

) • Pull from the stack config • Set a global variable • Add a 'region' parameter to every dynamic provider • Extend ResourceOptions, adding a 'region' parameter, and passing it to my provider in my resource constructor Unfortunately all of these are either have a lot of coupling or add a decent amount of boilerplate code to each provider.

not sure if this is the right room but I’m having issues getting the postgresql provider working (I’m converting from terraform). I’ve got something like this currently (disclaimer: I don’t know postgres very well) :

devRole = pulumi_postgresql.role.Role(f'developers-{i}',
                connection_limit=100,
                create_database=True,
                login=True,
                name=userlist[i],
                password='password',
                skip_reassign_owned=True,
                opts=ResourceOptions(provider=postgresProvider, depends_on=[postgresProvider]))

running results in a

PostgreSQL version: dial tcp :5432: connect: connection refused

So I assumed it was because

tf2pulumi

didn’t handle

provider "postgresql"

, so I added that and set the provider but I’m still getting the same error. is there somewhere I can find a python example of how to use postgres with pulumi?

^^^ ignore the above, I unblocked myself and made some progress. rather than pass the address from the rds I created I just request the rds instance again and use that address

I have a configuration file in which I want to replace

<BUCKET>

with a

bucket.id

.

userScript.replace('<BUCKET_NAME>', bucket.id)

causes the error

"Expected type 'str', got 'Output[str]' instead"

, which makes perfect sense. Could someone please tell me how I can tell Pulumi to execute this replacement only after the bucket resource has been created? (similar to what I usually

pulumi.Output.concat()

).

hi there. I would like to write an Output to a string. I’m in the “convert Output to str” trip. I’ve seen Output.all, concat and apply but In can’t understand how

I would like to write in a file that output

You’re on the right track with Output.apply — you can pass a function to the apply to write to the file. One thing to be aware of is https://github.com/pulumi/pulumi/issues/3991 — so you’ll probably want to export whatever it is you’re writing to the file so that the program properly waits for the resource.

Thanks @faint-table-42725 I found. Thanks for pointing to me the issue

Hi All , I am new to Pulumi. I am trying to use python to deploy my infrastructure on AWS. How do I mention resource dependencies . For example , I create a role named myLambdaRole and wanted to refer this role when i deploy my lambda. How do I say that , my lambda is dependent on this role.

Hi team I have created deb package of my code where I am using all dependency but still I am getting this error

Previewing update (pulumi-test): 
 
    pulumi:pulumi:Stack pulumi-test-pulumi-test  Traceback (most recent call last): 
    pulumi:pulumi:Stack pulumi-test-pulumi-test    File \"/root/.pulumi/bin/pulumi-language-python-exec\", line 14, in <module> 
    pulumi:pulumi:Stack pulumi-test-pulumi-test      import pulumi 
    pulumi:pulumi:Stack pulumi-test-pulumi-test  ImportError: No module named 'pulumi' 
    pulumi:pulumi:Stack pulumi-test-pulumi-test  It looks like the Pulumi SDK has not been installed. Have you run pip install? 
    pulumi:pulumi:Stack pulumi-test-pulumi-test  If you are running in a virtualenv, you must run pip install -r requirements.txt from inside the virtualenv. 
    pulumi:pulumi:Stack pulumi-test-pulumi-test  error: an unhandled error occurred: Program exited with non-zero exit code: 1 
    pulumi:pulumi:Stack pulumi-test-pulumi-test  1 error; 6 messages 
  
Diagnostics: 
  pulumi:pulumi:Stack (pulumi-test-pulumi-test): 
    Traceback (most recent call last): 
      File \"/root/.pulumi/bin/pulumi-language-python-exec\", line 14, in <module> 
        import pulumi 
    ImportError: No module named 'pulumi' 
    It looks like the Pulumi SDK has not been installed. Have you run pip install? 
    If you are running in a virtualenv, you must run pip install -r requirements.txt from inside the virtualenv. 
  
    error: an unhandled error occurred: Program exited with non-zero exit code: 1

If I go and check my python env then I am not getting any import error

root@b43ba04ea07e:/# /opt/ns/python-apps/resourcecreation/app-env/bin/python                                                                                                  
Python 3.7.4 (default, Sep  2 2019, 20:44:09) 
[GCC 5.4.0 20160609] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import pulumi
>>>

In this I am not getting this error but I have created API which will call the CLI so I am executing my code on my machine

/opt/ns/python-apps/resourcecreation/app-env/bin/python /opt/ns/python-apps/resourcecreation/app-env/lib/python3.7/site-packages/resourcecreation/app.py

Then I am getting error So Can any one tell me is it any way that while calling pulumi commands I point my python binary instead of taking machine default binary?

Hi all, not sure if this is a python sdk specific problem, but my hope is that you have some good input for me on this one. I have wrestled for a while to get my SQL database to scale properly in Azure by change the DTU or pricing tier using the

requested_service_objective_name

param with the

sql.Database

resource. However, it seems like the

import_

param mess things up by try to re-import/create the database again, even if it already exists.

sql.DatabasesClient#CreateImportOperation: Failure sending request: StatusCode=400 -- Original Error: Code="Failed" Message="The async operation failed." InnerError={"unmarshalError":"json: cannot unmarshal array into Go struct field serviceError2.details of type map[string]interface {}"} AdditionalInfo=[{"code":"0","details":[{"code":"0","message":"There was an error that occurred during this operation : '\u003cstring xmlns=\"<http://schemas.microsoft.com/2003/10/Serialization/>\"\u003eError encountered during the service operation. ; Exception Microsoft.SqlServer.Management.Dac.Services.ServiceException:Target database is not empty. The import operation can only be performed on an empty database.; \u003c/string\u003e'","severity":"16","target":null}],"innererror":[],"message":"There was an error that occurred during this operation : '\u003cstring xmlns=\"<http://schemas.microsoft.com/2003/10/Serialization/>\"\u003eError encountered during the service operation. ; Exception Microsoft.SqlServer.Management.Dac.Services.ServiceException:Target database is not empty. The import operation can only be performed on an empty database.; \u003c/string\u003e'","target":null}]

The change of the price tier and DTU works fine behind the scenes. After refreshing the resources everything works as expected. Am I doing something wrong here, or is there some way of ignoring the import during resource updates?

database = sql.Database(
        resource_name="my_database",
        name="my_database",
        resource_group_name=resource_group.name,
        server_name=sql_server.name,
        requested_service_objective_name=config.get('database.service_objective') or "S0", 
        import_={
                 "administrator_login":sqlserver_username,
                 "administrator_login_password": sqlserver_random_password.result,
                 "operation_mode":"Import",
                 "authentication_type":"SQL",
                 "storage_key":storage_account_bacpac.primary_access_key,
                 "storage_key_type":"StorageAccessKey",
                 "storage_uri":storage_blob_uri
         })

My initial thought was to check if the database already exists with

sql.get_database

and change the import block based on the result, but that seems like a messy solution?

Hi all. I would like to run test (unit test) or functional ones. I don’t find a lot of articles on that. May you suggest me something?

specific of pulumi/python

Hi all. I've contributed the first cut of an Azure example but there was a Travis build error - anything I need to do? https://github.com/pulumi/examples/pull/611

I am using aws_python how to I get the ipv6_cidr_block of my vpc and pass it into my ipv6_cidr_block for my Subnet?

Hi there! Im struggling for creating same

iam

roles that gonna be used for multiple resources (example : group of lambdas that use the same role). The problem is that the lambdas are created in a

for

loop . If I put the creation of the iam role inside as

role.arn

I get duplicates URN... Any advise/example how I should do this ?

Anybody got a good approach to replace terraform's

cidrsubnet

in pulumi?

how I can wait explicitly for an output if it's not a straight dependency o a resource? https://gist.github.com/MattBlack85/8fc884341cbc85d9dc0252286fb3c938 at line 37 I need the elastic IP value that I have to inject into the instance init script

but the only thing I get is a rendered

aws ec2 associate-address --instance-id $INSTANCE_ID --allocation-id <pulumi.output.Output object at 0x7f9fb51fd700>\n--==BOUNDARY==--\n"

What's the best way to achieve this

for

loop here?

vnet = network.VirtualNetwork('ssvc-vnet',
  resource_group_name=resource_group.name,
  name="ssvc-vnet-{0}".format(stack_name),
  address_spaces=[ssvc_vnet_space],
  subnets= [
    for subnet in subnet_config:
      {
        name=subnet[0],
        address_prefix=subnet[1],
        virtual_network_name=vnet.name
      )
      },
  ]
)

It's syntactically incorrect, so I'm wondering what the best way to get this done anyways might be.

is there any way to get the AWS region out of pulumi.Config()

?