https://pulumi.com logo
#general
Title
# general
f

fast-florist-41572

10/20/2021, 1:11 PM
Is there any easy way without juggling configs/stacks to be able to deploy pulumi to something locally without it trying to store the state remotely? The example is, we have local kubernetes clusters for dev and I want to configure them with baseline services, such as ingress and databases etc. But it doesn't make sense that it has a remote stack tracking it as it is only needed for each dev locally
p

prehistoric-activity-61023

10/20/2021, 1:28 PM
not really sure what you want to achieve but if you just want to use pulumi to configure LOCAL k8s cluster, you can use local state backend
f

fast-florist-41572

10/20/2021, 1:30 PM
Thanks @prehistoric-activity-61023 that's what I suspected. Its just cumbersome as people will need to login and logout depending on what you are doing. The idea is to have consistency with deployments. Whether I deploy yaml/helm to a local env vs our staging and prod. I would prefer it either all uses pulumi or not
p

prehistoric-activity-61023

10/20/2021, 1:31 PM
do you want to deploy the same project to local envs as well as staging/prod?
I think you can specify the state backend in
Pulumi.yaml
file but I’m not sure you can do the same on stack level.
another option I’d check, I’m pretty sure it was possible to override this using env variable
so in case of local development, I’d wrap this in a shell script that would use that env var
f

fast-florist-41572

10/20/2021, 1:34 PM
Ah awesome, that could actually work nicely. Let me give that a try thanks
p

prehistoric-activity-61023

10/20/2021, 1:35 PM
so normally you will use Pulumi Service (or other remote state backend of your choice)
and in case of local development you’d use:
Copy code
PULUMI_BACKEND_URL=file://./local-state.json pulumi up -s local