Application status does not change from "Deploying" on Error

[rtpro]

Steps to reproduce:

1. Deploy the app in the same namespace it already exists
2. Application status changes to "Deploying"
3. Get the error below in terminal
4. app status stays in "Deploying"

application_1    | INFO:     172.21.0.1:60348 - "GET /api/v1/user/me HTTP/1.1" 200 OK
application_1    | INFO:     172.21.0.1:60356 - "GET /api/v1/organization/kubernetes-configuration HTTP/1.1" 200 OK
application_1    | INFO:     172.21.0.1:60348 - "GET /api/v1/application/list HTTP/1.1" 200 OK
application_1    | INFO:     172.21.0.1:60352 - "GET /api/v1/template/list HTTP/1.1" 200 OK
application_1    | INFO:     172.21.0.1:60352 - "GET /api/v1/kubernetes/namespace/list?context_name=jovianx-admin-jovianx-system-gke_jovianx-dev-workload_us-central1-a_jovianx-dev HTTP/1.1" 200 OK
application_1    | INFO:     172.21.0.1:60352 - "OPTIONS /api/v1/application/install HTTP/1.1" 200 OK
application_1    | INFO:     172.21.0.1:60352 - "POST /api/v1/application/install HTTP/1.1" 200 OK
task-executor_1  | INFO:procrastinate.worker.worker:Starting job application__run_pre_install_hooks[17815](application_id=15)
task-executor_1  | INFO:procrastinate.jobs:Deferred job application__install_components[17816](application_id=15)
task-executor_1  | INFO:procrastinate.worker.worker:Job application__run_pre_install_hooks[17815](application_id=15) ended with status: Success, lasted 0.087 s
task-executor_1  | INFO:procrastinate.worker.worker:Starting job application__install_components[17816](application_id=15)
task-executor_1  | ERROR:managers.applications:Failed to install application component. Failed to install chart "bitnami/redis". Release with name "redis" already exists.
task-executor_1  | ERROR:procrastinate.worker.worker:Job application__install_components[17816](application_id=15) ended with status: Error, lasted 8.525 s
task-executor_1  | Traceback (most recent call last):
task-executor_1  |   File "/home/app/hub/services/helm/subcommands/install.py", line 69, in chart
task-executor_1  |     output = await self._run_command(command)
task-executor_1  |   File "/home/app/hub/services/helm/subcommands/base.py", line 56, in _run_command
task-executor_1  |     return await run(command=command, environment=self.environment, directory=directory)
task-executor_1  |   File "/home/app/hub/utils/shell.py", line 94, in run
task-executor_1  |     raise NonZeroStatusException(
task-executor_1  | exceptions.shell.NonZeroStatusException: Error: INSTALLATION FAILED: cannot re-use a name that is still in use
task-executor_1  | 
task-executor_1  | 
task-executor_1  | During handling of the above exception, another exception occurred:
task-executor_1  | 
task-executor_1  | Traceback (most recent call last):
task-executor_1  |   File "/home/app/hub/managers/applications.py", line 286, in install_component
task-executor_1  |     return await self.helm_manager.install_chart(
task-executor_1  |   File "/home/app/hub/managers/helm/manager.py", line 86, in install_chart
task-executor_1  |     return await helm_service.install.chart(
task-executor_1  |   File "/home/app/hub/services/helm/subcommands/install.py", line 75, in chart
task-executor_1  |     raise ReleaseAlreadyExistsException(
task-executor_1  | exceptions.helm.ReleaseAlreadyExistsException: Failed to install chart "bitnami/redis". Release with name "redis" already exists.
task-executor_1  | 
task-executor_1  | During handling of the above exception, another exception occurred:
task-executor_1  | 
task-executor_1  | Traceback (most recent call last):
task-executor_1  |   File "/usr/local/lib/python3.10/site-packages/procrastinate/worker.py", line 231, in run_job
task-executor_1  |     task_result = await task_result
task-executor_1  |   File "/home/app/hub/services/procrastinate/tasks/application/install_flow.py", line 62, in install_applicatoin_components
task-executor_1  |     await asyncio.gather(*[
task-executor_1  |   File "/home/app/hub/managers/applications.py", line 298, in install_component
task-executor_1  |     raise ApplicationComponentInstallException(
task-executor_1  | exceptions.application.ApplicationComponentInstallException: Failed to install application component.
task-executor_1  | INFO:procrastinate.periodic:Periodic job application__check_health[17817](timestamp=1668857280) deferred for timestamp 1668857280 with id 17817
task-executor_1  | INFO:procrastinate.worker.worker:Starting job application__check_health[17817](timestamp=1668857280)
task-executor_1  | INFO:procrastinate.worker.worker:Job application__check_health[17817](timestamp=1668857280) ended with status: Success, lasted 2.815 s
application_1    | INFO:     172.21.0.1:39814 - "GET /api/v1/user/me HTTP/1.1" 200 OK
application_1    | INFO:     172.21.0.1:40498 - "GET /api/v1/organization/kubernetes-configuration HTTP/1.1" 200 OK
application_1    | INFO:     172.21.0.1:39814 - "GET /api/v1/application/list HTTP/1.1" 200 OK
application_1    | INFO:     172.21.0.1:40488 - "GET /api/v1/template/list HTTP/1.1" 200 OK
task-executor_1  | INFO:procrastinate.periodic:Periodic job application__check_health[17818](timestamp=1668857340) deferred for timestamp 1668857340 with id 17818
task-executor_1  | INFO:procrastinate.worker.worker:Starting job application__check_health[17818](timestamp=1668857340)
task-executor_1  | INFO:procrastinate.worker.worker:Job application__check_health[17818](timestamp=1668857340) ended with status: Success, lasted 2.792 s
task-executor_1  | INFO:procrastinate.periodic:Periodic job application__check_health[17819](timestamp=1668857400) deferred for timestamp 1668857400 with id 17819
task-executor_1  | INFO:procrastinate.worker.worker:Starting job application__check_health[17819](timestamp=1668857400)
task-executor_1  | INFO:procrastinate.worker.worker:Job application__check_health[17819](timestamp=1668857400) ended with status: Success, lasted 2.712 s

[vadim-zabolotniy]

Hi @rtpro! I have few questions to you.

1. what you actually expect when we trying to install component(Helm chart) that already exists in this namespace? Application's status should become error or deployed?
2. How you determine that application doesn't change its status? I tried to confirm that issue on current main and unable to reproduce it. I checked application in DB shell and in list of applications API endpoint in both of them application have status: error and health: unhealthy.

[rtpro]

1. Error with the helm error message in events