I'm trying to deploy my django rest framework app on Heroku. I read many other similar questions but I'm confused. My app structure is not right or I'm missing something.
This is my structure on git:
.gitignore
requirements.txt
src
|
--authorization
--core
--static
--staticfiles
--Procfile
--manage.py
--suacm
|
---asgi.py
---settings.py
---urls.py
---wsgi.py
authorization and core are apps under my django project. there wasn't static or staticfiles before heroku deploy. But it automatically created staticfiles. Then I also created static and followed instructions to make it work via changes in settings.py. It'd be awesome if someone help me figure out my problem on heroku and why it doesn't work.
This is Procfile:
web: gunicorn suacm.wsgi
web: gunicorn suacm:app
When I run app with this command my app works fine and run locally:
gunicorn suacm.wsgi:application
But I couldn't solve the error in deployed app.
With
heroku logs --tail
I receive errors starts like below:
2021-05-21T14:32:20.000000+00:00 app[api]: Build succeeded
2021-05-21T14:32:26.505788+00:00 heroku[router]: at=error code=H14 desc="No web processes running" method=GET path="/" host=suacm.herokuapp.com request_id=dfeba2ff-fa7d-4dfb-9337-706f50d286dc fwd="82.222.237.15" dyno= connect= service= status=503 bytes= protocol=https
2021-05-21T14:32:26.882608+00:00 heroku[router]: at=error code=H14 desc="No web processes running" method=GET path="/favicon.ico" host=suacm.herokuapp.com request_id=292525ae-bc4e-49c7-b979-e455bbfd6b95 fwd="82.222.237.15" dyno= connect= service= status=503 bytes= protocol=https
When I run this
heroku ps:scale web=1 --app suacm
I get this:
Scaling dynos... !
▸ Couldn't find that process type (web).
And finally when I try to run heroku locally under src folder with this command
src % heroku local web
I get this response:
[INFO] Starting gunicorn 20.1.0
[INFO] Listening at: http://0.0.0.0:5000 (35246)
[INFO] Using worker: sync
[INFO] Booting worker with pid: 35247
Failed to find attribute 'app' in 'suacm'.
[INFO] Worker exiting (pid: 35247)
.
.
If needed, this is my requirements.txt file:
django_environ==0.4.5
djangorestframework_simplejwt==4.6.0
django_filter==2.4.0
Django==3.1.3
djangorestframework==3.12.4
environ==1.0
PyJWT==2.1.0
gunicorn==20.1.0
django-on-heroku==1.1.2
whitenoise==5.2.0
This is my first time deploying a django app. I hope I gave enough information. Just ask if there is more information needed.
Put your Procfile in the root directory, not in the src and
change this
web: gunicorn suacm.wsgi
to
web: gunicorn src.suacm.wsgi
Then why it work on a local server?
In your local setup, you are launching the server from the src directory which is not the root directory!
But the in the production Heroku launches the application from the root directory, since there is no Procfile in the root directory you are getting these errors.
Related
So I tried to host my Django application to Heroku and at first it worked. Then I wanted to add some release commands so I did this:
release_commands.sh
python manage.py makemigrations
python manage.py migrate
And in my Procfile:
release: bash ./release_commands.sh
web: gunicorn todo.wsgi --log-file -
The second line in my Procfile is unchanged, so I guessed that the error must be coming from the release commands, but that's not the case as in my heroku logs I have this:
2021-02-17T19:41:31.675215+00:00 heroku[release.6232]: Process exited with status 0
2021-02-17T19:41:31.772693+00:00 heroku[release.6232]: State changed from up to complete
2021-02-17T19:41:34.021990+00:00 app[api]: Release v11 created by user georgi.hr.dim#gmail.com
This is the error message:
2021-02-17T19:41:43.594322+00:00 heroku[router]: at=error code=H14 desc="No web processes running" method=POST path="/accounts/login/?next=/" host=essentialtodo.herokuapp.com request_id=5dce4816-6343-44c3-9c8f
-171e79379abc fwd="78.90.0.206" dyno= connect= service= status=503 bytes= protocol=https
Just found out what the problem was. For some reason after adding the release commands my dynos were turned off. To fix this: go to your heroku app (on the heroku website) -> go to resources -> press the edit button on your dyno -> turn it on and confirm
.
I have a Flask app deployed on Heroku using Gunicorn. Some processes take longer than the default 30 seconds, so I need to make use of the --timeout setting.
When I run gunicorn --workers=4 --timeout=300 app:app on my computer, there's no problem. But when I deploy it on Heroku with a Procfile consisting of web: gunicorn --workers=4 --timeout=300 app:app the app works fine but the timeout still defaults to 30 seconds.
When I check the Heroku logs, this is the error message I get after 30 seconds. Note the 503 status.
2021-01-15T01:57:01.545230+00:00 heroku[router]: at=error code=H12 desc="Request timeout" method=POST path="/api" host=counterpoint-server.herokuapp.com request_id=0865b365-e8de-48f1-bf94-6325206ec7df fwd="73.165.51.85" dyno=web.1 connect=1ms service=30000ms status=503 bytes=0 protocol=https
Any idea why my Gunicorn settings are being ignored in production?
I have a Django app which works when I run it locally heroku local. When I push my code to heroku git push heroku main it pushes without any problem, but when I opening my site it showes an 503 error in console
1 GET https://XXX.herokuapp.com/ 503 (Service Unavailable)
inside my logs I get this
at=error code=H14 desc="No web processes running" method=GET path="/" host=omylibrary.herokuapp.com request_id=lotofnumbers fwd="somenumbers" dyno= connect= service= status=503 bytes= protocol=https
from this question I assume that the problem is in ProcFile if more precisely in web. My Procfile looks like this
web: gunicorn --chdir ./Lib library.wsgi --log-file -
I also tried this
heroku ps:scale web=1
from previous answer. What Is wrong here ?
also heroku ps returns No dynos on mysite
The problem was that I named ProcFile instead of Procfile I found the solution here I have checked logs of heroku while pushing my app by terminal and I got
remote: Procfile declares types -> (none)
so if you don't have web process running probably it's because of procfile or the code inside it
I made a website with django and tried deploying it using Heroku. After deploying, I can't access to it and Heroku log shows the following error :
at=error code=H14 desc="No web processes running" method=GET path="/favicon.ico" host=pur-beurre-oc8.herokuapp.com request_id=9e33d4c8-83d6-4214-baf8-e4d1023b43ff fwd="176.134.6.63" dyno= connect= service= status=503 bytes= protocol=https
After looking through some similar problems, I tried:
heroku ps:scale web=1
and got:
Scaling dynos... !
▸ Couldn't find that process type (web).
I tried to remove and add buildpacks:
heroku buildpacks:clear
heroku buildpacks:add --index heroku/python
but this last command return to me:
Error: Expected an integer but received: heroku/python
at Object.parse (/snap/heroku/3832/node_modules/#oclif/parser/lib/flags.js:17:19)
at Parser._flags (/snap/heroku/3832/node_modules/#oclif/parser/lib/parse.js:152:49)
at Parser.parse (/snap/heroku/3832/node_modules/#oclif/parser/lib/parse.js:113:28)
at Object.parse (/snap/heroku/3832/node_modules/#oclif/parser/lib/index.js:25:27)
at Add.parse (/snap/heroku/3832/node_modules/#oclif/command/lib/command.js:83:41)
at Add.run (/snap/heroku/3832/node_modules/#heroku-cli/plugin-buildpacks/lib/commands/buildpacks/add.js:7:36)
at Add._run (/snap/heroku/3832/node_modules/#oclif/command/lib/command.js:44:31)
So I do not know what I'm doing wrong
My Proficle is like this:
web: gunicorn pur_beurre.wsgi
I tried to change Procfile like this:
web: gunicorn pur_beurre:app
or this:
web gunicorn pur_beurre:app
but heroku ps:scale web=1 still not working
Edit
My Procfile was not in my app’s root directory, so I moved it and now heroku ps:scale web=1 returns:
Scaling dynos... done, now running web at 1:Free
But I still have the error:
at=error code=H14 desc="No web processes running" method=GET path="/favicon.ico" host=pur-beurre-oc8.herokuapp.com request_id=8cc800fa-cd49-43a9-8fc5-babf0028ca1f fwd="176.134.6.63" dyno= connect= service= status=503 bytes= protocol=https
I am fighting with this error while trying to get Django app to Heroku.
Below some details:
Requirements:
dj-database-url==0.5.0
Django==2.1.5
django-heroku==0.3.1
gunicorn==19.9.0
psycopg2==2.7.7
psycopg2-binary==2.7.6.1
pytz==2018.9
whitenoise==4.1.2
WSGI:
import os
from django.core.wsgi import get_wsgi_application
os.environ.setdefault("DJANGO_SETTINGS_MODULE", "folder.settings")
application = get_wsgi_application()
Procfile tried (below are options i used in separate files/deployment):
web: gunicorn 4_Django/folder/wsgi.py 0.0.0.0:$PORT
LOGs:
Procfile logs:
2019-01-29T14:18:23.785257+00:00 heroku[web.1]: Starting process with
command `gunicorn 4_Django/folder/wsgi.py 0.0.0.0:8023`
2019-01-29T14:18:26.388317+00:00 heroku[web.1]: State changed from
starting to crashed
2019-01-29T14:18:26.307448+00:00 heroku[web.1]: Process exited with
status 3
2019-01-29T14:18:27.000000+00:00 app[api]: Build succeeded
2019-01-29T14:18:32.222748+00:00 heroku[router]: at=error code=H10
desc="App crashed" method=GET path="/favicon.ico"
host=djangozadanie.herokuapp.com request_id=7102a7b5-2798-4317-988d-
a690c10c9a14 fwd="89.78.65.77" dyno= connect= service= status=503
bytes= protocol=https
During deployment i get dozen of lines like the below, but i am not sure it has anythong to do with the problem:
Found another file with the destination path 'style.css'. It will be
ignored since only the first encountered file is collected. If this is
not what you want, make sure every static file has a unique path.
That's not how you use gunicorn. It accepts a Python module path, not a file path, and doesn't take an IP or port on the command line.
As the docs show, actually all you need is just:
web: gunicorn folder.wsgi
assuming folder is your project name.