1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
|
#!/bin/bash -e
# See "Running in Production" in the API Star documentation at
# <https://github.com/encode/apistar#running-in-production>
# In development, gunicorn can automatically reload code as you edit it. To
# enable this feature, export `GUNICORN_CMD_ARGS='--reload'` before running
# this script.
#
# See:
# * <http://docs.gunicorn.org/en/stable/settings.html>
# * <http://docs.gunicorn.org/en/stable/settings.html#reload>
# If the `PORT` environment variable is set, we're probably running on a
# platform like Heroku that expects the application to bind to it. Otherwise,
# default to port 5000. gunicorn will print the value on startup either way,
# allowing administrators and developers to determine the server's base URL.
#
# See:
# * <https://devcenter.heroku.com/articles/runtime-principles#web-servers>
# * <https://devcenter.heroku.com/articles/dynos#web-dynos>
PORT=${PORT:-5000}
# Run the application with gunicorn. Note that the API Star documentation
# recommends setting `--workers`. We omit this option: on our target platform,
# this is controlled automatically using the `WEB_CONCURRENCY` environment
# variable, which has a computed default based on the resources available at
# runtime. Falling back to gunicorn's default configuration is safe enough, as
# an administrator who discovers this is the wrong configuration can easily set
# `WEB_CONCURRENCY` to override it.
#
# See:
# * <https://devcenter.heroku.com/articles/python-gunicorn#basic-configuration>
# * <http://docs.gunicorn.org/en/stable/settings.html#workers>
#
# The API Star documentation also recommends the Meinheld worker type. However,
# deploying Meinheld to Heroku causes the application to fail on startup:
#
# Error: class uri 'meinheld.gmeinheld.MeinheldWorker' invalid or not found:
# [Traceback (most recent call last):
# File "/app/.heroku/python/lib/python3.6/site-packages/gunicorn/util.py", line 134, in load_class
# File "/app/.heroku/python/lib/python3.6/importlib/__init__.py", line 126, in import_module
# mod = import_module('.'.join(components))
# return _bootstrap._gcd_import(name[level:], package, level)
# File "<frozen importlib._bootstrap>", line 978, in _gcd_import
# File "<frozen importlib._bootstrap>", line 936, in _find_and_load_unlocked
# File "<frozen importlib._bootstrap>", line 961, in _find_and_load
# File "<frozen importlib._bootstrap>", line 205, in _call_with_frames_removed
# File "<frozen importlib._bootstrap>", line 950, in _find_and_load_unlocked
# File "<frozen importlib._bootstrap>", line 655, in _load_unlocked
# File "<frozen importlib._bootstrap>", line 205, in _call_with_frames_removed
# File "<frozen importlib._bootstrap_external>", line 678, in exec_module
# File "/app/.heroku/python/lib/python3.6/site-packages/meinheld/__init__.py", line 1, in <module>
# from meinheld.server import *
# File "<frozen importlib._bootstrap>", line 978, in _gcd_import
# File "<frozen importlib._bootstrap>", line 961, in _find_and_load
# ImportError: libpython3.6m.so.1.0: cannot open shared object file: No such file or directory
# ]
#
# The `libpython3.6m.so` library does not appear to be present. As we don't need
# to squeeze every ounce of performance out of this application, I've removed
# the Meinheld worker and opened a support ticket to get an explanation from
# Heroku.
exec gunicorn app:app \
--bind="0.0.0.0:${PORT}"
|
