我在Heroku上运行了一个烧瓶应用程序。直到昨天一切正常。 现在由于某种原因,Web应用程序不断启动工作人员:
2017-09-28T07:02:06.869090+00:00 app[web.1]: [2017-09-28 07:02:06 +0000] [3882] [INFO] Booting worker with pid: 3882
2017-09-28T07:02:07.308558+00:00 app[web.1]: [2017-09-28 07:02:07 +0000] [3899] [INFO] Booting worker with pid: 3899
2017-09-28T07:02:07.626838+00:00 app[web.1]: [2017-09-28 07:02:07 +0000] [3901] [INFO] Booting worker with pid: 3901
2017-09-28T07:02:07.906617+00:00 app[web.1]: [2017-09-28 07:02:07 +0000] [3909] [INFO] Booting worker with pid: 3909
2017-09-28T07:02:08.701473+00:00 app[web.1]: [2017-09-28 07:02:08 +0000] [3926] [INFO] Booting worker with pid: 3926
2017-09-28T07:02:08.890857+00:00 app[web.1]: [2017-09-28 07:02:08 +0000] [3928] [INFO] Booting worker with pid: 3928
2017-09-28T07:02:09.543447+00:00 app[web.1]: [2017-09-28 07:02:09 +0000] [3936] [INFO] Booting worker with pid: 3936
2017-09-28T07:02:10.724376+00:00 app[web.1]: [2017-09-28 07:02:10 +0000] [3946] [INFO] Booting worker with pid: 3946
2017-09-28T07:02:11.771921+00:00 app[web.1]: [2017-09-28 07:02:11 +0000] [3962] [INFO] Booting worker with pid: 3962
2017-09-28T07:02:11.830635+00:00 app[web.1]: [2017-09-28 07:02:11 +0000] [3963] [INFO] Booting worker with pid: 3963
当然现在应用程序无法正常工作,发送给它的任何请求都会超时。 这是Procfile:
web: gunicorn app:app
还有requirement.txt:
scipy
numpy
flask
gunicorn
pillow
tensorflow
当我使用以下命令在本地运行时,该应用程序正常工作:
heroku local web
我已经向Heroku提交了一张票,但是如果有人可以在这里提供帮助,那将节省我很多时间。
由于