dmytrostriletskyi / heroku-load-balancer Goto Github PK

Process exited with status 1
2021-09-02T15:37:48.338258+00:00 app[web.1]: Traceback (most recent call last):
2021-09-02T15:37:48.338267+00:00 app[web.1]: File "src/entrypoint.py", line 61, in
2021-09-02T15:37:48.338268+00:00 app[web.1]: cli()
2021-09-02T15:37:48.338268+00:00 app[web.1]: File "/usr/local/lib/python3.6/dist-packages/click/core.py", line 764, in call
2021-09-02T15:37:48.338268+00:00 app[web.1]: return self.main(*args, **kwargs)
2021-09-02T15:37:48.338269+00:00 app[web.1]: File "/usr/local/lib/python3.6/dist-packages/click/core.py", line 717, in main
2021-09-02T15:37:48.338269+00:00 app[web.1]: rv = self.invoke(ctx)
2021-09-02T15:37:48.338269+00:00 app[web.1]: File "/usr/local/lib/python3.6/dist-packages/click/core.py", line 1137, in invoke
2021-09-02T15:37:48.338270+00:00 app[web.1]: return _process_result(sub_ctx.command.invoke(sub_ctx))
2021-09-02T15:37:48.338270+00:00 app[web.1]: File "/usr/local/lib/python3.6/dist-packages/click/core.py", line 956, in invoke
2021-09-02T15:37:48.338270+00:00 app[web.1]: return ctx.invoke(self.callback, **ctx.params)
2021-09-02T15:37:48.338270+00:00 app[web.1]: File "/usr/local/lib/python3.6/dist-packages/click/core.py", line 555, in invoke
2021-09-02T15:37:48.338270+00:00 app[web.1]: return callback(*args, **kwargs)
2021-09-02T15:37:48.338271+00:00 app[web.1]: File "src/entrypoint.py", line 53, in create_load_balancer
2021-09-02T15:37:48.338271+00:00 app[web.1]: identifier=pipeline_identifier,
2021-09-02T15:37:48.338271+00:00 app[web.1]: File "/heroku-load-balancer/src/heroku.py", line 62, in by_pipeline_identifier
2021-09-02T15:37:48.338271+00:00 app[web.1]: if application.get('stage') == 'production':
2021-09-02T15:37:48.338272+00:00 app[web.1]: AttributeError: 'str' object has no attribute 'get'

It looks like redirects do not work as they redirect you to http://main:8001/login/?next=/dashboard/ for example

I got this problem on heroku can you please help

Hi,

first of all thank you for this package it's exactly what I was looking for for my apps.

I need to specify a ratelimit and custom error pages in the load balancer.

I succeeded in setting up the ratelimit in the "constants.py" file.

`limit_req_zone $binary_remote_addr zone=limit:20m rate=10r/m;

server <
    listen {port};

    location / <

        limit_req zone=limit burst=5 nodelay;
        limit_req_status 429;

        proxy_pass http://main;
        proxy_next_upstream error timeout invalid_header http_502 http_503 http_504;
        proxy_set_header X-Real-IP $remote_addr;
        proxy_set_header Host $http_host;
        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header X-Forwarded-Proto $scheme;
        proxy_set_header X-Forwarded-Host $host;
    >`

On the other hand, after several tries I can't display my specific page if the ratelimit is reached.

In my Nginx configuration I had this

error_page 429 =429 /error_429.json; location = /error_429.json { root <%= ENV['HOME'] %>; }

==>

error_page 429 =429 /error_429.json; location = /error_429.json < root /; >

I tried several things but I think it is the root that is not working. How to get the variable ENV HOME from Heruku so that it points where it should.

I added the file error_429.json in the src folder and in the root of the project.

Sorry I'm not comfortable with Python at all.

Thanks

let's say I wanna balance between two apps, is it possible to have a 90% / 10% balance ?
(or is your code always balancing at equal rate 50 / 50 ?)

From the Heroku documentation,
HTTP Routing:

So scaling an app’s capacity to handle web traffic involves scaling the number of web dynos:

$ heroku ps:scale web+5

A random selection algorithm is used for HTTP request load balancing across web dynos

Stated another way: when you scale up a Heroku app to multiple dynos (server instances), the platform automatically routes incoming requests using a randomized load-balancing algorithm.

So, what problem does this project solve? It seems like this would make your Heroku app more complex & brittle, harder to maintain, and more difficult to onboard new developers.