What is the purpose of X-Mx-ReqToken header? - http

What is the use of X-Mx-ReqToken HTTP header? Almost all tutorials on enabling CORS in nginx whitelist the X-Mx-ReqToken header. But I can't find any information on the purpose of the header.

X-Mx-ReqToken is a header used to supply a profiler key for Mendix Runtime:
Member Data Documentation
final String com.mendix.systemwideinterfaces.core.IProfiler.PROFILER_KEY = "X-Mx-ReqToken" [static]
Basically, this is just a case of copying & pasting the same snippet over and over again until it made it to every nginx CORS tutorial out there. But unless you actually use said profiler, it's safe to omit.

Related

Extending artifactory's pypi repos with plugins

I am trying to migrate a legacy system to use artifactory. However I have two blockers:
the old scripts require PyPixmlrpc, which artifactory doesn't support
they also make use of upload_docs, not supported by artifactory's pypi implementation either
a smaller issue, the old scripts call register and they expect 200 instead of 204 http status code.
Would it be possible for me to write a plugin to implement this?
Looking at https://www.jfrog.com/confluence/display/RTF/User+Plugins I couldn't find a callback for when POST /api/pypi/<index-name> is requested.
If I can make
work for the methods we actually use, to just pretend it deployed docs and to respond with the correct status code I will be happy enough.
As you say, there is no plugin hook for the Pypi API endpoints. It would be possible to use the altResponse endpoint to customize artifact downloads, but then you would be restricted to GET requests with no request body, which is also not a good option for you.
I think the most viable approach would be to define a custom executions endpoint. With this, you can specify the acceptable method, read the body, and set your own response code and body. The main shortcoming with this is that you can't customize the path (it's always /api/plugins/execute/[execution_name]), but this can be worked around.
Execution endpoints can take params in the following form:
/api/plugins/execute/[execution_name]?params=[param_name]=[param_val]
Say your plugin takes a param path, which represents the API path your old scripts are going to call. Then you can set your base URL to /api/plugins/execute/[execution_name]?params=path=/, so that the API path is appended to the param. Alternatively, you can use nginx or another reverse proxy to rewrite the original API path to this form.
(Since you'll be using XML-RPC, I don't suppose you'll need to worry about any of this path stuff, but I'm including it anyway for completeness.)
Some issues with this approach:
Execution endpoints only allow String responses, so sending binary data in the response body might be finnicky. However, no such limitation exists with the request body.
If you need more than one request method, you'll need more than one execution endpoint. This means you'll need to use a reverse proxy to rewrite each method to a separate endpoint. Again, since XML-RPC just uses POST, this probably won't be an issue for you.
Execution endpoints can't customize response headers. Therefore, if your scripts expect a particular Content-Type or other header, you'll need to use a reverse proxy to insert it into the response.

event espresso CORS is wrong

I was hired to write a wordpress plugin which involves an ajax request to the website's eventespresso api.
I got it working fine locally (calling the live site's api from my local server), but when I activate the plugin on the live site, it throws:
Failed to load http://example.com/wp-json/ee/v4.8.36/events: The
'Access-Control-Allow-Origin' header has a value 'http://opt.local'
that is not equal to the supplied origin. Origin
'http://www.example.com' is therefore not allowed access.
My local domain is "http://opt.local", and the live site is http://example.com.
This error suggests to me that it only wants to allow access from my local setup, and not from the live site, which isn't even cross origin! Maybe I caused it to cache the wrong thing in development?
So a few more tests revealed that the cors settings are correct for everything except the specific route I need.
> curl -I "http://example.com/wp-json"
Access-Control-Allow-Origin: http://example.com
> curl -I "http://example.com/wp-json/ee/v4.8.36"
Access-Control-Allow-Origin: http://example.com
> curl -I "http://example.com/wp-json/ee/v4.8.36/events"
Access-Control-Allow-Origin: http://opt.local
I was able to make it work by using ee/v4.8.35 (a lower api patch version) but hopefully, there is a better solution.
I helped develop the EE4 REST API.
Ya it sounds like some issue where the webserver or a proxy or something is caching the Access-Control-Allow-Origin header.
There's no code in the EE4 REST API that controls that header, that's actually handled by the WP API (on which the EE4 REST API is built).
The relevant code is in wp-includes/rest-api.php in the function rest_send_cors_headers(). That calls get_http_origin(), whose value can be filtered using the filter http_origin.
So you might want to try adding something like
function my_plugin_force_correct_http_origin($http_origin) {
return 'http://example.com';
}
add_filter('http_origin', 'my_plugin_force_correct_http_origin');
that will ensure the PHP code is sending the correct Access-Control-Allow-Origin header.
If that doesn't resolve the issue, I would verify rest_send_cors_headers() is getting called at all (you could temporarily put a line like echo 'called rest_send_cors_headers!';die; inside that function to check).
If it is getting called, and my suggested filter doesn't help, you could try tagging your question with 'wordpress-rest-api'. Also, I would be curious to see if http://example.com/wp-json/ee/v4.8.36/events?limit=50 has the same problem.

How can I avoid zero-byte files on Cloudfront edge locations?

We've just discovered that one of Cloudfront's edge locations is returning a zero-byte file for one of our javascript assets. The invalidation is running right now, but I'm beginning to think this phenomenon may be the source of widespread but strangely un-reproducible bugs that our customers have been reporting for months now.
We're using Cloudfront with a Custom Origin (Nginx serving static files from an EC2 server). It would appear that with every deploy to our application that introduces new asset names (e.g. changed file version), we have a non-zero chance of one or more Cloudfront edge locations getting a 0-byte file.
Is there any way to avoid this?
Is there any way to detect this?
[sentiment redacted]
There is a very similar problem which has been discussed on the AWS forum. It seems to boil down to your server not sending a Content Length header with your custom origin.
Note the excerpt from the forum, which may be related:
Unfortunately your origin doesn't appear to provide a Content-Length
header. Without a Content-Length header CloudFront can't determine
that a truncated object was received and will cache it. If your origin
can send a Content-Length header any truncated objects will not be
cached. See the Developer Guide for more details.
Try adding the ContentLength header, that should do the trick.

Custom HTTP Headers with old proxies

Is it true that some old proxies/caches will not honor some custom HTTP headers? If so, can you prove it with sections from the HTTP spec or some other information online?
I'm designing a REST API interface. For versioning I'm debating whether to use version as a part of the URL like (/path1/path2/v1 OR /path1/path2?ver=1) OR to use a custom Accepts X-Version header.
I was just reading in O'Reilly's Even Faster Websites about how mainly internet security software, but really anything that has to check the contents of a page, might filter the Accept-Encoding header in order to reduce the CPU time used decompressing and reading the file. The books cites that about 15% of user have this issue.
However, I see no reason why other, custom headers would be filtered. On the other hand, there also isn't really any reason to send it as a header and not with GET is there? It's not really part of the HTTP protocol, it's just your API.
Edit: Also, see the actual section of the book I mention.

Best method of post processing Nginx reverse proxy response

I'm doing some researching on switching from Apache to Nginx as a reverse proxy in front of a Grails application on the backend. I'm playing around with some URL rewriting and have run into an issue with the response being sent back from my back end. I can handle the location header rewrite but I'm wondering what the best way to process the actually content is for link and such.
Is nginx_substitutions_filter the preferred method or is there another module that folks use to do content replacement in the response body?
I've thought about creating a Grails plugin to handle rendering the correct content based on additional request headers but now I'm thinking that would be best handled outside the application to allow for the most flexibility and loose coupling.
Are there any articles about best practices for doing URL rewriting/response post processing for reverse proxy scenarios?
You can use the Lua module to capture the response and manipulate it like a Lua string. Silly example to upper case the output:
res = ngx.location.capture('/some/path')
ngx.print(string.upper(res.body))
see http://wiki.nginx.org/HttpLuaModule#ngx.location.capture
If you want to replace only the headers, HeadersMore 3rd party module is great for that.
Other than that, susbstiution module seems to be the only option.
But I would suggest you make the backend return the correct page. Modifying every response uses resources and takes time.

Resources