Want to test what's going on under the hood of http, so I did something like this:
socket = TCPSocket.open('localhost',4567)
request = "GET / HTTP/1.1\r\nHost:localhost\r\n\r\n"
socket.print request
But sinatra give me an error:
[2012-08-01 14:04:11] ERROR bad Request-Line `'.
ec2-184-169-164-27.us-west-1.compute.amazonaws.com - - [01/Aug/2012:14:04:11 UTC] "" 400 313
- ->
Any idea?
This might depend on web server, some web servers require some headers (I have had the same problem and solvedit by adding Accept header), and also some firewalls require User-Agent, so try this one:
socket.puts "GET / HTTP/1.1\r\n"
socket.puts "Host: localhost\r\n"
socket.puts "Accept: text/html,*/*\r\n"
socket.puts "User-Agent: Mozilla/4.0 (Windows; MSIE 6.0; Windows NT 6.0)\r\n"
socket.puts "Connection: close\r\n"
socket.puts "\r\n"
also note about space after : (it is rare) but a web server might require that!
also using puts might be better than print. (in my opinion!)
Related
From "Test in Web Chat" window in Azure Dashboard, I entered "hello" which is correctly received in my backend, but I'm still figuring out how to reply.
Note: WORDS IN CAPS means variable.
curl -d "grant_type=client_credentials&client_id=CLI-ENT-ID&client_secret=SECRET&scope=https%3A%2F%2Fapi.botframework.com%2F.default" -X POST https://login.microsoftonline.com/botframework.com/oauth2/v2.0/token
Response: {"token_type":"Bearer","expires_in":3600,"ext_expires_in":0,"access_token":"VERY_LONG_STRING"}
curl -d '{"type":"message","from":{"id":"mybot#SOME_ID","name":"mybot"},"conversation":{"id":"CONVERSATION_ID"},"recipient":{"id":"RECIPIENT_ID","name":"You"},"locale":"en","text":"hello too","replyToId":"CONVERSATION_ID|0000008"}' -H "Content-Type: application/json" -H "Authorization: Bearer VERY_LONG_STRING" -X POST https://api.botframework.com/v3/conversations/CONVERSATION_ID/activities/CONVERSATION_ID%7C0000008
Response: The page cannot be displayed because an internal server error has occurred.
Request from my app (also same error):
POST /v3/conversations/CONVERSATION_ID/activities/CONVERSATION_ID%7C0000008 HTTP/1.1
Accept: application/json
Content-Type: application/json
Authorization: Bearer VERY_LONG_STRING
User-Agent: Java-SDK
Content-Length: 273
Host: api.botframework.com
Connection: Keep-Alive
Accept-Encoding: gzip,deflate
{"type":"message","from":{"id":"mybot#SOME_ID","name":"mybot"},"conversation":{"id":"CONVERSATION_ID"},"recipient":{"id":"RECIPIENT_ID","name":"You"},"locale":"en","text":"hello too","replyToId":"CONVERSATION_ID|0000008"}
What do I miss?
Thanks to comment from Eric Dahlvang, I figure it out.
By default, ApiClient generated from swagger will have its basePath hardcoded to https://api.botframework.com.
Before replying, take the serviceUrl from incoming Activity and use it to change the basePath.
I have created some curl command to send a POST to my server where I am listening on that port for input to trigger additional action. The command is the following (Just masked the URL):
curl -v -H "Content-Type: application/json" -X POST -d "{\"Location\":\"Some Name\",\"Value\":\"40%\"}" http://example.com:8885/
I get the following output from curl:
About to connect() to example.com port 8885 (#0)
Trying 5.147.XXX.XXX...
Connected to example.com (5.147.XXX.XXX) port 8885 (#0)
POST / HTTP/1.1
User-Agent: curl/7.29.0
Host: example.com:8885
Accept: /
Content-Type: application/json
Content-Length: 40
upload completely sent off: 40 out of 40 bytes
However after that curl does not close the connection. Am I doing something wrong? Also on the server I only receive the POST as soon as I hit ctrl+c.
It sits there waiting for the proper HTTP response, and after that has been received it will exit cleanly.
A minimal HTTP/1.1 response could look something like:
HTTP/1.1 200 OK
Content-Length: 0
... and it needs an extra CRLF after the last header to signal the end of headers.
I'm a bit rusty on this, but according to section 6.1 of RFC7230, you might need to add a Connection: close header as well. Quoting part of the paragraph:
The "close" connection option is defined for a sender to signal
that this connection will be closed after completion of the
response. For example,
Connection: close
in either the request or the response header fields indicates that
the sender is going to close the connection after the current
request/response is complete (Section 6.6).
Let me know if it solves your issue :-)
Is there a question mark in link ?
I found that my link had question mark like http... .com/something/something?properties=1 and i tried header connection: close but it was still active so i tried then removing ?properties etc. and it worked...
Recently I have gotten a Raspberry Pi, and my first project I decided to build was a simple server that returns 'Hello World' to any and all clients. This is the code:
require 'socket' # Provides TCPServer and TCPSocket classes
puts 'initializing, standby'
server = TCPServer.new('localhost', 2345)
loop do
socket = server.accept
puts 'hello, this is alien'
request = socket.gets
STDERR.puts request
response = "Hello World!\n"
socket.print "HTTP/1.1 200 OK\r\n" +
"Content-Type: text/plain\r\n" +
"Content-Length: #{response.bytesize}\r\n" +
"Connection: close\r\n"
socket.print "\r\n"
socket.print response
socket.close
end
When I run that, the command line outputs initializing, standby. But then when I go to a browser and put in http://localhost:2345/anything, it returns connection refused. Changing localhost to its IP address or hostname does not work either. I have successfully SSHed to the raspberry pi. What is the problem, and how do I fix it?
Your code:
server = TCPServer.new('localhost', 2345)
means the server only listens connections from localhost. You can access it from the pi that is:
curl localhost:2345
But if you want to access the server from outside the pi, the server has to listen connections from anywhere like the following:
server = TCPServer.new('0.0.0.0', 2345)
I have a simple terminal shell command here:
curl -v -m 60 -H 'Accept-Language: en-US,en;q=0.8,nb;q=0.6' -A "Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.9.2.3) Gecko/20100401 Firefox/3.6.3" http://imeidata.net -e http://imeidata.net/iphone/model-check http://imeidata.net/iphone/model-check?sn=C8PKTRF5DTC1
When I run it for first time I get full result (Page is about 23KB)
When I run this command for second time I get only a sample page
(about 17KB)
I am still able to visit this website, so my IP is not blocked but only cURL request are denied.
Again if I change my IP.. same will happen.
Why do my requests get blocked?
Any solution will be highly appreciated.
Thank you for helping.
I would fire it with "-k" # the end of the request.
So fire your request + " -k" like below:
curl -v -m 60 -H 'Accept-Language: en-US,en;q=0.8,nb;q=0.6' -A "Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.9.2.3) Gecko/20100401 Firefox/3.6.3" http://imeidata.net -e http://imeidata.net/iphone/model-check http://imeidata.net/iphone/model-check?sn=C8PKTRF5DTC1 -k
get '/test' do
session[:my_session_id] = generate_random_id()
puts 'begin haml debug'
haml :"static/haml_page", :locals=>{:session_id => session[:my_session_id]}
end
I see in a log that a page above is constantly called twice:
begin haml debug
127.0.0.1 - - [02/Nov/2012 00:00:01] "GET / HTTP/1.1" 200 4317 1.5421
127.0.0.1 - - [02/Nov/2012 00:00:01] "GET /js/base/jquery.pjax.002902.js HTTP/1.1" 304 - 0.0234
[2012-11-02 00:00:01] WARN Could not determine content-length of response body. Set content-length of the response or set Response#chunked = true
127.0.0.1 - - [02/Nov/2012 00:00:01] "GET /css/docs.002902.css HTTP/1.1" 200 165 0.1086
.................................
begin haml debug
127.0.0.1 - - [02/Nov/2012 00:00:04] "GET / HTTP/1.1" 200 4317 1.9288
It makes me have some issues. Why is this happening?
I've moved to Puma server insted of Webrick because of similar issues.
Unfortunately I've lost example code with this problem.
In any case, if you have such problems please learn what brouser does:
Developers tool > Network (tab) will show exact sourse of request if it exists
Try to narrow this issue/bug by reducing code i.e. comment all JavaScripts, change page contents to 'Hello Wold' and observe is problem still happens
Share your code:)
Sorry for posting here, I don't know how to post this as addition to your qestion.
This is a hack, but if you really need to get it to only run the code once:
Create a global boolean variable. in the route, wrap everything in a conditional on the boolean. if false, set it true, run your code, and set it false again.