I have web service hosted on glassfish server. which basically translates English world to Hindi word.
suppose I post a request like close I get निकट in response but when I post a request something like 100 % off I get
The remote server returned an error: (500) Internal Server Error as a response
My glassfish server failed to handle % sign.
what should I do to overcome this error? screenshot
Use %25 for % because the percent ("%") character serves as the indicator for percent-encoded octets in Glasfish, it must be percent-encoded as "%25" for that octet to be used as data within a URI. For more information, refer link below :
http://urlencoder.urlencode.in/url/decoder/character-encoding-glassfish-url-encode
Related
I have created a new project using the template 'Building a SOAP WebService Test Plan' in jmeter (followed the steps in the tutorial from the apache jmeter website to set this up). The responses aren't being accepted and in the results log I am seeing the message:
1446205258738,20995,Soap Request,Non HTTP response code: org.apache.http.conn.HttpHostConnectException,Non HTTP response message: Connection to http://(server name).com refused,Number of Users 1-10,text,false,2273,1,1,0
It's making the connection to http rather than https which I think is why the connection is not being made. Is there anyway to change this? I've tried using the different implementations in the soap request but this hasn't worked, currently using HttpClient4.
Please see the http request details
HTTP REQUEST DETAILS
As per #Rage answer (who should answer in response not comment , I'll be happy to remove my answer if he does), just change :
Protocol [http] value to https
See:
As per your last comment, you modified to https and now you face another issue , getting a 404.
So most probably your Path or Server Name or IP are wrong, as the hoster or developer for the correct connection informations.
I am using jetty-9.2.2 with CometD-3.0.1. I am seeing below warning in my setup. It comes ~4,5 times in a day.:
2014-08-28 08:50:53.712:WARN:oejh.HttpParser:qtp607635164-15194: badMessage:
400 Illegal character for HttpChannelOverHttp#5946f125{r=1,a=IDLE,uri=-}
There is no details that can be debugged from the warning message. I have already logged a request https://bugs.eclipse.org/bugs/show_bug.cgi?id=443049 to provide detailed warning.
Meanwhile I want to know what is causing this warning? Can I ignore this or some messages are lost because of this?
Change https to http in the url.
I had the same error, then found out it's because my application did not support https, so jetty cannot recognize the https encrypted request.
Update May 2017
For Jetty 9.3+ users, you might see a log message that makes this response code more clear.
See Header parse error after upgrade to Jetty 9.3 for details.
Original Answer
The Bad Message: 400 Illegal Character can occur during parsing of a bad HTTP Request.
That is the HTTP error response that the client sees.
Some (not all) situations in which it can occur.
The EOL is not "\r\n" (CR + LF) (HTTP spec requirement)
The HTTP Method token is either not recognized or has invalid whitespace after it
The HTTP Version is not recognized or has invalid characters
HTTP Header name does not follow spec
HTTP Header value does not follow spec
This message is common on public (internet facing) servers.
You have bad HTTP requests coming in. Why?
A legitimate HTTP client has a bug
A legitimate HTTP client is not following the HTTP spec
A non HTTP client attempted to connect to your server (such as attempting to use non-encrypted HTTP on a SSL/TLS/HTTPS port, or even something as odd as an SMTP/IMAP email client attempting to talk to your HTTP port)
A malicious client is attempting to probe your system for weaknesses
This error can be caused, as it was for me, by a silly little mistake.
When testing on my localhost Jetty instance, I received a very similar 400 Illegal Character message. Then I realized why. I had simply assumed application address on my local Jetty was:
https://localhost:8080
whereas the correct address was unsecured:
http://localhost:8080
No problems after that.
Jetty is cautious about detailed error messages that include user sent data, as these can be part of an attack - even if echo'd just to a terminal.
However, we can do better and log some sanitised data. Acting on the bugzilla
Well, I met this problem because I mistook the "http://" as "https://"
I'm using libcurl (c++) library to make a request to an IIS 7.5 server. The transaction is a common SOAP webservice
Everything is working fine, my requests send an "Expect 100-continue" flag, and the server responds with a 100-continue and inmediately after that a 200 ok code along with the web service response.
But from time to time, the client receives a 100-continue message and after that, another 100 code. This makes the client report an error, as it expects a final status code right after the server 100 code. I read in W3C HTTP1.1 protocol:
An origin server that sends a 100 (Continue) response MUST
ultimately send a final status code, once the request body is
received and processed, unless it terminates the transport
connection prematurely.
The word "ultimately" makes me loose the track. Is it possible/common that a server sends several 100 codes after a final status code?
If anyone has faced this issue before, can point me to any explanation on how to handle multiple 100 response codes with libcurl?
Thanks in advance
The current spec says this on 100-continue:
The 100 (Continue) status code indicates that the initial part of a
request has been received and has not yet been rejected by the server. The server intends to send a final response after the request has been fully received and acted upon.
When the request contains an Expect header field that includes a
100- continue expectation, the 100 response indicates that the server wishes to receive the request payload body, as described in
Section 5.1.1. The client ought to continue sending the request and discard the 100 response.
If the request did not contain an Expect header field containing the 100-continue expectation, the client can simply discard this interim response.
The way I read it, it is not supposed to be more than one 100-continue response header and that's why libcurl works like this. I've never seen this (multiple 100 responses) happen and I've been doing HTTP for a while (I am the main developer of curl). To change this behavior I would expect you'd need to patch libcurl slightly to allow for this to happen.
It is not related to CURLOPT_FAILONERROR.
I suspect it's because there is an unhandled error that is not handled by the client properly. Make sure you set the CURLOPT_FAILONERROR flag.
See this SO post for more information.
I'm creating web services to make an Android app talk to a server.
I'm using Tomcat for the server and a JAX RS lib to convert data between Java objects and JSON during the transfers.
What I want is to be able to view the actual HTTP requests and responses the framework is creating.
A web browser (like Firefox with Firebug) won't do because I also need to check the request bodies generated by the Android app.
With the Servlet API I could intercept and get the request data with a filter, but not the response (I believe) after it was written by the framework.
Wireshark might be a solution (I'm trying it right now), but seems to be a bit to much. I need to worry about properly filtering my messages and didn't figure out yet how to get my HTTP messages properly formatted (plain text w/o the hexa content).
I wonder if there is a simpler way to do this from inside my application or from tomcat.
Try Tomcat's AccessLogValve. You can configure it to log incoming request headers as well as outgoing response headers. Here's an example:
<Valve className="org.apache.catalina.valves.AccessLogValve" directory="logs"
prefix="localhost_headers_access_log." suffix=".txt"
pattern="%t "%r" %s ||| %{Content-Length}o ||| %{User-Agent}i" />
Notice the %{xxx}o pattern - that's for response header. You might want to use it as %{X-JSON}o. Also notice the %{xxx}i pattern - that's for request headers.
Nest the <Valve/> element within a <Context/>, <Host/> or <Engine/> elements and you'll be all set.
I ended up using Wireshark with the following filter:
http.content_type == "application/json" || http.accept ==
"application/json"
To get the content decoded, right click on the json part of you package ("JavaScript Object Notation: application/json") > Copy > Bytes > Printable text only
I'm developing a RESTful application that integrates with other webservices.
My question is, which HTTP status should I return if my client posts data that is invalid for one of those webservices? For example, if it posts a name that is invalid for a webservice that my application uses, which of the 4** status codes should I return, considering it's a user input error?
Some considerations I've made, and why I'm not comfortable of using them:
400: The data is invalid, but not the request format itself
403: The server is not refusing to respond, although the data is invalid
406: The error is in a provided parameter, not in the "accept" header
412: The error has nothing to do with "If-Match" header
So, what would you use in this case?
Stick with 400, or have a look at 422 (which may be close to what you need).
In real life, HTTP status codes for REST and other web services can be vague and hard to clearly specify. Things also get interesting if your client is actually talking to a proxy server and that proxy sends back its own status. If there's a problem in your web service (perhaps below your app) you may just get 500.
In the past I would opt for returning 200 and using your own JSON-or-whatever structure for returning error information for your client.