Suppose client call server using GET API, is it possible that server send a response but client misses that response??
If yes how to handle such situation, as I want to make sure that client receives the data. For now I am using second REST call by client as ack of first.
It is certainly possible. For example if you are using a site with a REST API and a request is just sent to the API and your internet connection dies when the answer is supposed to arrive, then it is quite possible that the server has received your request, successfully handled it, even sent the response, but your computer did not receive it. It could be an issue on a server responsible for transmitting the request as well. The solution to this kind of issue is to have a timeout and if a request timed out, then resend it until it is no longer timed out.
Related
We have 2 Windows services (same machine) that communicate on top of HTTP Protocol.
On specific machine we see the HTTP POST being sent from the client (Windows service) and arrives to the server (Windows service listening to REST CALLs) - 2 times, meaning i get 2 exact HTTP Post request on the service, but we see on client it was executed only 1 time.
Before going to wireshark/analyze the HTTP protocol, I wish to understand what explain this behavior.
When going to https://www.rfc-editor.org/rfc/rfc7231#section-4.3.3
"the origin server SHOULD send a 201 (Created) response containing a Location header
field that provides an identifier for the primary resource created"
I guess we should look in wireshark for 201 response? And if no response? Does the HTTP or network framework for my C# application is retrying the POST on the server side? because we dont see 2 requests sent from client code.
POST reply behavior
While true, more often than not the server replies with a 200-ok status code and some extra information.
Whether this is by mistake or to avoid chatty apis or some other architecture/design consideration, only the developer can tell.
So in theory you get a 201 with an identifier and then make a GET request with said identifier to retrieve details.
In practice a lot of times this does not occur. So it is not safe to assume this behavior.
Your problem
I highly doubt that there is a built in mechanism that retries post. There are plenty of reasons for that:
Duplicating entries. Imagine creating a PayPal payment. If the network has an error and you just did not receive the answer, the built in mechanism will charge you twice.
There are libraries that do that only when you are sure that the request is idempotent, that is the post contained some sort of identifier and the second request will fail.
First, the calls are HTTP GET (not POST).
We define the URL with hostname/FQDN, the solution to avoid duplicated calls was to work with ip address instead of hostname when sending the Rest API.
This is the long explanation of the problem, no root cause yet.
Used both Wireshark/Process Monitor to diag, not sure for the root cause.
Process Monitor: Filtering to display network calls
Wireshark: Filter to show only HTTP
The Client send a single HTTP Get request to:
/DLEManagement/API/Engine/RunLearningPeriod
The call was executed at 11:08:16.931906
We can see 2nd call at 11:08:54.511909 - We did not trigger.
HTTP Get executed from *Server.exe (in red) and the Server is at *Management.Webservice.exe (in red).
We see that a *Client.exe (Antivirus process, in blue) is sending TCPCopy packets in the window between we sent and received.
Also, we can see that the first request was made with APIPA IPv6 and the 2nd call is IPv4, We checked the network interface and it was disabled.
Wireshark screenshot:
Process Monitor screenshot:
Network configuration:
How is possible to handle timeouts in time consuming operations in a REST API. Let's say we have the following scenario as example:
A client service sends a request to insert a resource through a REST API.
Timeout elapses. The client thinks the insertion failed.
REST API keep working and finishes the insertion.
Client do not notify the resource insertion and it status is "Failed".
I can think I a solution with a message broker to send orders to a queue and wait until they are solved.
Any other workaround?
EDIT 1:
POST-PUT Pattern as has been suggested in this thread.
A Message Broker (add more complexity to the system)
Callback or webhook. Pass in the request a return url that the server API can call to let the client know that the work is completed.
HTTP offers a set of properties for invoking certain methods. These are primarily safetiness, idempotency and cacheability. While the first one guarantees a client that no data is modified, the 2nd one gives a promise whether a request can be reissued in regards to connection issues and the client not knowing whether the initial request succeeded or not and only the response got lost mid way. PUT i.e. does provide such a property, i.e.
A simple POST request to "insert" some data does not have any of these properties. A server receiving a POST request furthermore processes the payload according to its own semantics. The client does not know beforehand whether a resource will be created or if the server just ignores the request. In case the server created a resource the server will inform the client via the Location HTTP response header pointing to the actual location the client can retrieve information from.
PUT is usually used only to "update" a resource, though according to the spec it can also be used in order to create a new resource if it does not yet exist. As with POST on a successful resource creation the PUT response should include such a Location HTTP response header to inform the client that a resource was created.
The POST-PUT-Creation pattern separates the creation of the URI from the actual persistence of the representation by first firing off POST requests to the server until a response is received containing a Location HTTP response header. This header is used in a PUT request to actually send the payload to the server. As PUT is idempotent the server simply can reissue the request until it receives a valid response from the server.
On sending the initial POST request to the server, a client can't be sure whether the request reached the server and only the response got lost, or the initial request didn't make it to the server. As the request is only used to create a new URI (without any content yet) the client may simply reissue the request and in worst case just create a new URI that points to nothing. The server may have a cleanup routine that frees unused URIs after a certain amount of time.
Once the client receives the URI, it simply can use PUT to reliably send data to the server. As long as the client didn't receive a valid response, it can just reissue the request over and over until it receives a response.
I therefore do not see the need to use a message-oriented middleware (MOM) using brokers and queues in order to guarantee reliable messaging.
You could also cache the data after a successful insertion with a previously exchanged request_id or something of that sort. But I believe message broker with some asynchronous task runner is a much better way to deal with the problem especially if your request thread is a scarce resource. What I mean by that is. If you are receiving a good amount of requests all the time. Then it is a good idea to keep your responses as quickly as possible so the workers will be available for any requests to come.
In fiddler, how to terminate the request in between before it reaches the host. For eg I send a request and I want to put a breakpoint in that request so that I'm not able to receive the response. Basically, I want to inspect the response before it is returned to the original caller and how my service is behaving if there's a connection lost or some other termination by which the request was unable to reach the server. Any answer is highly appreciated. Sorry for any flaw, I'm a newbie in using Fiddler. :)
Fiddler offers several mechanisms for interfering with requests. If your goal is to simply kill the request without returning a response, you can create a rule with an Action of *drop or *reset rules in Fiddler's AutoResponder.
*drop will close the client connection immediately without sending a response. The closure is graceful at the TCP/IP level, returning a FIN to the client.
*reset will close the client connection immediately without sending a response. The closure is abrupt at the TCP/IP level, returning a RST to the client.
Alternatively, you can have Fiddler return any HTTP/4xx or HTTP/5xx to the client.
Lastly, you could use a breakpoint to allow you to manually manipulate the request before it's sent to the server, and/or manipulate the server's response before it's sent to the client. Use the bpu command in the QuickExec box to break on a given URL (e.g. bpu sample.asp).
I have a running system that process short and long running operations with a Request-Response interface based on Agatha-RRSL.
Now we want to change a little in order to be able to send requests via website in Json format so i'm trying many REST server implementation that support Json.
REST server will be one module or "shelve" handled by Topshelf, another module will be the processing module and the last the NoSQL database runner module.
To talk between REST and processing module i'm thinking about a servicebus but we have two types of request: short requests that perform work in 1-2 seconds and long requests that do work in 1 minute..
Is servicebus the right choice for this work? I'm thinking about returning a "response" for long running op with a token that can be used to request operation status and results with a new request. The problem is that big part of the requests must be used like sync request in order to complete http response.
I think I have also problems with response size (on MSMQ message transport) when I have to return huge list of objects
Any hint?
NServiceBus is not really suitable for request-response messaging patterns. It's more suited to asynchronous publish-subscribe.
Edit: In order to implement a kind of request response, you would need to message in both directions, but consisting of three logical steps:
So your client sends a message requesting the data.
The server would receive the message, process it, construct a return message with the data, and send it to the client.
The client can then process the data.
Because each of these steps takes place in isolation and in an asynchronous manner there can be no meaningful SLA or timeout enforced between when a client sends a request and receives a response. But this works nicely for large processing job which may take several minutes to complete.
Additionally a common value which can be used to tie the request to the response will need to be present in both messages. Otherwise a client could send more than one request, and receive multiple responses and not know which response was for which request.
So you can do this with NServiceBus but it takes a little more thought.
Also NServiceBus uses MSMQ as the underlying transport, not http.
I'm trying to upload a file onto my personal server.
I've written a small php page that works flawlessy so far.
The little weird thing is the fact that I generate all the body of the HTTP message I'm going to send (let's say that amounts to ~4 mb) and then I send the request to my server.
The server, then, asks for an HTTP challenge and my delegate connection:didReceiveAuthenticationChallenge:challenge replies to the server with the proper credentials and the data.
But, what's happened? The data has been sent twice!
In fact I've noticed that when I added the progressbar.. the apps sends the data (4mb), the server asks for authentication, the apps re-sends the data with the authentication (another 4mb). So, at the end, I've sent 8mb. That's wrong.
I started googling and searching for a solution but I can't figure out how to fix this.
The case scenarios are two (my guess):
Share the realm for the whole session (a minimal HTTP request, then challenge, then data)
Use the synchronized way to perform an HTTP connection (things that I do not want to do since it seems an ugly way to handle this kind of stuff to me)
Thank you
You've run into a flaw into the http protocol: you have to send all the data before getting the response with the auth challenge (when you send a request with no credentials). You can try doing a small round trip as the first request in the same session (as you've mentioned), like a HEAD request, then future requests will share the same nonce.
Too late to answer the original requester, but in time if somebody else read this.
TL;DR: Section 8.2.3 of RFC 2616 describes the 100 Continue status which is all what you need (were needing) in such a situation.
Also have a look at sections 10.1.1 and 14.20.
The client sends a request with an "Expect: 100-continue" header, pausing the request before sending the body. The server uses the already received headers to make its decision whether this request may be accepted or not (if the entity –the body– to be received is not too large, if the user's credentials are correct...). If the request is acceptable for the server, it replies with a "100 Continue" status code, the client sends the body and the server replies with the final status code for that request. To the contrary, if the request is not acceptable, the server replies with a 4xx status code ("413 Request Entity Too Large" if the provided body size is... too large, or a "401 Unauthorized" + the WWW-Authenticate: header) and the client does not send the body. Being answered with a 401 status code and the corresponding WWW-Authenticate: information, the client can now perform the request again and provides its credentials.