Streaming over HTTP

I'm new to streaming, so hopefully this is a quick answer
question.....due to firewall restrictions I would need to STREAM
via HTTP protocol over (80)....I can't use RTMPT (which I was able
to get working) because of protocol restrictions.....however, all
the documentation keeps hinting that you can only progressive
download via HTTP...is this true?
However, other part of the doco say "you can stream via HTTP
if you have APACHE installed". My assumption was that all this was
doing was allowing Apache to serve the http "pages" and progressive
download the files....so, long story short, can a file be streamed
using FMIS over HTTP? Thanks for any help in advance.....

No... I think the docs are just using the term "streaming" so
the author didn't have to go into a long explanation of progressive
video (to save the non-technical users from information overload
maybe). You can't truly stream over HTTP. All HTTP flv delivery is
progressive.
That said, you can deliver a streaming-like experience using
an HTTP application server. See the following:
http://flashcomguru.com/index.cfm/2005/11/2/Streaming-flv-video-via-PHP-take-two

Similar Messages

Streaming APP-V Applications Over HTTPS

Hello,
Is there any documentation on streaming APP-V 5.0 over HTTPS? I've only found one articval online and its not that great. I setup my APP-V environment with https and none of the client machines will pull the packages and in event
view it tells me the package is corrupt. I re-packaged the same application and published them and still unable to stream to any clients.
I reverted my servers back to http and reset the clients are able to stream the application that are persented to them.
All servers are on a domain.
Any imput or documentation on streaming over https would be much appreciated.
Peter B Hardy

See Thamim's post:
http://virtualvibes.co.uk/ssl-and-https-with-app-v-5-0/
Regards, Michael - http://blog.notmyfault.ch -

Streaming video over http

Hello
I need your help to understand how streaming over http works
In our company we have about 600 users who will be able to view streaming video over http hosted in Internet
We have 100 Mbps Internet link . What can we expect to happen in the network when there is contention for resources to access video on our network which delivers best effort service for video
We have enough bandwidth 4 gigs in local area network to core , so there's no real bandwidth cruch
How to differentiate or monitor bandwidth used for http streaming as against http normal traffic
Sent from Cisco Technical Support iPad App

One step closer:
Changed the crossdomain.xml file (more permissive settings).
Changed the encoding on FMLE to vp6. Working somewhat (don't know what I did to make it start streaming through hds).
But at least now I can get the individual streams in the set manifest file to work:
http://localhost/hds-live/livepkgr/_definst_/livevent/livestream1.f4m
http://localhost/hds-live/livepkgr/_definst_/livevent/livestream2.f4m
http://localhost/hds-live/livepkgr/_definst_/livevent/livestream3.f4m
BUT when I try to play the streams through the set manifest file from http://localhost/liveevent.f4m I'm getting the following error:
"The F4m document contains errors URL missing from Media tag." I'll search the forums to see if anyone else has come across this problem.
I used the f4m config tool to make the file. These are the file's contents:
<manifest xmlns="http://ns.adobe.com/f4m/2.0">
<baseURL>http://localhost/hds-live/livepkgr/_definst_/liveevent/</baseURL>
<media href="livestream1.f4m " bitrate="200"/>
<media href="livestream2.f4m " bitrate="500"/>
<media href="livestream3.f4m " bitrate="1000"/>
</manifest>
Thanks

Anybody transmit video over HTTP?

use JMF to transmit video stream over HTTP but not RTP

I'm trying to do just this at the moment, to try to implement a low bitrate videoconferencing application. It is possible. The people behind Microsoft Portrait did it successfully. Check out their white paper at http://www.acm.org/sigs/sigmm/MM2001/ep/li-demo/.
But I don't have any idea how to do it. I'm not too good at networking stuff. Any help I can get?

App-V 5.0 Streaming over SCCM 2012 SP1 - High Availability / Load Balancig

Hello Anybody,
the default App-V 5.0 Streaming (SharedContentStoreMode = on) Deployment over SCCM belongs to a HTTP-Streaming over http:\\SCCM-Server\... . The whole content flows from SCCM to the App-V-Client. Is it possible to create there a high availability solution?
Yes, we can set a second distribution point. But in th App-V-Client are always only one Distribution Source.
Is it possible to set there an clustered adress of an other IIS? Is it possible to change the publishing Distribution Source with an SMB-Path to File-Cluster? Is there a trick with Source-Root-Parameter in App-V-Client? Or are ther anything else to create
a high availibility App-V-Streaming-Solution?
Yes, with App-V-Management-Server there is a high availability solution. - But we have only SCCM-Infrastruktur for App-V-Distribution.
Thank you, best regards
Daniel

Daniel, do you found solution?

How can I know the end of TCP stream and/or SOAP over HTTP flow

Hi all!
I want to read SOAP messages over HTTP, from sniffed TCP fragments.
How can I decide that the TCP and/or the HTTP fragmenst finished?
For example there is a SOAP message sliced to 5 TCP packets, and the first packet contains the HTTP header and some SOAP content too. The last packet contains the SOAP XML's ending XML-tag </soap:Envelope>.
I don't want to watch every packet's end like "if the end string is </soap:Envelope> then it is the last packet", I just simply want to know which is the last packet of that message.
The TCP connection won't be closed after the message arrives, and let's say the HTTP header doesn't contain a Content-Length field.

A TCP connection is just a stream of bytes. It doesn't care what those bytes are. HTTP is built on top of TCP and specifies the ability to make a request without closing the connection (HTTP 1.1 Keep Alive). So you need to understand the HTTP protocol, understand whether it's a Keep Alive connection or not, and then do the same thing a browser would to do understand when a reply has been completed and the connection is available for the next request. Otherwise, you'll just look like you're getting loads of unrelated data as you sniff the connection. Oh, and you'll probably need to understand the HTTP chunking protocol too.

HTTP 1.1: Using streams over socket; How ?

Can anyone give me a short example as how to handle streams over a persistent socket-connection ?
I've trouble handling this.
I can't "reget" the stream (create a new input/output stream over an existing socket-connection).
Is there someone who can help me ?
I'm getting confused !!!

Since no one helped I could only try to find out myself:
The problem is I used some example-source that fitted well for HTTP 1.0, but led to erroneous results when applied to HTTP 1.1
It's no problem in HTTP 1.1, but you need to read ALL input (headers and possibly content) before trying to process anything:
From the headers you need to read until an empty line is discovered.... Don't just break on return or newline, this will leave an extra character in the input:
Protocol requires both \r and \n. So you always break 1 char to early when scanning bytes and then concluding on a \r that the line is over.
Then you should possibly read 'contentlength bytes' (I've not tried it out yet) in case of POST.
Flushing is enough to write all data to output.
Open and close can't work. "Regetting" a stream is not possible, so all input for this request must be taken from inputstream before trying to process the next request.

Use HTTP Dynamic Streaming (HDS) and HTTP Live Streaming (HLS) to serve live streams to clients over HTTP

I have created a live stream of a video and it gets stored in live folder.
Now i need to use HTTP Dynamic Streaming (HDS) and HTTP Live Streaming (HLS) to serve live streams to clients over HTTP, publish the streams to the HTTP Live Packager service on Flash Media Server.
So what necessary steps do I need to follow to do that ??

You need to generate a manifest file using Configurator tool and placed it under the webroot directory.
C:\Program Files\Adobe\Flash Media Server 4.5\tools\f4mconfig\configurator

ACNS 5.5 / Troubleshooting bad quality VIDEO (MMS-over-http)

Hello community,
For an unexplanable reason I have to deal with an ACNS problem which is not my domain at all ! ACNS 5.5.15 is on a WAE-512-K9 working with CDM.
Please can someone advice me on how to start the basic troubleshooting steps on the following issue :
============================
- Users in Sydney are complaining of bad quality when they watch a video from an Intranet server in Frankfurt.
Video flow has bad quality (buffering ..) and/or stops before the end of the video.
- Users browsers are configured to use the WAE as local proxy on ports 3129 (.pac file). Requested Video on the intranet is in the Windows Media format and flows are MMS-over-http.
- No multicast distribution is used, neither content pre-positioning.
============================
We did a wireshark trace on the user PC watching a video which ends up unexpectidely with error code 500 => We follow the TCP stream, Time/Sequence graph is linear until ACNS sent [PSH,ACK] tcp flags to the user PC. Then PC answers with [FIN, ACK] etc...and the stream ends.
I have checked the show-tech (i have attached some commands from the sh tech) especially TCP statistics but I do not if the values are normal or not.
I am thinkink to open a TAC case but before contacting Cisco I prefer to try some basic tests if they can help.
Any recommendation on how to start a troubleshoot or any advice would be gretly appreciated.
Thanks.
Karim

Hi Defcom,
Thanks for the quick reply. I have tried changing and resetting the bandwidth limits at both ends last week at both ends but no joy!
I am going to try the solution on the Gerry's link above later but have to wait for my brother to be free as he has an airport as well(had a baby yesterday (hence why we are setting this up)).
Dave

Post XML over HTTPS in a loop

HI,
I have a requirement to post XMLs over HTTPS. I also am supposed to retrieve the response XML, interpret it and do some processing based on the response status in the response XML. This has to be done for each XML.
Currently I am establishing a new HTTPsURLConnection for each XML, wrting the XML to output stream, retrieving the response from input stream and then releasing the connection using disconnect() method.
Is this the recommended approach or should I be creating just one connection for posting all the XMLs? Also, does java use some kind of connection pooling while creating HTTPURLConnection?
I am frequently getting "SocketTimeoutException Exception --> connect timed out" while processing large number of XMLs. Is this exception due to large number of connections that I am creating?
This program is running as a stand alone program on unix box.
Regards,
Jacob

Is this the recommended approachYes.
or should I be creating just one connection for posting all the XMLs?No.
Also, does java use some kind of connection pooling while creating HTTPURLConnection? Yes.
I am frequently getting "SocketTimeoutException Exception --> connect timed out" while processing large number of XMLs. Is this exception due to large number of connections that I am creating?Could be, especially if you're doing them in parallel.

Making a call over HTTPS with LoadVars, XML.load(), and WebService - Yes or No?

Hello, do LoadVars, XML.load(), or WebService support HTTPS-based endpoints, Yes or No?
BACKGROUND
============
I've been trying to get a LoadVars to actually make a call to an HTTPS endpoint. There is nothing in the documentation that says it can't. I know that there's also XML.load() and WebService class, but from the looks of it they don't do HTTPS.
During my tests I have absolutely no issues with making calls to the same service over HTTP. When I change it to HTTPS I don't see HTTPStatus or even failures. Also, netstat on my server will show a connection being established with the endpoint when using HTTP but not when using HTTPS. I've also tried setting SSLVerifyCertificate to "false" in my Server.xml and after a restart of AMS it doesn't help, same symptom.
I've also googled and looked through all Adobe forum posts that I can find:
https://forums.adobe.com/message/4938426#4938426
https://forums.adobe.com/thread/1661461
https://forums.adobe.com/thread/782037
https://forums.adobe.com/message/74981
https://forums.adobe.com/message/5107735#5107735
https://forums.adobe.com/message/7815#7815
https://forums.adobe.com/message/53870#53870
https://forums.adobe.com/message/87797#87797
WebService Class - http://stackoverflow.com/questions/5619776/webservice-and-fms
The best I found from the posts above is a non-commital answer from adobe staff at https://forums.adobe.com/message/4938426#4938426 and a 3rd party person saying that Webservice doesn't work at http://stackoverflow.com/questions/5619776/webservice-and-fms.
All I need is an official supported/not-supported from the Adobe staff. Shouldn't be to hard after 5 years or so of ignoring the questions in the forum right?

Adobe, please provide some details to your current and possibly potential customers, in at least one of the many unanswered posts about making HTTPS requests from AMS.
P.S.
realeyes_jun,
RealEyes Media has been an inspiration to me for many years, and I would like to thank them for their efforts to better the media streaming community.
Also, would it be possible to please release the source to REDbug?

XML over HTTP between client and server

We are trying to pass XML between a client and servlet over HTTP.
          We used the code from the StockClient/StockServlet examples as a
          starting point but cannot get it to work. Basically we
          have a simple command line java client that is trying to access
          a VERY simple servlet. When the client tries to write data into
          the output stream associated with the connection I get:
          "Connection rejected: 'Login timed out after: '15000' ms....."
          I have read several postings that instruct me to raise the
          timeout limit, but as you can see, I surely don't need 15 seconds
          to write this data out! Is there something special I need to do?
          Does this have anything to do with known issue #10065
          (http://www.weblogic.com/docs51/release_notes/rn_knownprob51.html)
          I have followed all of the instructions in the example code
          (http://www.weblogic.com/docs51/classdocs/xml.html)...
          Any assistance is appreciated...
          here is the client code:
          import java.io.*;
          import java.net.*;
          public class TestClient
          public static void main(String aa[])
          URL url = null;
          HttpURLConnection urlc = null;
          PrintWriter pw = null;
          file://Commented lines indicate other things I have tried
          try
          url = new URL("http://localhost:7001/ParserServlet");
          file://urlc = url.openConnection();
          urlc = (HttpURLConnection)url.openConnection();
          file://urlc.setRequestProperty("Content-Type", "text/xml");
          urlc.setDoOutput(true);
          urlc.setDoInput(true);
          file://urlc.connect();
          pw = new PrintWriter(new OutputStreamWriter
          (urlc.getOutputStream()), true);
          pw.println("<?xml version='1.0'?><test>testing123</test>");
          pw.flush();
          file://urlc.disconnect();
          } catch(IOException ex) {
          System.out.println(ex.getMessage());
          Here is the servlet code:
          import javax.servlet.*;
          import javax.servlet.http.*;
          import java.io.*;
          import java.net.*;
          public class TestServlet extends HttpServlet
          public synchronized void init(ServletConfig config) throws
          ServletException
          super.init(config);
          System.out.println("Inside init()");
          public final void doPost(HttpServletRequest request, HttpServletResponse
          response)
          throws ServletException, IOException
          System.out.println("Inside doPost()");
          protected void doGet(HttpServletRequest req,
          HttpServletResponse resp)
          throws ServletException,
          java.io.IOException
          System.out.println("Inside doGet()");


          Jon,
          One thing is missed in your client code. When you use HTTP POST to send request,
          you have two ways to tell the Web server when to stop reading from your input and
          to start process your input: the first one is using "Content-Lenght" header property
          to specify how many bytes you want to send to your servlet, the seocnd is use "Transfer-Code:
          Chunked" and is much more complicated. I didn't see you pass "Content-Length" in
          your client code, in which case, the Web server (Weblogic) cannot know the end of
          your request data and could keep waiting for last byte to come out or waiting for
          the socket time out (that is what you get).
          Since you use servlet, not JSP, I would recommend to code in this way (it works fine
          for me, no guranttee for your situation):
          Client code: Use a big temprary string, or StringBuffer, or StringWriter to store
          all the request data (your xml file content) before you send out the request. After
          you finish to form your XML string, calculate the number of bytes (should equal to
          the length of the string) and add the request header as
          urlc.setRequestProperty("Content-Length", bytes_length);
          I will not suggest you using PrintWriter. Think use BufferedOutputStream constructed
          from URLConnection and write the bytes (use String.getBytes()) to the servlet and
          then flush.
          Servlet code: in the doPost() of your servlet, try to find the request data length
          by calling request.getContentLength(), then open the InputStream (think to use BufferedInputStream
          for performance). Read the contents from the InputStream byte by byte and counter
          the number of bytes. Once you get the number of bytes as specified via request Content-Length,
          break your reading loop and start whatever you want.
          Hope it helps.
          "Jon Clark" <[email protected]> wrote:
          >We are trying to pass XML between a client and servlet over HTTP.
          >We used the code from the StockClient/StockServlet examples as a
          >starting point but cannot get it to work. Basically we
          >have a simple command line java client that is trying to access
          >a VERY simple servlet. When the client tries to write data into
          >the output stream associated with the connection I get:
          >"Connection rejected: 'Login timed out after: '15000' ms....."
          >I have read several postings that instruct me to raise the
          >timeout limit, but as you can see, I surely don't need 15 seconds
          >to write this data out! Is there something special I need to do?
          >Does this have anything to do with known issue #10065
          >(http://www.weblogic.com/docs51/release_notes/rn_knownprob51.html)
          >I have followed all of the instructions in the example code
          >(http://www.weblogic.com/docs51/classdocs/xml.html)...
          >
          >Any assistance is appreciated...
          >
          >here is the client code:
          >import java.io.*;
          >import java.net.*;
          >
          >public class TestClient
          >{
          > public static void main(String aa[])
          > {
          > URL url = null;
          > HttpURLConnection urlc = null;
          > PrintWriter pw = null;
          >
          > file://Commented lines indicate other things I have tried
          > try
          > {
          > url = new URL("http://localhost:7001/ParserServlet");
          > file://urlc = url.openConnection();
          > urlc = (HttpURLConnection)url.openConnection();
          > file://urlc.setRequestProperty("Content-Type", "text/xml");
          > urlc.setDoOutput(true);
          > urlc.setDoInput(true);
          > file://urlc.connect();
          > pw = new PrintWriter(new OutputStreamWriter
          > (urlc.getOutputStream()), true);
          > pw.println("<?xml version='1.0'?><test>testing123</test>");
          > pw.flush();
          > file://urlc.disconnect();
          > } catch(IOException ex) {
          > System.out.println(ex.getMessage());
          > }
          > }
          >}
          >
          >
          >
          >Here is the servlet code:
          >
          >import javax.servlet.*;
          >import javax.servlet.http.*;
          >import java.io.*;
          >import java.net.*;
          >
          >public class TestServlet extends HttpServlet
          >{
          > public synchronized void init(ServletConfig config) throws
          >ServletException
          >
          >
          > super.init(config);
          > System.out.println("Inside init()");
          > }
          >
          > public final void doPost(HttpServletRequest request, HttpServletResponse
          >response)
          > throws ServletException, IOException
          > {
          > System.out.println("Inside doPost()");
          > }
          >
          > protected void doGet(HttpServletRequest req,
          > HttpServletResponse resp)
          > throws ServletException,
          > java.io.IOException
          > {
          > System.out.println("Inside doGet()");
          > }
          >}
          >
          >
          >
          >

Mangled file downloads over http problem in 10g

I have a web app running in an OC4J stand alone 10.1.3.3 and am having a problem with downloading files over http. Its a struts2 app whose file downloading impl is easy to use and standard code for writing to an http servlet response output stream.
Using the firefox plugin for Live Headers I can see that the headers are correctly added to the servlet response and I do get the file I want. However the file has been mangled with binary output around the text. This is the case for txt, word, or any other file.
This problem does not occur in Jetty or Tomcat. I've also ruled out file corruption while going in/out of the database since I can upload a file when running oc4j, turn off oc4j, start up my app in Jetty and retrieve the same file just fine.
The mime types are all accounted for and the problem exists regardless if I use a specific content type or just application/download. My browsers (firefox and ie) also recognize all files from the content disposition value "attachment; filename=myfilename.ext". Its just the file content that some how has been wrecked on the way out of the container.
Has anyone experienced this? I only found one or two unanswered posts elsewhere.
How can this be mitigated?
Thanks in advance.
Andrew

Figured it out when I realized it was in fact the data coming from the database that was corrupt. There were some older posts on the hibernate website that pointed to a single property that needs to go in the hibernate.properties file: hibernate.jdbc.use_streams_for_binary=true. Without it, Oracle returns the Blob locator consistently 86 bytes in length and therefore bad binary.

PDF content over https in websphere

I have a servlet that streams the result of some FOP transformations back to the browser. All this works fine over http but when we switch it to https, the browser throws up an alert stating that the content is unavailable.
Can anyone out there suggest where I should look with regards to websphere settings to support pdf over https?

Hello,
I am facing some problem in downloading the pdf. I am using IE6 with SP2. Its working fine on my local box in am using http on my local machine web browser and when i download i am getting the save and open pop up but when i deploy the application on UAT environment. We are using https on UAT environment. When i am trying to download the pdf file i am getting information pop. Which needs an extra step for the user to download the file.
I dont want to do some setting on the brwoser side to disable the activex control as it will require every customer to do the changes on browser side and most corporate dont have the access to change the browser setting.
Can we send some properties in http header which can supress the information pop up over https. Please help need to solve this quick !!!
~Manoj
Edited by: mkumar1980 on Oct 28, 2008 6:26 PM

RTMP Streaming Works, HTTP Streaming Does Not

I am using the following to stream video live:
- Windows Server 2008 with Flash Media Server 5.
- Flash Media Live Encoder 3.2
- This tutorial step by step to setup the manifests: Stream live multi-bitrate video over HTTP to Flash and iOS - YouTube
- In FMLE the following settings:
     - H264 Format, Main Profile, Level 3.1, Keyframe freq 4, 150, 500 and 1000 kpbs bit rates.
     - FMS URL: rtmp://<my-static-ip>/livepkgr
     - Stream: livestream%i?adbe-live-event=liveevent
- The liveevent.f4m and liveevent.m3u8 files are saved in the webroot folder
- Built in administrator console OSMF video player, with the following URL: http://<my-static-ip>/livepkgr/liveevent.f4m
FMLE is saying it connects without issue and that it publishes, and the server shows that the stream is connecting in the admin console. The one thing that does not appear to work is playing the stream back itself. I've tried any number of different URLS to play the stream, none work, they just give me a generic "We are having problems with playback. We appologize for the inconvenience." I know the stream connects because when I replace http://<my-static-ip>/livepkgr/liveevent.f4m with rtmp://<my-static-ip>/livepkgr/liveevent it plays the rtmp version.
Is there some further configuration I am missing on the server side? Something else I am missing?

Still nothing I can verify the embedded player is trying to access the stream URL I setup. I can still verify stream publishing in the admin console, but I cannot embed the video, still getting the same generic error.. I am streaming to my private server running Windows Server 2008 R2 that I host multiple sites with. I am beginning to think the URL doesnt recognize that I am trying to access the webroot folder for the manifests, but I don't know how to tell if this is the case or how to change this setting. Can anyone provide me an example of a proper html code to embed a video using the FMLE stream I am publishing that they can verify works with a similar setup to what I have?
Dustin Rogers
[email protected]
(780) 250-0200

Streaming over HTTP

Similar Messages

Maybe you are looking for