TCPServer + Spark access issue

Troubleshooting
1

I am having issues with both (TCPserver + Spark access) enabled and use in same code.
Once below code is called in setup(), all the Spark access (functions, variable) will not work.

static TCPServer server=TCPServer(23); // telnet defaults to port 23

    server.begin();                 // start listening for clients

I have uploaded my test code into following
https://www.spark.io/build/53f6be292367b10a250009a6

2

Hi @Dilbert

I don’t think anyone but the Spark team can access your build web IDE from that link, so if you could pop the code into a github or gist, it would be easier for everybody to see.

After talking to @Dilbert, I tried adding a Spark.variable() to the default webduino sample app, and that behaves strangely too. I can get it to work once, but then sometimes it just continues breathing cyan and the web server is non-responsive and the cloud thinks the core is offline.

I am hoping @mdma or @BDub could take a look at this test case (Sorry-I am busy with other projects right now). There are a number of small but persistent problems where using the cloud features somehow knocks other code over in ways that I don’t understand–maybe this is a good one to debug since it is fairly repeatable.

// This #include statement was automatically added by the Spark IDE.
#include "WebServer/WebServer.h"

/* This creates an instance of the webserver.  By specifying a prefix
 * of "", all pages will be at the root of the server. */
#define PREFIX ""
WebServer webserver(PREFIX, 80);

int count = 0;  //Spark variable counts accesses

/* commands are functions that get called by the webserver framework
 * they can read any posted data from client, and they output to the
 * server to send data back to the web browser. */
void helloCmd(WebServer &server, WebServer::ConnectionType type, char *, bool)
{
  digitalWrite(D7,HIGH);
  /* this line sends the standard "we're all OK" headers back to the
     browser */
  server.httpSuccess();

  /* if we're handling a GET or POST, we can output our data here.
     For a HEAD request, we just stop after outputting headers. */
  if (type != WebServer::HEAD)
  {
    /* this defines some HTML text in read-only memory aka PROGMEM.
     * This is needed to avoid having the string copied to our limited
     * amount of RAM. */
    P(helloMsg) = "<h1>Hello, World!</h1>";

    /* this is a special form of print that outputs from PROGMEM */
    server.printP(helloMsg);
    server.print(micros());
  }
  count++;
}

void setup()
{
    
  pinMode(D7,OUTPUT); //use the LED to indicate a request
  digitalWrite(D7,LOW);
  
  Spark.variable("Count", &count, INT);

  /* setup our default command that will be run when the user accesses
   * the root page on the server */
  webserver.setDefaultCommand(&helloCmd);

  /* run the same command if you try to load /index.html, a common
   * default page name */
  webserver.addCommand("index.html", &helloCmd);

  /* start the webserver */
  webserver.begin();

  
}

void loop()
{
  char buff[64];
  int len = 64;
  
  digitalWrite(D7,LOW);

  /* process incoming connections one at a time forever */
  webserver.processConnection(buff, &len);
}
3

Just wanted to acknowledge that I’ve added this to my backlog of things to do! Ping me again if anyone has any updates.

1 Like
4

Just wondering, is there a known issue list which had been confirmed by Spark member ?
I had been posting number of issues but sometime confusing whether it is already known issues of just my own mistake. If there is a list which link to reported issues, it will be great.

Especially, when the forum is getting bigger everyday. Searching for a right answer if getting more difficult as some may be too long and some are not related.

It would be great if there is a status column which can indicate (open, in progress, confirm bug, feature, resolved) just like bug tracking for public.

5

Everything is listed in https://github.com/spark/firmware/issues

I don’t think any of the Elites keep track of such issues but we do our best to help resolve as many as we can on top of our own free time

If you want it to be known to Spark, be sure to open an issue at github. The TCP issue has like 2-3 tickets opened already. :wink:

1 Like
6

Hi @Dilbert,

@kennethlimcp is totally right, you can search for known issues and their status on the various github pages for the projects, as well as our new support page ( http://support.spark.io ) which can help you find official answers to commonly asked questions. Also @BDub is officially on our firmware team :slight_smile:

Thanks!
David

2 Likes
7

OK, I just entered 2 issues which I have into GitHub for Spark team to keep track. Hope you will have solution soon :smile:

If I have feature request, should I enter in there too ?

1 Like
8

@BDub - any updates on this? Networking is one of the things I’m working on this this sprint!

9

Sorry Mat, I never got back to this one. I put it on my personal backlog and those items just seem to keep stacking up lately :wink: