• Contact
  • Register
    (If you're a human, don't change the following field)
    Your first name.
    (If you're a human, don't change the following field)
    Your first name.

    Enter your email above, then click "Sign Up" to join the STAC mail list and (optionally) register to access materials on the site. Click for terms.

Posted April 10, 2015 by jeffryfisher

Log text: "nothing dequeued before timeout..."

Status: Unsolved

I'm writing the custom N1 middleware adapter for Lightfleet. My initial design spins calling its getSenderMessage function with zero wait time. The screens and logs are flooding with dequeue timeout warnings. Does anyone here know how to turn them off? If there isn't currently any config setting that'll silence those warnings, then to whom can I write about creating one?

Comments

Posted on April 10, 2015 by TomHansen

Hi Jeffry,

I can get that message commented out if you need to use 0 for the timeout.  That said, most adapters end up using STAC_INFINITE_WAIT for the timeout. Would it work to let the STAC library do the spinning or did you need to get thread of control back more often on calls to getSenderMessage?

 

Posted on April 10, 2015 by jeffryfisher

Thanks for the reply!  To get up and running, my prototype design is as simple as I can make it (single-threaded alternating its checking between  sending and receiving). I know I'll probably need to split those tasks into two threads later, but for now I can learn much from what I have (and see just how much we can do while burning only half as many electrons) ...if I can keep the log files to a reasonable size.

So is there a config setting that will tell the getSenderMessage functions (both of them) to be silent? 

Thanks!

The speed with which applications in the trading process (algorithmic "black boxes", matching engines, smart order routers, etc.) can get information from and to the network is a critical--sometimes decisive--contributor to their overall latency. 

Read more