While it’s loading, do you get an empty new line added every few seconds?
If so, just try to hit the reset button on the Core and wait a few more seconds, then the message should pop up - it does with mine
Hi,
Brian and ScruffR. FYI, this is not for fun. I am looking for stable device for a few applications. If Core is not stable should I look at Photon instead? I wish I can share more information with you but not in such a pubic forum.
@Kiminator, we can ensure you, we are not doing this for fun either, but we want to help you and we are convinced, that the Core has proven to be a considerably stable device.
For example, I’ve frequently tested my suggestion from above, and it worked each time I tried.
I have just flashed the example code Brian has suggested above. And it does actually do what I expected - each time I tried it - and I’m sure Brian is getting the same results.
So I’d rather suspect that there must be something wrong at your own end.
Sure, it could be your Core hardware, but more likely are some other reasons, and we were trying to pinpoint them, together with you.
At the moment I’m not even convinced that your Core can propperly speak to the cloud (since you don’t get the publish events).
Hi @ScruffR
I am glad to hear what you just wrote. As I was testing Core since Xmas of this year. My first one had trouble with WiFi Credential etc… but that is something I can avoid.
As well, here are many other things I would like to discuss with you both, if you are from Spark… I mean business wise. Again, how can we exchange information in a more private way? As the applications I have now are highly confidential.
I am sure you guys are smart and can find the problem I am facing with this Core.
PS: The Chrome is still trying to load the firmware information from my Core… this is since 1 hour+ ago…
Thanks.
I myself am not on the Spark pay roll 
But you could always PM one of the guys like @harrisonhjones, @Dave, @AndyW.
Hi @Kiminator
I don’t work for Spark either–the “elites” are forum administrators and advisors, not employees. For real business issues I would recommend you PM @Dick. You can PM by clicking on a person’s at-name and then clicking the blue Private Message button. Email to hello@spark.io would also work. Spark has a lot of customers in the maker/hobbyist category, but they also have a growing number of corporate customers, so they are set up for both.
Just for the record, I have a bunch of Spark core that work fine for me doing things like scraping web page data and displaying it on displays, sending sensor information to the cloud, getting the time of day over NTP etc. It is a pretty small micro-controller so you have to program defensively to avoid crashes but once you get used to it, it is great!
1 Like
Hi! I was summoned! Only took me 14 minutes to fly here.
3 Likes
I must say I am impressed by the enthusiasm of @ScruffR and @bko in supporting user like me in this forum. And Dick glad that you join.
My background is, in short, in M2M businesses - right now there is a few growing needs from customers for various applications - Dick. Where are you based? I am in Montreal, Quebec, Canada. Would like to write you to discuss further.
It is now Friday night here… and I have a heck of a week - so I would take some time off now… should be with family too… 
Thanks… I hope you guys will help me going through this Deep Upgrade … soon.
1 Like
In addition, if you send an email to hello@spark.io ill make sure it gets in front of the right person. You can send more confidential information to that address, only spark employees have access to it.
1 Like
to this, you can always email us and we try to get back within a day or so if we can. This is generally the best start for business inquires as well. 
( hello@spark.io )
Thanks,
David
@Kiminator, I am an Elite but located in Ottawa if you should need “closer” help. 
Hi @Dave,
I will be writing something to Dick on Monday. Thanks.
Just to close this topic, I am working with Kiminator to resolve his issue. Basically, the problem is due to the fact that client.flush() does not work as expected and client.read() must be used to fully flush the incoming data buffer or else the CC3000 buffers will overflow causing chaos.
2 Likes
Well damn, that actually answers a thing that’s been troubling me this weekend. Wondered why flush() wasn’t living up to my expectations. Thanks @peekay123!
2 Likes
Hi @peekay123
So, should I code client.flush() before connect to prevent this to happen?
Would you mind adding an issue to the firmware repo describing the problem with flush() and and the current workaround?
1 Like
@Kiminator, did you test the example I emailed to you? I would say for now you can create you own “flush” function that reads all available return bytes from the server using client.available() and client.read() as in the example I sent you.
@mdma, I’ll log a issue but which repo did you want me to tag this against
(HAL)?
@mdma, I posted the issue in the HAL branch
Hi @peekay123,
I did not see anything from you.
But I reflashed the Core… and recoded my script to read then flush the “client”… and my last 2 attempts it works now.
Thanks.
1 Like
@Kiminator, that’s great news!
1 Like
Hi Paul,
It is working fine now… I am able to advance with my server testing… thanks a lot!
By the way what is your email? Want to send you NDA and see if we can work together.
1 Like