-->
Page 20 of 34

Re: node.dsleep() reliability - zombie mode

PostPosted: Sat Feb 21, 2015 7:24 am
by Markus Gritsch
@Derk: Would you mind attaching your code?

Re: node.dsleep() reliability - zombie mode

PostPosted: Sat Feb 21, 2015 10:04 am
by GeoNomad
epsoc wrote: The few misses are multiples of the 60 s sampling interval, so are most likely short network reachability problems (I am on Frontier, brief network 'outages' during the day are common


Thank you for all the information Derk. It is most helpful.

I have 2 ESP-01s running identical code and with the same resistor configurations.

Interestingly, one of them almost never misses an upload.

The other one often skips one and sometimes two connections. At this point, I haven't delved in to see at what point it is failing because the real test is battery life and stability, so I don't want to disturb the test, but I find it interesting that there would be such a big difference between two seemingly identical setups.

They are running 60 second interval temperature measurements with a DS18B20 on GPIO2.

Re: node.dsleep() reliability - zombie mode

PostPosted: Sun Feb 22, 2015 4:12 am
by fabix68
some development on this bug?
I tried everything but without any logical crashes. :oops:

Re: node.dsleep() reliability - zombie mode

PostPosted: Sun Feb 22, 2015 5:57 pm
by epsoc
Markus Gritsch wrote:@Derk: Would you mind attaching your code?


Markus, it's not LUA, I write my own firmware in C using the Espressif SDK. So I am not sure if it's usable for you. Besides it's really all bog standard stuff as found at many other places on the web or in the SDK examples and doesn't really add much value. Anyway, if you're interested in particular parts of it I think it would make more sense to publish that corresponding code snippet instead of the entire rather complex and confusing chunk of code. Let me know.

Derk