is there a reference somewhere that describes the built-in debugging output?
for instance
pm open,type:2 0
bcn_timout,ap_probe_send_start
ap_probe_send over, rest wifi status to disassoc
state: 5 -> 0 (1)
rm 0
pm close 7
f r0, scandone
i can gather some of it course
nope :) all those are closed libs
but you can get what it means if you know enough about wifi and stuff :P
some are obvious, you all figured out the state line?
looks like disconnect to me :D
well, yeah. lol. trying to figure out why though, lol
2 lines above
beacon timeout
not enough yield or too busy interface
afterwards, it would attempt to re-connect numerous times and never succeed. i'd reset it and it'd reconnect
yeah, I have seen that too
I would think it's a bug and something is not cleared on beacon timeout
like a bug internal to the SDK? aka, something i can do nothing about?
yes
for me to do something, I need to go to China :D but @igrr could maybe have a look, or tell someone to do it
guess i could setup a watchdog in the mainloop to check association and software reset?
@romuye123
hi
http://pastebin.com/embed_js/2HNhfLr8
i am using this very simple program to generate CLOCK on 2Mhz
but when i check on oscilloscope it shows a straight line
what should be the mistake i am doing ?
hi
http://pastebin.com/embed_js/2HNhfLr8
i am using this very simple program to generate CLOCK on 2Mhz
but when i check on oscilloscope it shows a straight line
what should be the mistake i am doing ?
wow :D many things are wrong
GPIO 11 is one
then you start transaction but send nothing so how do you expect any clock to come out?
no pinMode is needed for SPI
@me-no-dev is there anyway to get lower level stats, like link layer retries?
not that I know of
LwIP is as low as you can go
@me-no-dev GPIO 11 is one, i did not get you
SPI runs on pins 12-15
11 is used by the Flash chip
yes i changed those pins
used 12,13,14,15
remove them... you do not need to set the pins at all (except the slave select pin)
SPI.begin() does it for you
let me show you my final program
do you SPI.transfer() in your loop?
yes i do
i have connected ADE7753
was not shown in the code you posted
was going step by step
@me-no-dev one more thing. I'm experiencing some severe packet lag times, like it's buffering up in the AP due to retries. I have 16 modules on my AP and randomly will have two or three that do this. Rebooting the AP or module usually fixes it, but it sometimes comes back or happens on another module. I'm thinking it could have something to do with either RF calibration or too high of a speed trying to be negotiated. Here's an example of one, with an average RSSI of -56dbm - http://pastebin.com/P7v3vCDa. You ever experiencing anything like this or have and clues as to the cause?
never had that many devices at once on my WiFi and I don;t even have that many ESPs so :D
lol. well, i've had some people experience it with just a few modules so I'm not sure if quantity comes into play. i'm just more likely to notice it. I've read posts about having to re-flash esp_init_data_default.bin when migrating to certain SDK versions. does the Arduino IDE take care of this or is it something I should look into?
I've never noticed it flashed and I don't flash that section in my tool either. Do I need to flash it directly or can I manipulate it purely in RF_PRE_INIT() {} with SDK calls?
@forkineye I recently had a class with lots of users playing around with two different Wifi routers. One didn't like this much users connecting/dissconnecting at all, failing misserably. Turning on the brand name router, all was fine. So perhaps?
@holgerlembke I'm using a Ubiqitui Nanostation locoM2. I've thought about that as well and have tried adjusting some settings on it. Lowering the max link speed seemed to help at first, but the issue started occurring again. Doing more testing now.
I have one WiFi AP router that won't even correctly do UDP broadcast; contact the company and they state it works fine; I gave them a sample (and no, it wasn't ESP specific I found out) and they just shutup and won't talk to me anymore. Sigh.
So its useless for CoAP.
doh
why did they have to make UDP so reliable with link layer acks? arrrggg!!! i may end up just going back to a nRF24L01 for this
So its useless for CoAP.
To clarify, the WiFi AP (a lynksys) I have is useless for CoAP not the ESP.
Hi guys. I am using mdns with esp hostname. So i can access my device with http://esp.local/ url. If i have multiple devices in same network what happens? Which device i will access with this url? I didnt test it because i have only one esp :(
a random one :-D
Thanks. So i will append chipId into esp.
For make it unique
fuck! using WifiClientSecure instead of WiFIClient reduces available RAM by nearly 6KB.
My code uses
WiFiServer to listen on TCP ports. When a client connects, I do WiFiClient serverClient = myServer.available(); and get the IP address of the client using IPAddress sourceAddr = serverClient.remoteIP();. This works in most cases but NOT when the connection in made by nmap. In that case, the IPAddress is 0.0.0.0, always.
Does anyone know how (=where in the code) the IP address is populated? I can see it comes from
WiFiServer._client._pcb->remote_ip.addr -- but I can't fathom how it gets there.
(I can see using WireShark that the TCP connections are different... I can describe how, if it helps.)