[solved]Ping timeout's, but no reconnects

[valbuz99]

hi together.

i have some problems with a wemos D1 mini and espeasy.
setting the wifi with fix ip and force b/g (RSSI: -85db)

when i ping the device 3 of 10 pings are NOK.
on the espeasy side there are no reconnects. so some times my mqtt connection fails...

is there another way to check what's happen here?

thanks
pat

esp.PNG (24.68 KiB) Viewed 6530 times

[TD-er]

Disconnect the device and try to ping it.
Maybe there is another device on the network using the same IP?

[valbuz99]

No, it's not a ip adress problem.

BTW: The power is a dc-power supply.

[TD-er]

The RSSI of -85 can be a bit tricky sometimes. (on some nodes, no idea why some work fine with weak wifi signals and some constantly disconnect)
Can you try to rotate the node to see if the RSSI improves to something like -75 or better?
If that's possible, test to see if it may respond better to pings.

Also have a look at the timing stats to see if there are things taking long time to process. (stats are reset every time you reload that page)
Maybe if there is some blocking code that constantly takes > 100 msec to run, then that may also affect WiFi connectivity.

[valbuz99]

So...now the RSSI i have about -60db and the pings are OK!
(the router near the esp was a mesh repeater, now i configure it as a AP)

But sometimes i have retries from mqtt server (Message 37979 deleted after 11 retries) and don't know why..
perhaps you can see some errors in my timing stats.
in the att i send 3 screenshots.

thanks a lot

pat

[TD-er]

The timing stats look good, so there's not the problem with your MQTT server retries.

Since the controllers have lots and lots of parameters, I advice you to read about them first: https://espeasy.readthedocs.io/en/lates ... parameters
And maybe you should have a look at the timeout settings and/or minimum time between messages.

For example if your MQTT broker is not in the local lan, or it is running on a not so powerful computer, then you may want to increase the timeout.
100 msec is good for local network, but with these kind of issues, you may want to set the timeout a bit higher.

Also I guess the default of 10 retries is also a bit silly.
If one or two times fail, the rest will also fail, when sending to a MQTT broker.

[valbuz99]

It works now! I think it was the bad RSSI for alle the leaks

Thanks alot for help here!

pat