You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository was archived by the owner on Sep 6, 2023. It is now read-only.
Repository navigation
This repository was archived by the owner on Sep 6, 2023. It is now read-only.
Those changes were to correctly get/set the DNS info for the access point. I assume you're using your computer WiFi to connect to the AP? Do you get proper IP / routing info over DHCP? When you connect in the browser, are you using an IP or a name?
@BasedOnTechnology You're seeing the problem you are because the AP is giving out IP addresses for the "old" IP range of 192.168.4.0/24, when it should be giving out IPs from 1.1.1.0/24.
Do we want a crappy work-around for now? I was going to move the tcpip_adapter_get_dns_info call into the "read from" code block, and possibly add a small delay (in case someone calls ifconfig() and ifconfig(settings) back-to-back) for good measure.
Hi all. Is this just a race condition where the station is connecting to the AP before the AP is reconfigured?
Can that be resolved by calling ap.ifconfig() before ap.active(True)?
Hi all. Is this just a race condition where the station is connecting to the AP before the AP is reconfigured?
Can that be resolved by calling ap.ifconfig() before ap.active(True)?
No, I tested this myself, only connecting the station well after the AP was up. By not calling tcpip_adapter_get_dns_info() (note this is a get), the problem goes away. Or, if you add a short delay after calling it. By rights, that call shouldn't affect the DHCP server at all.
Okay, so the IDF folks cannot reproduce this; I doubt they're running MP. I'm trying to create a minimal recreation for them, and ... I can't reproduce it, either.
But it's definitely happening in MP.
Can someone look at the code below, and tell me if it's pretty close to matching the code flow in MP?
Activity
0xDBFB7 commentedon Nov 10, 2017
Has the AP API (heh) changed in the last little bit? I've seen some things regarding AP DNS changes, is that perhaps related?
MrSurly commentedon Nov 10, 2017
Those changes were to correctly get/set the DNS info for the access point. I assume you're using your computer WiFi to connect to the AP? Do you get proper IP / routing info over DHCP? When you connect in the browser, are you using an IP or a name?
MrSurly commentedon Nov 10, 2017
Yes, it's related. I'm investigating (my PR), though it's pretty weird; Might be an IDF bug.
MrSurly commentedon Nov 10, 2017
Opened issue with IDF: espressif/esp-idf#1257
@BasedOnTechnology You're seeing the problem you are because the AP is giving out IP addresses for the "old" IP range of 192.168.4.0/24, when it should be giving out IPs from 1.1.1.0/24.
Here's a workaround for now: https://github.com/MrSurly/micropython-esp32/tree/mp_issue_221_workaround
MrSurly commentedon Nov 10, 2017
@dpgeorge @nickzoic
Do we want a crappy work-around for now? I was going to move the
tcpip_adapter_get_dns_infocall into the "read from" code block, and possibly add a small delay (in case someone calls ifconfig() and ifconfig(settings) back-to-back) for good measure.0xDBFB7 commentedon Nov 13, 2017
@MrSurly Sorry, I didn't see this before. Thank you very much for the workaround.
nickzoic commentedon Nov 13, 2017
Hi all. Is this just a race condition where the station is connecting to the AP before the AP is reconfigured?
Can that be resolved by calling
ap.ifconfig()beforeap.active(True)?MrSurly commentedon Nov 14, 2017
No, I tested this myself, only connecting the station well after the AP was up. By not calling
tcpip_adapter_get_dns_info()(note this is a get), the problem goes away. Or, if you add a short delay after calling it. By rights, that call shouldn't affect the DHCP server at all.Changed debug pin, fixed AP mode problem micropython#221
MrSurly commentedon Dec 4, 2017
Okay, so the IDF folks cannot reproduce this; I doubt they're running MP. I'm trying to create a minimal recreation for them, and ... I can't reproduce it, either.
But it's definitely happening in MP.
Can someone look at the code below, and tell me if it's pretty close to matching the code flow in MP?