Conversation
RajkumarGara
approved these changes
Sep 14, 2026
…drop A single lost PONG on the office WiFi used to tear down a healthy TCP connection and force a reconnect every ~40s. Add HeartbeatMonitor to count consecutive misses and only drop the socket after three in a row. Raise the PONG wait from 2s to 5s, feed the watchdog before the blocking recv, and stop late PONGs from being written to the UART as device data.
…tness # Conflicts: # src/pi/PicoScriptDeployer.py
…s, non-blocking heartbeat The reconnect backoff slept up to 30s with the watchdog fed only once per attempt, so with the Pi down the board rebooted itself on the third try instead of backing off. Add sleep_fed() and a permanent 5s socket timeout so nothing can outlast the 8.4s watchdog window; a stalled network is now a reconnect, not a reboot. Move the registration hello inside the try so a server that hangs up right after accepting causes a reconnect rather than ending the program. Replace the setblocking toggle and fixed sleep with select.poll(), which wakes the instant the server sends a command and catches hangup/error events on the same pass. Loop on send() via send_all() so a partial send cannot truncate a command. Make the heartbeat non-blocking: HeartbeatMonitor now owns the PING/PONG timing and is asked once per pass. Strip a PONG that arrives in the same chunk as a command, mirroring how PtyServer already handles a PING glued to device data. Verified with 32 unit tests and a desktop simulation of the full loop against a fake server: backoff without watchdog starvation, invalid UTF-8 relayed without disconnecting, glued PONG separated from its command, server hangup detected immediately, and reconnect after three missed PONGs.
… join The original firmware called wlan.connect() once, unguarded. On the first boot after a flash the CYW43 radio is not ready and that call throws OSError EPERM every time (seen on two boards on 2026-09-11), leaving the board frozen at the REPL. A soft reset does not clear the radio; only a hardware reset does. connect_wifi() now retries with the same capped backoff TCP uses, feeding the watchdog throughout, treats a failed join status as a failed attempt instead of waiting forever, powers the radio down between attempts, and after five consecutive failures resets the hardware itself. The TCP connect loop checks wlan.isconnected() before each attempt and rejoins WiFi first if the link has dropped mid-run. Verified in the desktop simulation: two EPERM failures then a successful join; a mid-run WiFi drop followed by rejoin and TCP reconnect; and an AP that is never found, ending in machine.reset() after five attempts with the watchdog never starved.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.