On boards with PSRAM the MQTT client runs in its own task with a 5120
byte stack. Verifying an EC certificate chain against a CA, such as
Let's Encrypt ISRG Root X2, needs about 6 KB, so the task overflowed
and the board boot looped on every connect.
The espMqttClient constructors now take the stack size, and
startClient() passes EMSESP_MQTT_STACKSIZE_CA (8192) when TLS is on
with a CA, keeping the 5120 default for insecure TLS and plain MQTT.
The client is recreated when that changes. The stack usage log
reports against the size the task was created with.
uploadComplete() leaves _is_firmware / _is_filesystem set after a
successful upload. When the client then closes the connection,
handleDisconnect() sees the flags, treats it as an aborted upload and
sets the status back to NORMAL, cancelling the pending restart. The WebUI
gets away with it because its status poll turns the pending restart into
a requested one first; an upload from curl or a script is accepted
(md5_ok) but the device keeps running the old firmware until something
else restarts it.
Clear the flags once the upload has completed.
A restart requested from the WebUI (Restart button, firmware upload,
firmware update from URL, settings or customization upload) is set to
PENDING_RESTART and was only escalated to RESTART_REQUESTED when the
WebUI's SystemMonitor polled /rest/systemStatus. If the tab was closed,
nothing ever escalated it and the generic 5 minute status timeout reset
it back to NORMAL, so the restart was dropped and an uploaded firmware
was never booted.
- System::loop() now escalates a pending restart itself if it hasn't been
picked up within 10 seconds (the SystemMonitor polls every second)
- the generic status reset timeout skips PENDING_RESTART, and its timer is
cleared whenever the status is back to normal, so a start time left over
from an earlier status can't reset a later one (e.g. a pending restart
right after a firmware download) immediately
Proddy