-
Notifications
You must be signed in to change notification settings - Fork 638
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
INFLUXDB module freezes the devices #410
Comments
There was a memory leak with the host string introduced with the mDNS resolver feature in 1.12.0. I have fixed it in the dev branch (commit 7bed35f). |
Original comment by Marek Forkasiewicz (Bitbucket: fork04, GitHub: fork04): It seems to be stable now |
Released with 1.12.1 |
Original comment by Marek Forkasiewicz (Bitbucket: fork04, GitHub: fork04): Most of my INFLUXDB enabled devices are dead now |
Original comment by Marek Forkasiewicz (Bitbucket: fork04, GitHub: fork04): [INFLUXDB] Sending to server.lan:8086 Exception (9): ctx: sys
ets Jan 8 2013,rst cause:2, boot mode:(3,7) load 0x4010f000, len 1384, room 16 |
Does it have to do with the previous problem? It looks like an exception while sending a message... Do you have a steps to reproduce? |
Original comment by Marek Forkasiewicz (Bitbucket: fork04, GitHub: fork04): To reproduce this effect, You must enable Influx module and send periodically msg to influxdb and wait :) nothing more. Also I noticed something very strange , when I reflashed "frozen" esp without INFLUX included it`s booting normaly and work ok, when Influx IS included esp do not even connect to wifi and freeze again. Maybe something is stored in EEPROM partition? |
Original comment by Marek Forkasiewicz (Bitbucket: fork04, GitHub: fork04): Here is serial log from "frozen" esp and we have endless loop :) [RELAY] Saving mask: 1 Exception (9): ctx: sys
ets Jan 8 2013,rst cause:2, boot mode:(3,6) load 0x4010f000, len 1384, room 16 [INIT] ESPURNA 1.12.0 [INIT] CPU chip ID: 0xCED810 [INIT] Flash chip ID: 0x1640E0 [INIT] Flash sector size: 4096 bytes [INIT] BOARD: WEMOS_D1_MINI_RELAYSHIELD [INIT] Last reset reason: Software/System restart [TELNET] Listening on port 23 Exception (9): ctx: sys
ets Jan 8 2013,rst cause:2, boot mode:(3,6) load 0x4010f000, len 1384, room 16 |
@fork04 The problem seems to be the idbSend calls in the heartbeat method. It is called from the _mqttCallback in the mqtt.ino module, which is called when MQTT gets connected. Since the current InfluxDB implementation uses a sync client it just chokes in the callback... not a good idea. Two options here: use a flag to send the heartbeat instead of doing it from the callback (easy and safer, should do it anyway) or rework the influxDB client to be async (harder but a good idea). So I would eventually do both but will start with the flag thing to prevent the crash. |
Original comment by Marek Forkasiewicz (Bitbucket: fork04, GitHub: fork04): After many compilation attempts on windows I decided to go on virtualized Debian 9 and finally, I have stable build enviroment :) After 2h testing of newly builded 1.12.2B it seems be ok. Tomorrow I will deploy 1.12.2B on 5-6 devices for further tests. |
Good choice :) |
Originally reported by: Marek Forkasiewicz (Bitbucket: fork04, GitHub: fork04)
On my 2 different devices INFLUXDB module after 20-30 minutes of work, caused total freeze
The text was updated successfully, but these errors were encountered: