No connection to spark

[uk_brewer]

I’m still having connection issues, updated and flashed spark

https://termbin.com/vnw2

[uk_brewer]

Was just in the basement. Spark is rebooting again.

[Bob_Steers]

We’re currently testing a fix for the spark crashing when it’s nearly full. Can this be said about your Spark?

Your last log truncated halfway through your block output, so I can’t use that to verify.

[uk_brewer]

What do you mean by the spark being full?

[Bob_Steers]

Persistent memory being (almost) full. This happens at 50+ blocks or so.

[Elco]

Actually, no. This was a bug in RAM allocation. The out of memory handler caused a hangup of the thread manager. This is now fixed.

If you still have reboots with the latest update, can you provide an EEPROM dump so I can make an exact copy of your Spark to find the cause?
Instructions in this thread:

Updated to 4/06 release, now spark is in a SOS reboot loop

Can you try giving me aan EEPROM dump, so I can get an exact copy of your Spark? docker pull brewblox/firmware-flasher:edge docker run --privileged -it --rm -v /dev:/dev brewblox/firmware-flasher:edge trigger-dfu docker run --privileged -it --rm -v /dev:/dev -v /home/pi/brewblox/:/dump -v /dev:/dev brewblox/firmware-flasher:edge -c "dfu-util -d 2b04:d006 -a 0 -s 0x800C000:0x20000 -U /dump/eeprom.bin" curl https://bashupload.com/eeprom.bin --data-binary @/home/pi/brewblox/eeprom.bin

[uk_brewer]

Elco,

I’ve just removed all blocks and Brewblox and done a new install. I’ll keep you posted.

Thanks.

[uk_brewer]

Elco,

Followed instructions but not sure what or where the dump is saved. Also, still getting random issues of the spark not connecting to the service with spark now connected by usb.

When the UI is working noticed units showing °C even though I have °F selected.

[Bob_Steers]

The dump is uploaded by the last command. It should’ve printed a line starting with “wget https://bashupload.com/…”.

If you go to the quick action editor, does it still show degC?

[uk_brewer]

Bob/Elco,

I did the reset and force dfu on the spark yesterday, so far its been up for 19 hours. Spoke too soon, just lost it again.

[Elco]

Share your EEPROM dump if you want us to look at it.

[uk_brewer]

Elco,

https://bashupload.com/HaelU/eeprom.bin

Thanks.

[uk_brewer]

Todays log, no connection to spark. Spark is connected to the rpi by usb, and also displaying a wifi address.

https://termbin.com/rq1ef

[Elco]

Your log file seems incomplete, that’s weird.

It will not use USB, because you have --device-host=192.168.1.30 in your docker file, based on your previous complete log.

It will use that address and will not try anything else. Does that match the address displayed on your Spark?

You could try disabling ipv6 to see if that helps:

Some basic newbie questions

I’m seeing quite a lot of network interrupts, also in other services. Docker seems to have issues with ipv6, so it may be helpful to disable that on your Pi. brewblox-ctl down echo "net.ipv6.conf.all.disable_ipv6 = 1" | sudo tee -a /etc/sysctl.conf echo "net.ipv6.conf.default.disable_ipv6 = 1" | sudo tee -a /etc/sysctl.conf echo "net.ipv6.conf.lo.disable_ipv6 = 1" | sudo tee -a /etc/sysctl.conf sudo sysctl -p # Check results - this command should print "1" cat /proc/sys/net/ipv6/conf/all/dis…

Maybe if the tilt service restarts, it causes network problems for the spark service.

[uk_brewer]

Elco, pretty sure I already disabled ipv6. The ip address matches the spark, I have it as a reserved address on my router.

[Bob_Steers]

Could you please verify? You can do so using

cat /proc/sys/net/ipv6/conf/all/disable_ipv6

If this echoes “1”, it’s disabled.

[uk_brewer]

I get a 1 in the response.

[Elco]

I tried your eeprom on a spark here, and the connection is stable, without a single drop.
I have no clue right now why it would not be stable at your place.

Do you see any difference when the tilt is not online?

@j616s What is the restart interval when the tilt service cannot find a connection?
Is there any mechanism to prevent very frequent restarts of the container?
We see that container restarts shortly take down the host network because it is reconfigured. I think this only happens with ipv6, but perhaps it would be good to introduce a minimal restart interval of 1 minute + some randomness. Or if a container restart is not necessary that would be even better.

[j616s]

I don’t think I’ve added anything for restart intervals. The container is pretty close to the template. Happy to take a look though.

[uk_brewer]

Elco

I don’t have a tilt running right now. Currently running just to main fridge temp constant, don’t feel like I can trust this to run a brew session as is.