Hi @Martin-Kudláček ,

Sorry to wake this one up again.

But i hope you can reassure me a little bit more.

This high load kind of freaks me out, so i would really like to put a finger on whats happening

I have a lot of Neurons running (at least 15+) with very low load averages, always under 0,10

I also have a lot of (newer) Neurons running (some 20+) with very high load averages, like @cristian mentioned, way above 1,20

I haven't got influxdb running on those, and i'm not doing weird things as far as i know.

The only obvious difference I can find is that in the low-load unipi's ther's still the old drivers called NEURONSPI, and in the high-load unipi's they are called UNIPISPI.

I'm not sure this is the culprit, but it seems that the OS version is not really an issue

I have Debian stretch on the older ones, with low load, but i have also stretch installs that show the high load.

All unipi's with Buster have the high load for sure, but they're all UNIPISPI according to dmesg

Is there anything you are aware of that justifies the load difference between UNIPISPI and NEURONSPI?

Or do i have to look further... Or should i really not worry about this?

Most of these unipi's are in production environments long drives away, so if there's something to be fixed i would try to get that done before they break...

Thanks a lot, Tony, Fireware Netherlands

)

I looked into your package pinning suggestion.

apt-mark hold raspberrypi-bootloader
apt-mark hold python-tornado
apt-mark hold raspberrypi-kernel

before an upgrade at least prevents things getting broken

so at least there's a temporary prevention scheme

Hi Tomas,

Thanks for the suggestions... i'll keep an eye on the git issues ticket, do you want to keep this one alive here too?

Do I understand correctly that it's just a question of pinning the packages you mention too keep my running (remote)neurons safe for now?

i'm so happy I did the upgrade on my desk first, instead of doing it on the remote sites already

i have to say that normally i do apt-get update/upgrade on a regular basis on most of our systems without giving it much thought, so it's good to know that in this case it's a bit more dangerous :/

i'll get back to you after i tried an upgrade with pinned packages...

cheers, tony

here's an example flow that should work:

[{"id":"96ae23da.869f9","type":"websocket in","z":"5bcdfac1.29ad74","name":"","server":"","client":"e7a76108.fb387","x":286.0476608276367,"y":1473.309573173523,"wires":[["8f278ce2.4128c"]]},{"id":"8f278ce2.4128c","type":"function","z":"5bcdfac1.29ad74","name":"IN 1-7 -> RELAY 1-7","func":"var status = JSON.parse(msg.payload);\nvar msg = {};\nif (status.dev == \"input\" && status.circuit<8) {\n\t//leave the relay 8 to the temp control node\n\tif (status.value === 0) {\n\t\tmsg.payload = {\"dev\":\"relay\", \"circuit\":\"\", \"value\":\"0\"};\n\t\tmsg.payload[\"circuit\"] = status.circuit;\n\t}\n\telse {\n\t\tmsg.payload = {\"dev\":\"relay\", \"circuit\":\"\", \"value\":\"1\"};\n\t\tmsg.payload[\"circuit\"] = status.circuit;\n\t}\n\treturn msg;\n}\nelse return;\n","outputs":1,"noerr":0,"x":570.0476608276367,"y":1475.309573173523,"wires":[["f54e4693.87a0f8"]]},{"id":"f54e4693.87a0f8","type":"websocket out","z":"5bcdfac1.29ad74","name":"","server":"","client":"e7a76108.fb387","x":959.0476684570312,"y":1474.3095703125,"wires":[]},{"id":"204f3723.1da8d8","type":"inject","z":"5bcdfac1.29ad74","name":"R1 ON","topic":"","payload":"{\"cmd\":\"set\", \"dev\":\"relay\", \"circuit\":\"1_01\", \"value\":\"1\"}","payloadType":"str","repeat":"","crontab":"","once":false,"x":790,"y":1560,"wires":[["f54e4693.87a0f8"]]},{"id":"cd75e128.e55f8","type":"inject","z":"5bcdfac1.29ad74","name":"R1 OFF","topic":"","payload":"{\"cmd\":\"set\", \"dev\":\"relay\", \"circuit\":\"1_01\", \"value\":\"0\"}","payloadType":"str","repeat":"","crontab":"","once":false,"x":790,"y":1596,"wires":[["f54e4693.87a0f8"]]},{"id":"e7a76108.fb387","type":"websocket-client","z":"","path":"ws://localhost/ws","wholemsg":"false"}]

Hi guys,

I would like to tag along on this one

• little sidenote: this is probably a different problem than the problem mentioned earlier in this thread...
  in this case the board is actually not found on SPI at all, so no browser or socket problem i guess *

if i start out with image UniPian-Neuron-OS-2018-03-09.img all works fine out of the box

Evok fine everything fine

But when i just apt-get update / apt-get upgrade things get broken

Does anybody know where i need to look for this?

My guess is that its a kernel thing... it breaks after i reboot, not before taht

under here i have both "good" and "broken" log blurbs

thanks and greetz from holland

all is fine:

root@S103-sn488:/boot# uname -a
Linux S103-sn488 4.9.41-v7+ #1023 SMP Tue Aug 8 16:00:15 BST 2017 armv7l GNU/Linux

root@S103-sn488:/home/pi# dmesg | grep neuronspi
[ 6.246724] neuronspi: loading out-of-tree module taints kernel.
[ 6.257603] neuronspi: probe of spi0.3 failed with error -5
[ 6.270814] neuronspi: probe of spi0.2 failed with error -5

root@S103-sn488:/boot# python /opt/evok/evok.py
[I 180321 08:09:18 config:63] eprom: UniPi Neuron S103 version: 1.0 serial: 0x1e8
[I 180321 08:09:18 evok:1628] Starting using config file /etc/evok.conf
[I 180321 08:09:18 config:75] YAML Definition loaded: L20x.yaml, type: 3, definition count 0
[I 180321 08:09:18 config:75] YAML Definition loaded: M30x.yaml, type: 3, definition count 1
[I 180321 08:09:18 config:75] YAML Definition loaded: L50x.yaml, type: 3, definition count 2
[I 180321 08:09:18 config:75] YAML Definition loaded: S10x.yaml, type: 3, definition count 3
[I 180321 08:09:18 config:75] YAML Definition loaded: xS40.yaml, type: 3, definition count 4
[I 180321 08:09:19 config:75] YAML Definition loaded: M10x.yaml, type: 3, definition count 5
[I 180321 08:09:19 config:75] YAML Definition loaded: L30x.yaml, type: 3, definition count 6
[I 180321 08:09:19 config:75] YAML Definition loaded: M50x.yaml, type: 3, definition count 7
[I 180321 08:09:19 config:75] YAML Definition loaded: L51x.yaml, type: 3, definition count 8
[I 180321 08:09:20 config:75] YAML Definition loaded: xS30.yaml, type: 3, definition count 9
[I 180321 08:09:20 config:75] YAML Definition loaded: M40x.yaml, type: 3, definition count 10
[I 180321 08:09:20 config:75] YAML Definition loaded: M20x.yaml, type: 3, definition count 11
[I 180321 08:09:20 config:75] YAML Definition loaded: CUSTOM MODBUS DEVICE.yaml, type: 3, definition count 12
[I 180321 08:09:20 config:75] YAML Definition loaded: L40x.yaml, type: 3, definition count 13
[I 180321 08:09:20 config:75] YAML Definition loaded: xS10.yaml, type: 3, definition count 14
[I 180321 08:09:20 config:75] YAML Definition loaded: xS50.yaml, type: 3, definition count 15
[I 180321 08:09:20 config:75] YAML Definition loaded: evok-alias.yaml, type: 2, definition count 0
[I 180321 08:09:20 evok:1692] HTTP server listening on port: 8080
[I 180321 08:09:20 modbusclient_tornado:122] SPI client started
[I 180321 08:09:20 neuron:157] Reading SPI boards
[D 180321 08:09:20 owclient:377] Entering 1wire loop
[I 180321 08:09:20 neuron:172] No board on SPI 2
[I 180321 08:09:20 neuron:172] No board on SPI 3
[I 180321 08:09:21 web:2063] 101 GET /ws (::1) 7.53ms
[D 180321 08:09:21 evok:158] New WebSocket client connected
[I 180321 08:09:21 web:2063] 304 GET /rest/all (::1) 9.60ms

things got broken:

root@S103-sn488:/home/pi# uname -a
Linux S103-sn488 4.9.80-v7+ #1098 SMP Fri Mar 9 19:11:42 GMT 2018 armv7l GNU/Linux

root@S103-sn488:/home/pi# dmesg | grep neuronspi
root@S103-sn488:/home/pi#
nothing

root@S103-sn488:/home/pi# python /opt/evok/evok.py
[I 180321 08:24:14 config:63] eprom: UniPi Neuron S103 version: 1.0 serial: 0x1e8
[I 180321 08:24:14 evok:1628] Starting using config file /etc/evok.conf
[I 180321 08:24:14 config:75] YAML Definition loaded: L20x.yaml, type: 3, definition count 0
[I 180321 08:24:14 config:75] YAML Definition loaded: M30x.yaml, type: 3, definition count 1
[I 180321 08:24:15 config:75] YAML Definition loaded: L50x.yaml, type: 3, definition count 2
[I 180321 08:24:15 config:75] YAML Definition loaded: S10x.yaml, type: 3, definition count 3
[I 180321 08:24:15 config:75] YAML Definition loaded: xS40.yaml, type: 3, definition count 4
[I 180321 08:24:15 config:75] YAML Definition loaded: M10x.yaml, type: 3, definition count 5
[I 180321 08:24:15 config:75] YAML Definition loaded: L30x.yaml, type: 3, definition count 6
[I 180321 08:24:15 config:75] YAML Definition loaded: M50x.yaml, type: 3, definition count 7
[I 180321 08:24:16 config:75] YAML Definition loaded: L51x.yaml, type: 3, definition count 8
[I 180321 08:24:16 config:75] YAML Definition loaded: xS30.yaml, type: 3, definition count 9
[I 180321 08:24:16 config:75] YAML Definition loaded: M40x.yaml, type: 3, definition count 10
[I 180321 08:24:16 config:75] YAML Definition loaded: M20x.yaml, type: 3, definition count 11
[I 180321 08:24:16 config:75] YAML Definition loaded: CUSTOM MODBUS DEVICE.yaml, type: 3, definition count 12
[I 180321 08:24:16 config:75] YAML Definition loaded: L40x.yaml, type: 3, definition count 13
[I 180321 08:24:17 config:75] YAML Definition loaded: xS10.yaml, type: 3, definition count 14
[I 180321 08:24:17 config:75] YAML Definition loaded: xS50.yaml, type: 3, definition count 15
[I 180321 08:24:17 config:75] YAML Definition loaded: evok-alias.yaml, type: 2, definition count 0
[I 180321 08:24:17 evok:1692] HTTP server listening on port: 8080
[I 180321 08:24:17 modbusclient_tornado:122] SPI client started
[I 180321 08:24:17 neuron:157] Reading SPI boards
[I 180321 08:24:17 neuron:172] No board on SPI 1
[I 180321 08:24:17 neuron:172] No board on SPI 2
[D 180321 08:24:17 owclient:377] Entering 1wire loop
[I 180321 08:24:17 neuron:172] No board on SPI 3
[I 180321 08:24:17 web:2063] 101 GET /ws (127.0.0.1) 3.58ms
[D 180321 08:24:17 evok:158] New WebSocket client connected
[I 180321 08:24:17 web:2063] 304 GET /rest/all (127.0.0.1) 4.17ms
[E 180321 08:24:17 neuron:211] 'NoneType' object has no attribute 'do_scan'
Traceback (most recent call last):
File "/opt/evok/neuron.py", line 209, in scan_boards
yield self.modbus_cache_map.do_scan()
AttributeError: 'NoneType' object has no attribute 'do_scan'