Ip dropout

[Tokatubs]

Having a bit of an issue with a brand new Logic Machine 5.
Have set it up and knx tp is up and running. 

The latest days it loses its ip. So i cant connect to it. 
Taking out 24v and data cabel. And reconnecting it that way it some times comes back. And stays normal for some hours. 
Then again it loses its connection. And i cant connect to it. 
Leds are blinking and everything seems normal. But cant connect on ip. 

Advanced ip scanner shows that LM is offline. Restarting described above brings it back up. Tested ip and thats not the problem. 

Running latest firmware.

Any tips on what to try, have many other LM5 and have never seen this.

[Daniel]

I would bet that device has full memory and it keeps rebooting. Factory reset would help if you have backup.

[Tokatubs]

I would bet that device has full memory and it keeps rebooting. Factory reset would help if you have backup.

Device is brand new. Up and running for about a week. Memory and Cpu seems fine. Nothing in error log. 
Today again it was offline. Got the client to restart it. Now up again until next time. 


   

[Daniel]

I would monitor it over time as you see it only after reboot. Try this script

Code:

data = io.readfile('/proc/meminfo')
load = io.readfile('/proc/loadavg'):split(' ')[ 1 ]

memtotal = data:match('MemTotal:%s+(%d+)')
memavail = data:match('MemAvailable:%s+(%d+)')
memusage = math.floor((memtotal - memavail) / memtotal * 1000) / 10

log(load, memusage)

[Tokatubs]

Just run it once, and doesnt matter what type i run it as?

[Daniel]

You should run it periodically over time.

[Tokatubs]

I put it as resident script then. Seems like its giving values.

[Daniel]

Keep checking it periodically or create a trend. We want to see if it is not growing over time.

[Tokatubs]

How would i be able to create a trend from this script. I see the "number" value increasing. The rest is steady.

[Daniel]

Just write the values in log to two objects.

[Tokatubs]

I guess for you guys this is kindergarden stuff

But was trying to do this, but see that i need some help.
What is the simplest way to write make this script lines.

I created 2 virtual adresses, and then tried using grp.update with the Memtotal name inserted. But then i dont see how i can send it to correct virtual adress.
And i guess the grp.update is incorrect aswell. But learning by doing i guess.
A final push in the right direction ?

[Daniel]

Create two objects with 2byte floating dpt, In my example they have names CPU and Memory

Code:

data = io.readfile('/proc/meminfo')
load = io.readfile('/proc/loadavg'):split(' ')[ 1 ]

memtotal = data:match('MemTotal:%s+(%d+)')
memavail = data:match('MemAvailable:%s+(%d+)')
memusage = math.floor((memtotal - memavail) / memtotal * 1000) / 10

log(load, memusage)

grp.update('CPU', load)

grp.update('Memory', memusage)

[Tokatubs]

Today it happened again. 2 strange times marked in picture. During this time i cant log in via IP.

   

[Daniel]

The numbers are good, Maybe you have IP conflict? Try changing the IP.

[Tokatubs]

The numbers are good, Maybe you have IP conflict? Try changing the IP.

In the beginning it was set to dhcp. After issue started i put it to static. Its a technical net, so we are sure that no other has this address.

[Daniel]

I have no other ideas. It looks like hardware issue. Ask for replacement.