Information - LoRa server load test

[sebastien...@gmail.com]

Hello, 

I wanted to know if load tests were done on this LoRa server.

Here is the configuration in which I use the LoRa server: 

- RPi 3B + 

- IC808A

- Ubuntu 18.04 arm 64bit

- lorawan-server 0.6.7

- Erlang / OTP 20 [erts-9.2] [source] [64-bit] [smp: 4: 4] [ds: 4: 4: 10] [async-threads: 10] [kernel-poll: false] 

If yes, what the hardware configuration, number of devices, payload length, ...?

I regularly have LoRa server shutdowns and my questions are as follows: 

- Can an Erlang update fix the stability of the LoRa server?

- What is the maximum number of devices per GW on this type of configuration?

- I regularly have a Dwell greater than 20%, how can I correct this problem? By reducing the number of devices? by reducing the amount of message? By reducing the size of the payload?

Thank you so much for your help.

Have a good day!

[Александр Тищенко]

Hello Sebastian,

As for myself, I didn't do any serious load tests on a raspberry Pi HW, but I guess 20-30 "standard" devices or even more should be ok.

To the serious stuff:

- What do you mean by having a server shutdown? How does this happen? What is in the logs? It must not shutdown out of its own accord, not at all.

- Which dwell time you are talking about? How does it manifest itself?

Cheers,

Alex

вт, 27 окт. 2020 г. в 14:08, Sebastien Carobbio <sebastien...@gmail.com>:

Hello, 

I wanted to know if load tests were done on this LoRa server.

Here is the configuration in which I use the LoRa server: 

- RPi 3B + 

- IC808A

- Ubuntu 18.04 arm 64bit

- lorawan-server 0.6.7

- Erlang / OTP 20 [erts-9.2] [source] [64-bit] [smp: 4: 4] [ds: 4: 4: 10] [async-threads: 10] [kernel-poll: false] 

If yes, what is the hardware configuration, number of devices, payload length, ...?

Another point:

I regularly have LoRa server shutdowns and my questions are as follows: 

- Can an Erlang update fix the stability of the LoRa server?

- What is the maximum number of devices per GW on this type of configuration?

- I regularly have a Dwell greater than 20%, how can I correct this problem? By reducing the number of devices? By reducing the amount of messages? By reducing the size of the payload?

[Sebastien Carobbio]

Hello Alex,

I think that I have a problem with Mongo Connector on my LoRawan server.

In attachment, we can find crash.log file.

I see all messages in Dashboard view and rxFrame view but I haven't saved messages in my mongo BDD (DB name: Database, collection name: Collection).

For information, I have around 700 devices connected on this GW. Below, a picture of my GW dashboard.

Can you help me to understand why datas aren't saved??

Thanks a lot for your help!

Le lun. 2 nov. 2020 à 11:03, Sebastien Carobbio <sebastien...@gmail.com> a écrit :

Hello Alex,

I think that I have a problem with Mongo Connector on my LoRawan server.

In attachment, we can find crash.log file.

I see all messages in Dashboard view and rxFrame view but I haven't saved messages in my mongo BDD (DB name: Database, collection name: Collection).

For information, I have around 700 devices connected on this GW. Below, a picture of my GW dashboard.

Can you help me to understand why datas aren't saved??

Thanks a lot for your help!