Too many problems with orange pi 2G-IOT and its OS

Armen

Hi
Finally I receive my board. at first what can I say:
1. WiFi is very unstable: In Debian image WIFI doesn't work at all, Ubuntu image can connect to network but periodically loses connection and reconnect
2. Cannot boot from first power on: For booting need reconnect power until boot usually 5-8 times. I try with other power supply but same result

malvcr

crossmax replied at 2017-5-30 21:28
Hi,
Then it seems that the problems I'm having when testing the device are normal. I have a lot ssh  ...

There are several issues with these machines.

1) On Ubuntu, before updating, they won't keep the MAC address.  On Rasbian, after updating, they won't keep the MAC address.  Although I can't guarantee that the MAC will survive (for that check how to store and retrieve the last used MAC address on Linux).

2) On Ubuntu, they try fieressly to find the WIFI registration domain (the country).  This happens because that it is not defined in the default configuration.

When they are doing that, the WIFI becomes very unstable ... there are some operations that constantly close and reopen the WIFI connection.


So ... after updating and upgrading.

Complete the CRDA parameters

In /etc/default/crda   ....   define a DOMAIN (for example, US).  I think it must be the same defined in WPA_supplicant.conf

REGDOMAIN=US

And also, configure the Domain with "iw".  I am not sure if this is necessary because of the CRDA configuration, but I let the command in case it helps.

sudo iw reg set US


This won't resolve all the WIFI issues (it is still a little unstable on the connection speed), but will make life much more easier and will avoid several disconnections and to full the logs with strante CRDA messages.

crossmax

malvcr replied at 2017-6-17 23:24
There are several issues with these machines.

1) On Ubuntu, before updating, they won't keep the ...

Hi,I have re-tested the new OS versions (Ubuntu, Debian and Raspbian) to know if has considerable improvements. Wifi still hateful despite making the changes you mention. In my case, the /etc/default/crda file did not exist and I created it.
The wifi instability makes it impossible to work with it, impossible even OS setting up with some package without losing a couple of hours. And work throught serial port is tedious too. I can't test the WiringPi library without despair.

For all of that, I'm still waiting for a good linux OS image. I hope find it soon.

cmrahman

I got the board but i can't get any output from TTL .  Any clue?

Reinhard

You already see nothing during booting?
Also no hieroglyphs? Then check the function of your USB to TTL adapter.Note: Some distributions use baud rate 921600.
The DIP switches are used to assign the USB ports and have no influence on the serial data transfer.

olauah

My 2 week experience with the 2G-IOT, using mostly the ubuntu image, but also tested Debian and Armbian, is that:

• It is highly sensitive to the power supply and even if it boots, some things may fail later due to "unsuited" supply. Tested with 2A, 3A chargers (all original from Apple, Huawei and Blitzwolf), as well as with 3A DC-DC step down modules, and a HP lab power supply. A small resistance in the cable (e.g 0.5-1ohm) is enough to cause trouble, even if the supply provides 5.000V stable. It also not just a matter of max Amps available but also the noise produced by the supply. A capacitor in the DC line may help, however I only saw some changes with a 5000uf capacitor, which is not practical.
  

• Using an lab power supply and a oscilloscope I determined that while idle, the power consumption is ~60% of the time around 90mA, then it increases to ~250mA for the remaining 40% of the time, and the cycle restarts. During this time, it will repeatedly demand almost 2A during ~50ns. (do not have the traces here, and the values are from memory, but I can provide them later). This is not a friendly behaviour for any power supply and I assume it could be producing some of interference.
  

• GSM Modem works ok if the power supply is reasonable. The first symptom of a bad power supply seems to be failure to detect the SIM card insertion, resulting in CME ERROR: 10.
  

• Wifi is terribly unstable and almost not usable for a continuous session. It seems to be the first component to fail with a "less good" power supply. There are many packet losses and high jitter (crda installed, power save off).
  

Also, the behavior is different depending on the traffic direction:

Tested in a wifi network in channel 1, with no other network in the same channel and no traffic in the wireless network, using WPA and a Huawei 802.11ac router.
Ping from 2G-IOT (102.168.1.20) to another wifi host (192.168.1.67):
PING 192.168.1.67 (192.168.1.67) 56(84) bytes of data.
64 bytes from 192.168.1.67: icmp_seq=1 ttl=64 time=3.29 ms
64 bytes from 192.168.1.67: icmp_seq=2 ttl=64 time=67.3 ms
64 bytes from 192.168.1.67: icmp_seq=3 ttl=64 time=47.7 ms
64 bytes from 192.168.1.67: icmp_seq=4 ttl=64 time=42.6 ms
64 bytes from 192.168.1.67: icmp_seq=5 ttl=64 time=19.9 ms
64 bytes from 192.168.1.67: icmp_seq=6 ttl=64 time=47.9 ms
64 bytes from 192.168.1.67: icmp_seq=7 ttl=64 time=43.2 ms
64 bytes from 192.168.1.67: icmp_seq=8 ttl=64 time=22.3 ms
64 bytes from 192.168.1.67: icmp_seq=9 ttl=64 time=39.9 ms
64 bytes from 192.168.1.67: icmp_seq=10 ttl=64 time=35.4 ms

--- 192.168.1.67 ping statistics ---
12 packets transmitted, 10 received, 16% packet loss, time 11048ms
rtt min/avg/max/mdev = 3.293/37.005/67.381/16.971 ms



simultaneous ping from the wifi host (192.168.1.67) to the 2G-IOT (192.168.1.20)
PING 192.168.1.20 (192.168.1.20): 56 data bytes
64 bytes from 192.168.1.20: icmp_seq=0 ttl=64 time=552.454 ms
64 bytes from 192.168.1.20: icmp_seq=1 ttl=64 time=82.929 ms
64 bytes from 192.168.1.20: icmp_seq=2 ttl=64 time=204.815 ms
64 bytes from 192.168.1.20: icmp_seq=3 ttl=64 time=559.674 ms
64 bytes from 192.168.1.20: icmp_seq=4 ttl=64 time=80.399 ms
64 bytes from 192.168.1.20: icmp_seq=5 ttl=64 time=307.805 ms
64 bytes from 192.168.1.20: icmp_seq=6 ttl=64 time=564.616 ms
64 bytes from 192.168.1.20: icmp_seq=7 ttl=64 time=91.208 ms
64 bytes from 192.168.1.20: icmp_seq=8 ttl=64 time=367.757 ms
64 bytes from 192.168.1.20: icmp_seq=9 ttl=64 time=571.671 ms
64 bytes from 192.168.1.20: icmp_seq=10 ttl=64 time=100.303 ms
64 bytes from 192.168.1.20: icmp_seq=11 ttl=64 time=372.485 ms

--- 192.168.1.20 ping statistics ---
13 packets transmitted, 12 packets received, 7.7% packet loss
round-trip min/avg/max/stddev = 80.399/321.343/571.671/197.338 ms

crossmax

olauah replied at 2017-8-4 20:44
My 2 week experience with the 2G-IOT, using mostly the ubuntu image, but also tested Debian and Armb ...

Very good analysis that shows what we have all been commenting. The 2G-IOT is terribly unstable making it impossible to use

zoums

The wifi is so bad.

x893

Agree, board very bad. Unstable and no normal OS support.
(last check with OrangePi_2G-IOT_Debian_GSM_Package.rar)
Wants use for small LORA gateways but unpossible.
Go back to NanoPi S2 with external GSM modem.

crossmax

x893 replied at 2017-9-10 02:20
Agree, board very bad. Unstable and no normal OS support.
(last check with OrangePi_2G-IOT_Debian_GS ...

And out of curiosity, why do you use for gateway a NanoPi S2 and not use a Raspiberry Zero W? Maybe for OS boot from emmc?

Thanks