0
n0n1ck

Orange Pi zero - freezing randomly

Recommended Posts

Armbianmonitor:

Hello,

 

I have been struggling for some time with random freezes. I am running application with i2c0 Fram and UART1 for transferring/receiving data based on MAX3232. The freezes cause no response from Orange PI (debug UART too), only unplug power helps.
I read that it may be a problem with powering but i tried many USB power adapters 2A, 3A and ATX power (28A) connected to GND and 5v, nothing helps. I have lowered cpu max frequency to 816Mhz  and still the same.  All necessary data uploaded via armbianmonitor.

Share this post


Link to post
Share on other sites
8 minutes ago, n0n1ck said:

I have been struggling for some time with random freezes.


Have you tried latest stable kernel or legacy? Also there?

Share this post


Link to post
Share on other sites

Now i updated to Linux orangepizero 5.4.8-sunxi and removed xradio_wlan  cfg80211 mac80211 modules

Share this post


Link to post
Share on other sites

 

I am suffering from the same problem, but my orange freezes only the script that is running in the background, the operating system remains running without a problem. already managed to solve your problem?

Share this post


Link to post
Share on other sites

Probably yes, i used performance on Orange Pi and Nano Pi Neo 1 and no hangs from March

Share this post


Link to post
Share on other sites

I have some 20 Orange Pi Zero's around and I tried quite a few images, the last one was Devuan. I installed it two days ago and this morning I found it crashed. Anyway, I fear the the hardware is definitly unstable. I tried to limit the CPU freq, but no way. A reboot does not solve the issue, you really have to give it a power cycle (!) Some boards can run for months, other will crash after a few hours. I ordered some Raspberry Pi Zero-W now and I will replace these boards. And I have two Orange Pi PC boards (H3): same issue. Switching on the WiFi reduces the uptime a lot so I blacklisted the xr819 module on all boards. I just installed Armbian buster, just to see what happens before they end up in the trash bin. I definitively lost confiance in these boards. I have lost too much time with this crap. Sorry boys.

Share this post


Link to post
Share on other sites

Oh, and freezing also means that some services still work. E.g. ssh still listens on tcp port 22 but no ssh connection possible. Time is sometimes weird, like 1937 or 2178. Sometimes nullmailer works as I receive messages that a certificate is out of date, due to a local time of 21 dec 1937 when the cert was not valid yet.

Share this post


Link to post
Share on other sites

 

2 hours ago, Richard Lucassen said:

I tried quite a few images, the last one was Devuan


If Armbian doesn't work, then you can suspect this:
 

2 hours ago, Richard Lucassen said:

I fear the the hardware is definitly unstable.

Share this post


Link to post
Share on other sites

Here, I have two GPIO OPiZero switches running Armbian. Same Armbian, same software, same hardware, wifi module blacklisted. One is rarely used and works fine. The other one is behaving weirdly. Fortunately this time I'm able to access the board by ssh (which is not always the case):

 

$ timeout 1s sleep 5
timeout: error while loading shared libraries: ld-lijux-armhb.so.3: cannot open shared object file: No such file or directory

 

The one that works:

 

$ timeout 1s sleep 5

(sleep just killed after 1 second)

 

Ok, to make it weirder:

The crashed one:

find /usr/ -name '*ld-lijux*'

find /lib/ -name '*ld-lijux*'
[nothing]

 

The working device:

find /usr/ -name '*ld-lijux*'

find /lib/ -name '*ld-lijux*'
[nothing]

 

Searching for the error message I stumble upon cross compiler issues. Anyway, another issue:

 

# apt update
Segmentation fault

 

Reboot: nothing changes, same issues (!!!) The uSD is ok. Already replaced the uSD twice. Power supply is ok (5.10 V) Only a power cycle resolves these issues (temporarely). Ok, it might be a production batch issue. But: I have two H3 Orange Pi PC's having the same weird behaviour. I also run some Raspberry Pi's: never a problem. And btw: switch on the WiFi chip and you can wait for it.

 

I just installed Armbian Buster and set the governor to "performance" as mentioned above, just to see what happens. And I suspect the hardware because a power cycle is needed. Bit rot. And I did not complain about Armbian! :-)

Share this post


Link to post
Share on other sites
38 minutes ago, Richard Lucassen said:

Power supply is ok


Most of the troubles are caused by a bad PSU/cable combo.

 

How can you prove that its OK? I have around 30 mUSB cables at home and around 70% of them are broken - can only slow charge (low current) mobile phones / transfer data, but my PSU is O.K.

Our test rig reports https://dl.armbian.com/_test-reports/2020-05-31_17.23.38.html that most of the boards (1-3 are sometimes having issues) are surviving extreme stress testings. With wifi enabled and tested. 

 

I can't say its not a hardware problem. Possible. It can be a board quality issue, but before claiming that ... triple check PSU and cabling.

Share this post


Link to post
Share on other sites
39 minutes ago, Richard Lucassen said:

Power supply is ok (5.10 V)

Do not measure the output voltage of your supply. That value is worth nothing. It does not include the resistance of the wire and the connectors and these values can add up badly.

 

Measure for example at the GPIO 5V and GND pins. You will notice quite some fluctuations when putting a load to the board.

Share this post


Link to post
Share on other sites

I got a old but often updated installation on my OPi Zero which does run stable (not much used performance).

 

Do you got the same hardware-limits like

hardware limits: 480 MHz - 1.01 GHz
available frequency steps: 480 MHz, 648 MHz, 816 MHz, 960 MHz, 1.01 GHz

when doing a query with

cpufreq-info

?

Because most H2/H3 CPUs are unstable over 1008Mhz/1.01GHz

 

My stable OPi Zero has the following config:
System diagnosis information has been uploaded to http://ix.io/2obt

Share this post


Link to post
Share on other sites

The 5.10V is applied directly to the board, not via USB. There is a PCB connected to the OPi Zero with a 5V/4A PSU on it. I know USB power supplies are crap. I got some usb chargers here that have as low as 4.65V as output voltage. I don't use these ones. But most important phenomenon: after a reboot the problems remain!

And that's quite weird, must be some bitrot somewhere. And almost all of these boards suffer from these problems. As I bought most of these boards all in one order (also the PC H3 version) must be some hardware glitch somewhere.

 

@ Guidol: just the default cpu frequencies. I never overclock a cpu. I wanted to insert this image with the "insert image from URL" button, but that does not work.

Share this post


Link to post
Share on other sites
24 minutes ago, Richard Lucassen said:

@ Guidol: just the default cpu frequencies. I never overclock a cpu. I wanted to insert this image with the "insert image from URL" button, but that does not work.

should work ;) 

OPi-Zero.jpg

Share this post


Link to post
Share on other sites

No, the URL field got red :-( And I don't see the image here. I just tried FireFox and there the image shows up. Brave browser apparently has problems with it.

Share this post


Link to post
Share on other sites
On 6/3/2020 at 3:09 PM, Richard Lucassen said:

Here, I have two GPIO OPiZero switches running Armbian. Same Armbian, same software, same hardware, wifi module blacklisted. One is rarely used and works fine. The other one is behaving weirdly. Fortunately this time I'm able to access the board by ssh (which is not always the case):

 

$ timeout 1s sleep 5
timeout: error while loading shared libraries: ld-lijux-armhb.so.3: cannot open shared object file: No such file or directory

 

The one that works:

 

$ timeout 1s sleep 5

(sleep just killed after 1 second)

 

Ok, to make it weirder:

The crashed one:

find /usr/ -name '*ld-lijux*'

find /lib/ -name '*ld-lijux*'
[nothing]

 

The working device:

find /usr/ -name '*ld-lijux*'

find /lib/ -name '*ld-lijux*'
[nothing]

 

Searching for the error message I stumble upon cross compiler issues. Anyway, another issue:

 

# apt update
Segmentation fault

 

Reboot: nothing changes, same issues (!!!) The uSD is ok. Already replaced the uSD twice. Power supply is ok (5.10 V) Only a power cycle resolves these issues (temporarely). Ok, it might be a production batch issue. But: I have two H3 Orange Pi PC's having the same weird behaviour. I also run some Raspberry Pi's: never a problem. And btw: switch on the WiFi chip and you can wait for it.

 

I just installed Armbian Buster and set the governor to "performance" as mentioned above, just to see what happens. And I suspect the hardware because a power cycle is needed. Bit rot. And I did not complain about Armbian! :-)

 

Try adding mem=512M to kernel parameters in boot.cmd/boot.scr, smells like defective memory.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
0