System crahed often

[PoldiMUC]

Hello, everyone,
I installed a new DietPi system on an RPI3 in February. Unfortunately the system crashes regularly. After a reboot it runs again for a few days until another crash occurs. When booting shows me the motto that there are problems and I should execute the command: "dmesg -l err,crit,alert,emerg".
The following messages look like this:

Code: Select all

root@Homeserver:~# dmesg -l err,crit,alert,emerg
[    0.808381] vc_vchi_sm_init: failed to open VCHI service (-1)
[    0.808387] [vc_sm_connected_init]: failed to initialize shared memory service
[    6.870468] vc_sm_cma_vchi_init: failed to open VCHI service (-1)
[    6.870481] [vc_sm_connected_init]: failed to initialize shared memory service
[203024.080284] BUG: Bad page state in process iobroker.js-con  pfn:11dc6
[203024.089059] page:bd5b8fd8 count:0 mapcount:0 mapping:bd5b0000 index:0x1
[203024.096880] flags: 0x0()
[203024.100574] raw: 00000000 00000100 00000200 bd5b0000 00000001 00000000 ffffffff 00000000
[203024.110899] raw: 00000000
[203024.114648] page dumped because: non-NULL mapping

Unfortunately, I don't understand this at all. Can someone give me a hint where I can start looking for the problem?

While I was writing this here, another system (RPI4) has errors after updating to the current version (DietPi v6.29.2). Before the update, the motto has reported no errors to me. But I don't know if there were any because I didn't execute the command: "dmesg -l err,crit,alert,emerg". Here the error messages of the second system:

Code: Select all

root@HomeserverV4:~# dmesg -l err,crit,alert,emerg
[    0.261382] bcm2708_fb soc:fb: Unable to determine number of FB's. Assuming 1
[    0.261682] raspberrypi-firmware soc:firmware: Request 0x00048003 returned status 0x80000001
[    0.261706] bcm2708_fb soc:fb: Failed to allocate GPU framebuffer (-22)
[    0.261741] bcm2708_fb soc:fb: probe failed, err -22
[    0.354959] vc_vchi_sm_init: failed to open VCHI service (-1)
[    0.354964] [vc_sm_connected_init]: failed to initialize shared memory service
[    5.414642] vc_sm_cma_vchi_init: failed to open VCHI service (-1)
[    5.414648] [vc_sm_connected_init]: failed to initialize shared memory service

I notice that the error message:
vc_vchi_sm_init: failed to open VCHI service (-1)
and
[vc_sm_connected_init]: failed to initialize shared memory service
now occurs on both systems. Is this possibly a bug in DietPi?

Many thanks Jochen


Translated with www.DeepL.com/Translator (free version)

[Joulinar]

Hi,

many thanks for your report. Basically MOTD is not a message of your system that something is not working correctly. It's more an information about useful thinks that can be done on your system. So don't worry on the current MOTD requesting to check error messages

Regarding the messages: vc_vchi_sm - VC shared memory errors you see are expected. Kinda unnecessary overhead and confusing. So not a bug but a feature As well the number in front of the message is the up-time of the system in seconds. So these messages are produced within the first 5-6 seconds after you put power on and they are not related to any crash happen hours or days later.

the more important part is this one

Code: Select all

[203024.080284] BUG: Bad page state in process iobroker.js-con  pfn:11dc6
[203024.089059] page:bd5b8fd8 count:0 mapcount:0 mapping:bd5b0000 index:0x1
[203024.096880] flags: 0x0()
[203024.100574] raw: 00000000 00000100 00000200 bd5b0000 00000001 00000000 ffffffff 00000000
[203024.110899] raw: 00000000
[203024.114648] page dumped because: non-NULL mapping

could be some kernel issues. hard to say what it caused. Can you maybe change the power supply? Or do you use some overclocking?