Error,

This morning the machine operator said the hadwheel started acting erratic. I told him to close LinucCNC and reopen it. When he did this comes up.

I am assuming the machine was stable, with acceptable latency before.
What may have changed on the machine?
Was new hardware added? (maybe a real-time unfriendly USB device.)
Was the computer shutdown with a dead cmos battery causing it to reset to a default setting with poor latency?

Yes, the machine ran all night and and most of the morning untill it got weird. It has been running the same program for a week.

It may be nothing, but this looks weird to me:

why is it mounting a file system at this point and what is it recovering from?

[ 1254.467145] EXT4-fs (sda2): warning: maximal mount count reached, running e2fsck is recommended
[ 1254.468391] EXT4-fs (sda2): recovery complete
[ 1254.468574] EXT4-fs (sda2): mounted filesystem with ordered data mode

PCW wrote:

It may be nothing, but this looks weird to me:

why is it mounting a file system at this point and what is it recovering from?

[ 1254.467145] EXT4-fs (sda2): warning: maximal mount count reached, running e2fsck is recommended
[ 1254.468391] EXT4-fs (sda2): recovery complete
[ 1254.468574] EXT4-fs (sda2): mounted filesystem with ordered data mode

None of that makes any sense to me.

Good news, for now. It is working. I ran memtestx86 and after a few hours and no errors found I rebooted and it is working fine.

Bad news it's back, not running

Same dmesg error?

sounds a bit like failing hardware

Yes saying error I am running memtest

I would clone the hard disk or replace the hard disk with a backup immediately. Looks like the disk is about to fail, or it too late and has failed already.

EXT4-fs (sda2): warning: maximal mount count reached, running e2fsck

this has been flagged on the Linux file system. Sometimes typical power failures ect.. causes corruption and can be tagged as needing to be checked with e2fsck. But in your case a steady build that is acting "strange" indicates a failing disk. Or corrupt data is being written for other reasons, faulty hardware ; memory ; drive controller.



EDIT: have you ran e2fsck to check the file system?

Link here

This should reset the mount count warning. The issue is perhaps you have corrupted data, either from faulty drive or from multiple powerfailures ect without running e2fsck afterwards. (not shutting down completely prior to power off)

my .02

Realtime system did not load

Have you tried to reload linuxcnc after you get his error? Sometimes when I get this error a reload will bring it back up.