Page 1 of 1

Apocalyptic Error Message..

Posted: 2018/10/03 05:02:50
by Gemesys
The Error Message of the Apocalypse! :o
===============================

Looking at a process-list on my new bleeding-edge Linux Kernel (4.18.11 - as of last week), running in CentOS-7.4,
I noticed what looked like the "Error Message of the Apocalypse" - pretty much every horrible/awful thing that could
happen to your little computing machine appears to be being flagged here.

See, I did a "ps aux" from a CLI root account. I log into to root and check stuff, *before* I
login to a user-account, and run "startx" to fire-up the GUI desktop Xwindows/Gnome thing.

(If you want to disable the GUI at CentOS boot, and thus know what is going on, you can disable the GUI
obfuscator on CentOS-7 with "plymouth" and "dracut" commands, which are named for towns in Mass., USA)

(as root...)
plymouth-set-default-theme details
dracut -f
reboot

But, if you do a "ps aux | more" to see what is running on the latest kernel, in CentOS 7.4, this FULL line below
appears, since the BASH command shell line wraps. If you just start a terminal window in Gnome/Xwindows, the
command-line parms passed to the processes are truncated (That means you don't see about 90% of this line...)

------------------------------------------------------------------------------------------------------------
root 894 0.0 0.2 217012 8312 ? Ss 22:16 0:00 /usr/bin/abrt-watch-log -F BUG: WARNING: at WARNING: CPU: INFO: possible recursive locking detected ernel BUG at list_del corruption list_add corruption do_IRQ: stack overflow: ear stack overflow (cur: eneral protection fault nable to handle kernel ouble fault: RTNL: assertion failed eek! page_mapcount(page) went negative! adness at NETDEV WATCHDOG ysctl table check failed : nobody cared IRQ handler type mismatch Machine Check Exception: Machine check events logged divide error: bounds: coprocessor segment overrun: invalid TSS: segment not present: invalid opcode: alignment check: stack segment: fpu exception: simd exception iret exception: /var/log/messages -- /usr/bin/abrt-dump-oops -xtD
------------------------------------------------------------------------------------------------------------

But in BASH command-line mode, outside of a GUI, you see it all. And it scared the heck out of me. Read it out loud.
Use it to scare the little children who come to your door at Halloween.

But turns out, it is just a bunch of search parameters being passed to an ABRT process,
that then takes these search strings, and looks in /var/log/messages to see if anything
awful has happened that matches any of these awful things in this list, and thus needs to be flashed
as a notification to Mr. or Ms. Computer User. You know, like those ABRT warnings we get all the
time about DAC-READ-ERRORS? Those ones we can safely ignore? This is similar.

Turns out this goofy construct is being replaced by a lovely "systemd-journal" watcher that will
be watching for awfulness, and will not need to pass the awfulness-strings thru the process command
line, and get shown on the system process list. (And give minor heart-attacks to old boys like me...)

This is explained in an email message (where all the good Linux documentation lives, eh?)

https://lists.centos.org/pipermail/cent ... 57821.html

What is basically happening is that process "abrt-watch-log" is running, and watching /var/log/messages,
and that it is using the strings supplied by "abrt-dump-oops".

You can list all the scary "-oops" strings, with the command:

abrt-dump-oops -m

which will just print a table of these strings that "abrt-watch-log" is watching for.

Simple, and nothing to worry about. Whew.
:)