[svlug] (forw) [DNG] Linux system can be brought down by sending SIGILL to Systemd

Rick Moen rick at linuxmafia.com
Sat May 25 09:58:03 PDT 2019


Quoting Michael Eager (eager at eagerm.com):

> Agreed, except the OP mentioned writing a script to explictly send ILL 
> to proc 1.  Not an accident.

Maybe I wasn't paying enough attention (am still jetlagged), but my
interpretation was that it was a typo in composition _of_ a script, and
thus still an accident.

Typos happen is shell scripts.  I have the figurative scars to prove it.

> It looks like the problem is in kill(2) not in init or systemd.  It is 
> kill(2) man page that says that it will only send "good" signals to proc 
> 1.  It isn't the init process that seems broken.

Is that indeed the problem?  The reporter said he crosschecked the same
thing _with a different init_, and it didn't happen.  I really didn't
look further at the time, and I'm not likely to do do so today, either,
as it's shaping up to be a busy Saturday, e.g., I have cooking to do in  
preparation for today's CABAL meeting (4pm to midnight, Menlo Park, 
http://linuxmafia.com/cabal/ ).

Anyway, I really have no horse in this race, so I really didn't want to
get that deep in the weeds.  I was just really amused.  TBH, I really
don't care enough to go re-read and study the situation more.  It
suffices that I'm glad my system isn't likely to fall over if I leave
the 'k' off 'kill', even if I'm UID0.


> Which tool are you talking about?  

I'm going to have to beg off, Michael, apologies, as I'm actually out of
time.  

> So, someone should file a bug against kill(2).  Right?

Even though it happens only with one init and not others?  Colour me
skeptical.




More information about the svlug mailing list