RE: {bb} Kernel error from bbd on RH linux 9

On Thu, 2003-05-15 at 10:28, Wilson, Eric wrote:
> I believe a triggered alert initiates the condition, however, I don't believe
> the condition is corrected when the status goes back to green. 

One of the issues here revolves around your OS's handling of the SIGCHLD
signal (and handler) and whether signal handlers are inherited by child
processes from the parent.

I suspect Red Hat is about to do something funky here, which wouldn't
surprise me one bit. I get this feeling because version 9 is issuing
warnings about SIGCHLD being set to SIG_IGN when it previously did
not...as in "we're telling you about this now because in the future we
may just kill -9 your program."

>From Red Hat's Seawolf (7.1) man page on signal(2):


According  to  POSIX  (B.3.3.1.3) you must not set the action for
SIGCHLD to SIG_IGN.  Here the BSD and SYSV behaviours differ, causing
BSD software that sets the action for SIGCHLD to SIG_IGN to fail on
Linux.

CONFORMING TO
       ANSI C


Big Brother forks....a lot. A child of the main process will spawn child
processes of its own. In many cases, the parent and child processes need
to be ignorant of each other. In some cases, the parent needs to wait
for the child to finish. Some of the "wait" calls have preprocessor
defines around them (ZOMBIE) as does the initial signal call to set
SIGCHLD to SIG_IGN. Since one of the goals of Big Brother is the ability
to compile and work on as wide a variety of *nix platforms as possible,
this will be a "sticky" issue, especially on the Linux platform.

Linux has already fractioned into many clones. There were 2 big efforts
to bring these clones back under some "commonality," most notably the
Linux Standard Base and United Linux. UL is more of a commercial venture
and is somewhat shaky IMHO...the Linux Standard Base is really
comprehensive and has no commercial interests what so ever. The LSB says
signal should behave according to CAE Specification, January 1997,
System Interfaces and Headers (XSH),Issue 5 (ISBN: 1-85912-181-0, C606).
I have no idea what this document says.

I know I'm not providing any kind of solution or suggestion...just
offering up a little research.

-- 
--Jeff
Sr. Unix Systems Administrator & Engineer
SAIR Linux and GNU Certified Professional

This email may contain privileged and confidential information. It is
intended only for the use of the adressee(s). I'll further add that if
this message is not digitally signed or encrypted using one of my keys,
than I can deny composing it in the first place.

--
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-==-=-=-=-=-=-=-=-=-=-=-=
To unsubscribe from this list send e-mail to mailto:[email protected]
with unsubscribe bbd in the BODY of the message.