[PATCH 3/5] init: fix race with utmp and children

Petter Reinholdtsen pere at hungry.com
Sun Nov 22 21:05:58 UTC 2009

[Petter Reinholdtsen]
> What happen when getty fail to start?  If I read the code correctly,
> a bogus utmp entry will be written then.  Where did I misunderstand?

I had a closer look, and concluded that the normal child handling in
init would take care of the utmp entry and change the INIT_PROCESS
entry to a DEAD_PROCESS entry if the exec failed.

The patch is now committed to upstream svn.

Happy hacking,
Petter Reinholdtsen

More information about the initscripts-ng-devel mailing list