namespaces.7: Note treatment of PID namespace "init" process with respect to signals

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
This commit is contained in:
Michael Kerrisk 2013-01-16 01:25:16 +01:00
parent 33a3c1b8ec
commit e17d07c17b
1 changed files with 12 additions and 0 deletions

View File

@ -345,6 +345,18 @@ the kernel terminates all of the processes in the namespace.
This behavior reflects the fact that the "init" process
is essential for the correct operation of a PID namespace.
Only signals for which the "init" process has established a signal handler
can be sent to the "init" process by other members of the PID namespace.
This restriction applies even to privileged processes,
and prevents other members of the PID namespace from
accidentally killing the "init" process.
However, within ancestor namespaces
the "init" process is treated as a normal user process:
any process can\(emsubject to the usual permission checks described in
.BR kill (2)\(emsend
any signal to the "init" process,
including signals that may result in its termination.
PID namespaces can be nested.
When a new PID namespace is created,
the processes in that namespace are visible