[prev in list] [next in list] [prev in thread] [next in thread] 

List:       hurd-bug
Subject:    Re: [PATCH 3/8] initscripts: use pidof -s /sbin/init for robustness
From:       Justus Winter <4winter () informatik ! uni-hamburg ! de>
Date:       2013-07-22 15:14:28
Message-ID: 20130722151428.13666.13032 () thinkbox ! jade-hamburg ! de
[Download RAW message or body]

Quoting Pino Toscano (2013-07-22 15:57:45)
> Alle martedì 9 luglio 2013, Justus Winter ha scritto:
> > pidof -s returns the first matching process. Since readdir(2) on
> > procfs returns the processes in ascending order, this returns the
> > first process. This is more robust if there is more than one
> > /sbin/init process.
> > 
> > Currently on Hurd there are two /sbin/init processes.
> 
> Are you sure there are two /sbin/init process around, when using 
> sysvinit? Back then in my experiments, I remember one /hurd/init (as of 
> now) and /sbin/init, and I had to fix procfs so `pidof /sbin/init` would 
> actually match exactly and not go into fuzzy search (which would find 
> /hurd/init and /sbin/init).

Sounds crazy, but yes, I'm somewhat sure:

Mounting local filesystems...done.
+ grep init
root         1  - Sslo  0:00.02 /hurd/init root=device:hd0s1 console=com0
root         6  - Sslo  0:00.01 init boot
root        12 co Ssfow 0:00.00 init boot
root        13 co So    0:00.01 /bin/sh /etc/init.d/rc S
root       416 co Sow   0:00.03 /bin/sh /etc/init.d/mountall.sh start
root       493 co So    0:00.00 grep init
+ pidof /sbin/init
12 6
+ set +x

Justus

[prev in list] [next in list] [prev in thread] [next in thread]