Bug 28146

Summary: RETVAL=$? isn't always reliable due to daemon()/fork()
Product: [Retired] Red Hat Linux Reporter: Pekka Savola <pekkas>
Component: opensshAssignee: Tomas Mraz <tmraz>
Status: CLOSED NOTABUG QA Contact:
Severity: medium Docs Contact:
Priority: medium    
Version: 7.0CC: dr
Target Milestone: ---   
Target Release: ---   
Hardware: i386   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2005-02-02 15:49:14 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description Pekka Savola 2001-02-17 16:07:17 UTC
This is probably a more generic problem due to a coding style, but it happened to me with 
OpenSSH, so I'll put it in the record here.

(just trying to start sshd when it's already running)

# ./sshd -d
[snip]
socket: Invalid argument
debug1: Bind to port 22 on 0.0.0.0.
fatal: Cannot bind any address.
# echo $?
255

# ./sshd
# echo $?
0

With './sshd', the same fatal message is printed to syslog.

---

Now, the init script e.g. does:

---
        action $"Starting $prog: " /usr/sbin/sshd
        RETVAL=$?
---
now RETVAL is 0 (as above), even though sshd really fatal()'ed out with 255 exit code.

Kevin Steves <stevesk.com> explained on openssh-univ-dev list:
---
this one has forked and detached from the terminal at the point of that
error.  its parent does exit(0) in daemon() which is what the shell
sees.
---

So, it'd appear RETVAL or exit codes are not always too reliable.  Luckily in this case it doesn't really
matter that RETVAL is mistaken to be 0.

Comment 1 Tomas Mraz 2005-02-02 15:49:14 UTC
This is not a bug but rather a feature of the current SYS V init
scripts style code. The RETVAL can reflect only errors before the
daemonization happens. There is nothing we can do with that.