Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

For bugs related to Red Hat Enterprise Linux 5 product line. The current stable release is 5.10. For Red Hat Enterprise Linux 6 and above, please visit Red Hat JIRA https://issues.redhat.com/secure/CreateIssue!default.jspa?pid=12332745 to report new issues.

Bug 463205

Summary:	mistake in /etc/init.d/functions file's status function
Product:	Red Hat Enterprise Linux 5	Reporter:	Balazs Zachar <balazs.zachar>
Component:	initscripts	Assignee:	initscripts Maintenance Team <initscripts-maint-list>
Status:	CLOSED DUPLICATE	QA Contact:	BaseOS QE <qe-baseos-auto>
Severity:	low	Docs Contact:
Priority:	medium
Version:	5.2	CC:	notting
Target Milestone:	rc
Target Release:	---
Hardware:	All
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2008-09-24 14:15:35 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 Balazs Zachar 2008-09-22 15:36:50 UTC

Description of problem:
I have to run two service with the same name (and same binary but different config) but different pidfile and I have to monitor these processes (with the status argument with its init scripts).
Situation: One service from the two is running (A) and the second is not running (B).
I would like to "monitor" the B service with its init script using the status argument and I get that: this service is running...(BUT NOT)
The init scripts's status argument call the functions file's status function with the -p $PIDFILE argument to determine the process's status. But because the status function in the functions file firstly do pid="$(__pids_pidof "$1")" where $1 is the program's name it didn't use my PIDFILE argument and here it gets the A process's pid and returns with the A process status and not with the B's status. I think it should use the pidfile first to determine the process's pid if somebody give that argument...

Version-Release number of selected component (if applicable):
# rpm -qf /etc/init.d/functions 
initscripts-8.45.19.EL-1

How reproducible:
see Description

Comment 1 Bill Nottingham 2008-09-23 19:17:17 UTC

        # Find pid.
        __pids_var_run "$1" "$pid_file"
        if [ -z "$pid_file" -a -z "$pid" ]; then
                pid="$(__pids_pidof "$1")"
        fi

I'm confused - how is this not using the pidfile?

Comment 2 Balazs Zachar 2008-09-24 07:05:40 UTC

(In reply to comment #1)
>         # Find pid.
>         __pids_var_run "$1" "$pid_file"
>         if [ -z "$pid_file" -a -z "$pid" ]; then
>                 pid="$(__pids_pidof "$1")"
>         fi
> 
> I'm confused - how is this not using the pidfile?

I had problem with the status function not with the killproc... (killproc works fine)

status function's begin:
    status() {
        local base pid pid_file=

        # Test syntax.
        if [ "$#" = 0 ] ; then
                echo $"Usage: status [-p pidfile] {program}"
                return 1
        fi
        if [ "$1" = "-p" ]; then
                pid_file=$2
                shift 2
        fi
        base=${1##*/}

        # First try "pidof"
        pid="$(__pids_pidof "$1")"
        if [ -n "$pid" ]; then
                echo $"${base} (pid $pid) is running..."
                return 0
        fi

Here, you first try the pidof before you use the pidfile to determine the process's pid. So, pidof find the A process's pid and give me the A process's status instead of B.

Comment 3 Bill Nottingham 2008-09-24 14:15:35 UTC


*** This bug has been marked as a duplicate of bug 440658 ***