Bug 53979

Summary: Apache 1.3.20-15 terminates after 7 killall -USR1 httpd calls
Product: [Retired] Red Hat Raw Hide Reporter: nobody+ctolley
Component: apacheAssignee: Nalin Dahyabhai <nalin>
Status: CLOSED RAWHIDE QA Contact: David Lawrence <dkl>
Severity: low Docs Contact:
Priority: low    
Version: 1.0   
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: 2001-09-24 21:35:14 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 nobody+ctolley 2001-09-24 21:35:10 UTC
From Bugzilla Helper:
User-Agent: Mozilla/4.0 (compatible; MSIE 5.5; Windows NT 5.0)

Description of problem:
changed logrotation from:

/var/log/httpd/access_log {
    missingok
    postrotate
        /usr/bin/killall -HUP httpd 2> /dev/null || true
    endscript
}

/var/log/httpd/agent_log {
    missingok
    postrotate
        /usr/bin/killall -HUP httpd 2> /dev/null || true
    endscript
}

/var/log/httpd/error_log {
    missingok
    postrotate
        /usr/bin/killall -HUP httpd 2> /dev/null || true
    endscript
}

/var/log/httpd/referer_log {
    missingok
    postrotate
        /usr/bin/killall -HUP httpd 2> /dev/null || true
    endscript
}

to:

/var/log/httpd/*log {
    daily
    rotate 2
    missingok
    postrotate
        /usr/bin/killall -USR1 httpd
    endscript
    compress
}

Version-Release number of selected component (if applicable):


How reproducible:
Always

Steps to Reproduce:
1. /etc/init.d/httpd restart
2. for i in 1 2 3 4 5 6 7 8 9 10;do echo -n $i":";killall -USR1 
httpd;echo;sleep 5;done
3. After the 7th killall, "8:httpd: no process killed", and a ps auxf 
shows no httpd processes running.	

Actual Results:  I've vaired the sleep value from 5 to 30 with the same 
results.  I even wondered if it was that the server was taking no traffic, 
so I simulated light load on the server...same results.

Expected Results:  Graceful restart shouldn't terminate the daemon 
completely.  Version 1.3.14 is not succeptable to this, no matter how many 
killall's are issued, nor how fast they are issued.

Additional info:

Error logs only show the last graceful restart being issued:

tail /var/log/error_log

[Mon Sep 24 16:06:47 2001] [notice] SIGUSR1 received.  Doing graceful 
restart

No other error messages.  No core files found on system.

Comment 1 nobody+ctolley 2001-10-02 17:45:30 UTC
Closing this.  Downloaded most recent sources of apache SRPMS and associated 
packages, recompiled, and now there are no problems.