Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 14 Mar 2006 13:30:07 GMT
From:      Bernd Ueberbacher <bernd@uebi.net>
To:        freebsd-gnats-submit@FreeBSD.org
Subject:   i386/94446: FreeBSD 6.0 crash with forkbomb
Message-ID:  <200603141330.k2EDU72B001710@www.freebsd.org>
Resent-Message-ID: <200603141340.k2EDeIrb075845@freefall.freebsd.org>

next in thread | raw e-mail | index | archive | help

>Number:         94446
>Category:       i386
>Synopsis:       FreeBSD 6.0 crash with forkbomb
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    freebsd-i386
>State:          open
>Quarter:        
>Keywords:       
>Date-Required:
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Tue Mar 14 13:40:18 GMT 2006
>Closed-Date:
>Last-Modified:
>Originator:     Bernd Ueberbacher
>Release:        6.0-RELEASE
>Organization:
Private
>Environment:
FreeBSD uebi.mynet.at 6.0-RELEASE-p4 FreeBSD 6.0-RELEASE-p4 #0: Fri Feb 17 14:33:41 CET 2006     root@uebi.mynet.at:/usr/obj/usr/src/sys/UEBI  i386
>Description:
Hi there!

I am using 2 Tyan Transport GS-10 as router with FreeBSD 5.4 and quagga-0.98.5. Everything worked fine, but when I changed the Hardware (now Dell PowerEdge SC1425, XEON 3.0 GHz) I installed FreeBSD 6.0 and quagga-0.99. It seemed to work fine, too, but on the FIRST day i had 2 crashes! The box has completelly forzen both times. I checked the logfiles but there where only a few  "em0: RX overrun" messages, nothing else. I changed back to the Tyan router and investigated a bit more. I read that the em0 errors shouldn't be responsible for the crash and thought I could make some stress tests. I installed the forkbomb tool and when I start it with "forkbomb -f --runasroot" I crash the machine within a couple of seconds. The warning "maxproc limit exceeded by uid 0" shows up and the box freezes.

I tried it with 6.0, 6.1-BETA3 and 5.4. It seems that it only affects RELENG_6. When i try the same on 5.4 (same hardware) the machine gets unusable (also reports "maxproc limit exceeded by uid 0"), but as soon as I hit ctrl+c everything goes back to normal.

I was wondering if I made a mistake, but no matter what hardware, what kernel (GENERIC or custom), ... it happens all the time on RELENG_6. Today I tried it on a Sun Enterprise 220 (sparc64) and the same thing happened.

As Goldmember would say: Isn't this weird? ;-)

Please try it on one of your 6.0 boxes and see if the same problem occurs. Excuse my poor english and please don't flame me if I'm not right with what I write, but it just seems odd to me that this happens with a fresh install of 6.0 but not with 5.4


Have a nice day,
greets from Austria

Bernd





>How-To-Repeat:
pkg_add -r forkbomb; rehash; forkbomb -f --runasroot *GG*
>Fix:

>Release-Note:
>Audit-Trail:
>Unformatted:



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200603141330.k2EDU72B001710>