Bus error while processing comment

[Hiren Patel]

Stefan Bethke wrote:
> I'm on:
> FreeBSD krokodil.zs64.net 8.0-CURRENT FreeBSD 8.0-CURRENT #12: Fri Jun  
> 12 06:29:20 UTC 2009    root at lokschuppen.zs64.net:/usr/obj/usr/src/sys/ 
> EISENBOOT  amd64
> with FreeBSD sources from June 11.
> 

If I'm not mistaken, is -current not the testing/unstable edition? can 
you reproduce this with the latest stable release?

> I've been having this issue for a couple of months now (nagios isn't  
> critical for me, so it had a rather low priority), and I tried various  
> things to fix or at least work around it, including wiping /var/spool/ 
> nagios.  I now remember that nagios kept running every time I wiped  
> the files, and started crashing shortly after I acknowledged a problem.
> 

can you reproduce this with minimal configs?
the thread you posted looks very old, my thinking is that it should have 
been fixed in 3.x if the problem was in 2 so long ago.

> I've compiled nagios with CFLAGS=-g, and I get this in gdb:
> 
> # gdb base/nagios
> GNU gdb 6.1.1 [FreeBSD]
> Copyright 2004 Free Software Foundation, Inc.
> GDB is free software, covered by the GNU General Public License, and  
> you are
> welcome to change it and/or distribute copies of it under certain  
> conditions.
> Type "show copying" to see the conditions.
> There is absolutely no warranty for GDB.  Type "show warranty" for  
> details.
> This GDB was configured as "amd64-marcel-freebsd"...
> (gdb) r /usr/local/etc/nagios/nagios.cfg
> Starting program: /var/ports/work/net-mgmt/nagios-devel/nagios-3.1.2/ 
> base/nagios /usr/local/etc/nagios/nagios.cfg
> [New LWP 100227]
> [New Thread 800c021c0 (LWP 100227)]
> 
> Nagios 3.1.2
> Copyright (c) 2009 Nagios Core Development Team and Community  
> Contributors
> Copyright (c) 1999-2009 Ethan Galstad
> Last Modified: 06-23-2009
> License: GPL
> 
> Website: http://www.nagios.org
> Nagios 3.1.2 starting... (PID=55280)
> Local time is Tue Jul 28 09:21:56 UTC 2009
> Warning: Host 'tivo' has no services associated with it!
> [New Thread 800c511c0 (LWP 100449)]
> 
> Program received signal SIGBUS, Bus error.
> [Switching to Thread 800c021c0 (LWP 100227)]
> 0x000000000044869a in get_next_comment_by_host (host_name=0x800c6f960  
> "slingbox", start=0x800c488e0) at ../common/comments.c:632
> 632		for(;temp_comment && compare_hashdata(temp_comment- 
>  >host_name,NULL,host_name,NULL)<0;temp_comment=temp_comment->nexthash);
> (gdb) bt
> #0  0x000000000044869a in get_next_comment_by_host  
> (host_name=0x800c6f960 "slingbox", start=0x800c488e0) at ../common/ 
> comments.c:632
> #1  0x0000000000447ad1 in delete_host_acknowledgement_comments  
> (hst=0x800c16800) at ../common/comments.c:301
> #2  0x00000000004362e6 in handle_host_state (hst=0x800c16800) at  
> sehandlers.c:731
> #3  0x0000000000412955 in process_host_check_result_3x  
> (hst=0x800c16800, new_state=0,
>    old_plugin_output=0x800c486a0 "CRITICAL - slingbox.lassitu.de: Host  
> unreachable @ 44.128.127.15. rta nan, lost 100%",
>    check_options=0, reschedule_check=1, use_cached_result=1,  
> check_timestamp_horizon=15) at checks.c:3744
> #4  0x00000000004117ac in handle_async_host_check_result_3x  
> (temp_host=0x800c16800, queued_check_result=0x800c6b0c0) at checks.c: 
> 3380
> #5  0x000000000040a7c7 in reap_check_results () at checks.c:206
> #6  0x000000000042574d in handle_timed_event (event=0x800c86fe0) at  
> events.c:1307
> #7  0x0000000000424cc8 in event_execution_loop () at events.c:1002
> #8  0x000000000040a3c2 in main (argc=2, argv=0x7fffffffea58) at  
> nagios.c:833
> (gdb) p temp_comment
> $1 = (comment *) 0x5a5a5a5a5a5a5a5a
> 
> 5a is a guard value filled in by FreeBSD's malloc when certain config  
> flags are set.  Quoting from the man page:
>      J       Each byte of new memory allocated by malloc(), realloc() or
>              reallocf() will be initialized to 0xa5.  All memory  
> returned by
>              free(), realloc() or reallocf() will be initialized to  
> 0x5a.
>              This is intended for debugging and will impact  
> performance nega-
>              tively.
> 
> It seems plausible that some structure was free'd before the loop in  
> get_next_comment processes that structure.
> 
> 
> Stefan
> 


------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july