[Nut-upsuser] upsd crashes with a "broken pipe" error

Zach La Celle lacelle at roboticresearch.com
Mon Jan 3 13:29:37 UTC 2011


On 12/29/2010 10:00 AM, Zach La Celle wrote:
> On 12/29/2010 08:34 AM, Charles Lepple wrote:
>> On Dec 27, 2010, at 9:36 AM, Zach La Celle wrote:
>>
>>> I ran this in debug mode and captures the backtrace.
>>>
>>> root@*********:/etc/nut# upsd -D
>>> Network UPS Tools upsd 2.4.3
>>>   0.000000     listening on 0.0.0.0 port 3493
>>>   0.000354     Connected to UPS [rack1ups]: apcsmart-rack1ups
>>>   2.550554     User upsmon at 127.0.0.1 logged into UPS [rack1ups]
>>> *** glibc detected *** upsd: free(): invalid next size (fast): 
>>> 0x00000000012c9870 ***
>>
>> Can you give us some background information about this system? What 
>> OS and version, who built the package, etc.
>>
>> Do you have valgrind available?
>>
>> Your version of glibc probably has some more thorough memory 
>> corruption detection algorithms than the default - "man malloc" on 
>> one of my systems suggests that setting the MALLOC_CHECK_ environment 
>> variable to either 1 or 2 will print some additional diagnostics.
>>
>
> Output of uname -a:
> Linux www 2.6.32-27-server #49-Ubuntu SMP Thu Dec 2 02:05:21 UTC 2010 
> x86_64 GNU/Linux
>
> Information about glibc:
> GNU C Library (Ubuntu EGLIBC 2.11.1-0ubuntu7.6) stable release version 
> 2.11.1, by Roland McGrath et al.
> Compiled by GNU CC version 4.4.3.
> Compiled on a Linux >>2.6.24-28-server<< system on 2010-11-17.
>
> I didn't know about that MALLOC_CHECK_ variable: I've set it and am 
> running the software again to see if I get a better error.
>
> I can install valgrind if necessary.  Let me see what happens this time.
>
> _______________________________________________
> Nut-upsuser mailing list
> Nut-upsuser at lists.alioth.debian.org
> http://lists.alioth.debian.org/mailman/listinfo/nut-upsuser

After setting the variable, the only further information I have is this:
upsd: free(): invalid pointer: 0x000000000
1d20f30 ***
Segmentation fault

I'll install valgrind and see if I can catch any memory leaks that way: 
otherwise, I'm not sure exactly what else to do.



More information about the Nut-upsuser mailing list