Re: fs_fs core dumps in checksum code

On 04/13/2012 06:33 PM, Daniel Shahaf wrote:
> Blair Zajac wrote on Fri, Apr 13, 2012 at 13:08:12 -0700:
>> To do valgrind well, do I need to recompile APR with specific flags to
>> enable pool debugging?
>>
>
> Other than --enable-pool-debugging=N (aka -DAPR_POOL_DEBUG=N) (where
> N is placeholder for a number) ?

Yes, besides that?

Besides valgrind and enabling pool debugging, does anybody have any
recommendations for memory debugging besides valgrind in a production
environment, something like using a library that you LD_PRELOAD to
intercept malloc(), free(), etc?

These core dumps occur infrequently so if there's a way to leave some
debugging in that doesn't cost too much of a hit, that would be great.

Blair
Received on 2012-04-14 03:45:16 CEST