[Pvfs2-developers] server side memory problems in trunk?

[Sam Lang]

I was just seeing something similar with lookup.  It turns out the  
PINT_decode in server_state_machine_start does some in-place  
decoding, so we can't free the unexpected buffer right after that  
(call the BMI_unexpected_free).  I'm just going to move the free back  
down to server_state_machine_complete.

-sam

On Aug 23, 2006, at 3:17 PM, Phil Carns wrote:

> I am having some problems with builds from the latest cvs head  
> snapshot.    The server keeps crashing while I try to run an ACL  
> test case, but I think that the problem is more fundamental than  
> that.  If I watch the server in valgrind, this is a sampling of  
> errors that come out when just mounting and performing an "ls - 
> alh".  This is with a newly created storage space:
>
> ==7512== Invalid read of size 1
> ==7512==    at 0x401C8E7: strncmp (in
> /usr/lib/valgrind/x86-linux/vgpreload_memcheck.so)
> ==7512==    by 0x80B1E3A: PINT_eattr_is_prefixed (pvfs2-server.h:133)
>
> ==7512== Invalid read of size 1
> ==7512==    at 0x8056866: trove_keyval_read_list (trove.c:498)
>
> ==7512== Invalid read of size 1
> ==7512==    at 0x401CA60: memcpy (in
> /usr/lib/valgrind/x86-linux/vgpreload_memcheck.so)
> ==7512==    by 0x805CB6B: dbpf_keyval_read_list_op_svc (dbpf- 
> keyval.c:916)
>
> ==7512== Invalid read of size 1
> ==7512==    at 0x808E4B1: PINT_string_next_segment (str-utils.c:159)
>
> ==7512== Invalid read of size 1
> ==7512==    at 0x401C638: strlen (in
> /usr/lib/valgrind/x86-linux/vgpreload_memcheck.so)
> ==7512==    by 0x80A50C4: lookup_read_directory_entry (lookup.sm:545)
>
> ==7512== Invalid read of size 1
> ==7512==    at 0x805647B: trove_keyval_read (trove.c:291)
>
> ...
>
> Anyone know what's going on?
>
> -Phil
> _______________________________________________
> Pvfs2-developers mailing list
> Pvfs2-developers at beowulf-underground.org
> http://www.beowulf-underground.org/mailman/listinfo/pvfs2-developers
>