In the old log, we can see the affected vfswrap_getwd function being called and flagged in a few places:
==9522== 36 bytes in 3 blocks are definitely lost in loss record 123 of 498
==9522== at 0x4C31D2F: realloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==9522== by 0x8A8B15F: getcwd (getcwd.c:84)
==9522== by 0x55A452D: vfswrap_getwd (vfs_default.c:2227)
In thw new log file, there is no "vfs" match anywhere.
This is the call I used:
smbclient //localhost/pub -U ubuntu%ubuntu -c "pwd;dir;cd dir1;dir;pwd;cd dir11; pwd; dir; cd /; cd dir2; pwd; dir; cd /"
And pub was created as:
# tree /pub
/pub
├── dir1
│ └── dir11
└── dir2
[pub]
path = /pub
read only = no
guest ok = no
And I had this global option set, to not use a cache and provoke more leaks:
[global]
...
getwd cache = no
Ok, got something.
First, the full logs: paste.ubuntu. com/p/CCR4fbFf8 4/ paste.ubuntu. com/p/Hmw6QGHVq P/
old samba: http://
new samba: http://
In the old log, we can see the affected vfswrap_getwd function being called and flagged in a few places:
==9522== 36 bytes in 3 blocks are definitely lost in loss record 123 of 498 valgrind/ vgpreload_ memcheck- amd64-linux. so) c:2227)
==9522== at 0x4C31D2F: realloc (in /usr/lib/
==9522== by 0x8A8B15F: getcwd (getcwd.c:84)
==9522== by 0x55A452D: vfswrap_getwd (vfs_default.
In thw new log file, there is no "vfs" match anywhere.
This is the call I used:
smbclient //localhost/pub -U ubuntu%ubuntu -c "pwd;dir;cd dir1;dir;pwd;cd dir11; pwd; dir; cd /; cd dir2; pwd; dir; cd /"
And pub was created as:
# tree /pub
/pub
├── dir1
│ └── dir11
└── dir2
[pub]
path = /pub
read only = no
guest ok = no
And I had this global option set, to not use a cache and provoke more leaks:
[global]
...
getwd cache = no