New autofs in sid, please push to testing
A message that is not in the archive (sent off-list or not yet received): [email protected]
A message that is not in the archive (sent off-list or not yet received): [email protected]
Quoted 6 lines · Steve Langasek
Quoted 4 lines
I've uploaded autofs 4.1.3+4.1.4beta2-8 to sid, which fixes (to the best of my knowledge -- I can't reproduce the problems myself) two important bugs, and contains no other changes. vorlon promised me a freeze exception for the upload, so letting it through to sarge would be nice :-) Thanks.
Hmm, memory leak:
Quoted 8 lines
[...]
++int
is_local_mount(const char *hostpath)
++{[...]
Memory is allocated for hostname on every call, and never freed.
I've fixed this in 4.1.3+4.1.4beta2-10 (just uploaded); it's a simple fix (free(hostname) added three or four places -- it's being used in printf()s etc. in various code branches, so I can't just free it once).
Quoted 6 lines
The
get_best_mountfunction worries me, as it's writing to a caller-provided pointer (char *what) with no bounds checking. This appears to be the intended *use* of the pointer, so this isn't a blocking issue, it's just something that ought to be cleaned up at some point (including making sure there's no exploitable buffer overflow here when using things like LDAP or Hesiod maps).
Cc-ing the autofs list :-)
/* Steinar */
Signature · 1 line
Homepage: http://www.sesse.net/
[email protected]
maintainerQuoted 1 line · Steinar H. Gunderson
On Thu, May 12, 2005 at 09:41:52PM -0700, Steve Langasek wrote:
snip ...
Quoted 6 lines
Quoted 6 lines
The
get_best_mountfunction worries me, as it's writing to a caller-provided pointer (char *what) with no bounds checking. This appears to be the intended *use* of the pointer, so this isn't a blocking issue, it's just something that ought to be cleaned up at some point (including making sure there's no exploitable buffer overflow here when using things like LDAP or Hesiod maps).
Perhaps.
I'll have a look but bounds checking should be done way before this so I'm not so concerned about this. Checking that it is done at the right spot is the important thing (I think it is in the latest release).
Ian