[Gluster-devel] glfs_readdir_r is painful

Eric Blake eblake at redhat.com
Wed Oct 30 17:18:29 UTC 2013

I'm trying to use glusterfs-api, but ran into some questions on usage
(currently targetting Fedora 19's glusterfs-api-devel-3.4.1-1.fc19.x86_64).

Why is there no glfs_readdir?  POSIX readdir_r() is broken by design; it
assumes that the caller is providing a large enough buffer, but does not
take a buffer size as a parameter, so it is very easy to trigger buffer
overflow bugs, since struct dirent contains only a variable-length
d_name member.

Read this for more details about how hard it is to use readdir_r
properly: http://womble.decadent.org.uk/readdir_r-advisory.html

Read this for proof that POSIX is considering ditching readdir_r
altogether and merely require sane thread-safety requirements on
readdir() (since at least that way there are no buffer overflow
attacks): http://austingroupbugs.net/view.php?id=696

The only safe way to use readdir_r is to know the maximum d_name that
can possibly be returned, but there is no glfs_fpathconf() for
determining that information.  Your example usage of glfs_readdir_r()
suggests that 512 bytes is large enough:
but I don't know if that is true.  You _do_ have the advantage that
since every brick backing a glusterfs volume is using an xfs file
system, then you only have to worry about the NAME_MAX of xfs - but I
don't know that value off the top of my head.  Can you please let me
know how big I should make my struct dirent to avoid buffer overflow,
and properly document this in <glusterfs/api/glfs.h>?  Furthermore, can
you please provide a much saner glfs_readdir() so I don't have to worry
about contortions of using a broken-by-design function?

Eric Blake   eblake redhat com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 621 bytes
Desc: OpenPGP digital signature
URL: <http://supercolony.gluster.org/pipermail/gluster-devel/attachments/20131030/9564b6b3/attachment-0001.sig>

More information about the Gluster-devel mailing list