Your message dated Tue, 07 Nov 2006 20:55:38 +0200
with message-id <[EMAIL PROTECTED]>
and subject line libsasl2: Please use /dev/urandom instead of /dev/random
has caused the attached Bug report to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what I am
talking about this indicates a serious mail system misconfiguration
somewhere. Please contact me immediately.)
Debian bug tracking system administrator
(administrator, Debian Bugs database)
--- Begin Message ---
Package: libsasl2
Version: 2.1.19-1.5sarge1
Severity: grave
Justification: renders package unusable
The use of /dev/random blocks all processes that use libsasl2 when the
random pool is (almost) empty. This makes the packages (and the services
that depend on it) completely unusable.
Please use /dev/urandom instead (or at least make this a runtime option).
Thanks!
Ivo De Decker
-- System Information:
Debian Release: 3.1
Architecture: i386 (i686)
Kernel: Linux 2.6.17.8-ugent-pe
Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968)
Versions of packages libsasl2 depends on:
ii libc6 2.3.2.ds1-22sarge3 GNU C Library: Shared libraries an
ii libdb4.2 4.2.52-18 Berkeley v4.2 Database Libraries [
-- no debconf information
--- End Message ---
--- Begin Message ---
Version: 2.1.22-1
Hi,
This bug has been fixed in version 2.1.22-1 (and earlier pre-versions in
experimental). The change has been documented in README.Debian.
On Mon, 30 Oct 2006 12:14:21 -0800, Steve Langasek wrote:
> SASL is a security-sensitive library. Far from being "grave", any
> request that it be changed to not require good randomness as input is
> a wishlist bug, and one I would recommend the maintainers not
> implement.
The security of the Linux random number generator has been challenged in
a paper (Analysis of the Linux Random Number Generator, Gutterman,
Pinkas, Reinman, March 6, 2006) which has in turn been criticised by
kernel developers. There is a summary with links to the paper and the
discussion at http://lwn.net/Articles/184925/
The randomness produced by /dev/urandom is the same as that produced
by /dev/random up to the point when /dev/random would block. After
this, /dev/urandom will continue to supply random numbers, but they will
be cryptographically weaker.
Exactly how much weaker, and how this affects security, is a hard
question to answer. To my knowledge, there have been no attempts at
answering it. However, it seems that it is not a practical concern, and
that the blocking behaviour of /dev/random is a much greater (and
common) problem than the potential, theoretical, cryptographic flaws
of /dev/urandom.
As a rule of thumb, it has been suggested that /dev/random is needed
only when an application is doing things like generating crypto-keys or
certificates which are used over a long period of time. Cyrus SASL
doesn't, and thus it could well use /dev/urandom.
Thanks,
--
Fabian Fagerholm <[EMAIL PROTECTED]>
signature.asc
Description: This is a digitally signed message part
--- End Message ---
