security update as time permit it Reproducible: Always
dev-perl/HTML-Parser-3.63 is already in the tree. perl bug: <http://rt.perl.org/rt3/Public/Bug/Display.html?id=69973> gmane link: <http://article.gmane.org/gmane.comp.lang.perl.perl5.porters/74707>
Thanks. My understanding is that only perl-5.10.1 is affected by the crash issue. Can you confirm that our latest stable version of perl is not affected? If yes, we'll stabilize HTML-Parser-3.63 only; if no, we'll have to patch our latest stable perl as well.
I wonder if this is actually a security bug at all. Someone else from security, please advise.
(In reply to comment #2) > My understanding is that only perl-5.10.1 is affected by the crash > issue. Can you confirm that our latest stable version of perl is not affected? Yes, no segfault with 5.8.8 and the test script from rt69973. perl-5.10.1 is fixed too.
CVE-2009-3627 (http://nvd.nist.gov/nvd.cfm?cvename=CVE-2009-3627): The decode_entities function in util.c in HTML-Parser before 3.63 allows context-dependent attackers to cause a denial of service (infinite loop) via an incomplete SGML numeric character reference, which triggers generation of an invalid UTF-8 character.
Patch here: http://github.com/gisle/html-parser/commit/b9aae1e43eb2c8e989510187cff0ba3e996f9a4c
*** Bug 293128 has been marked as a duplicate of this bug. ***
This is a bug in stable HTML-Parser that triggers a vulnerability in unstable Perl. Since the bug in Perl is being fixed (bug 293128), there is nothing to do on this bug for security. As a precautionary measure I am adding architectures to stable the latest HTML-Parser release via this bug.
Arches, please test and mark stable: =dev-perl/HTML-Parser-3.64 Target keywords : "alpha amd64 arm hppa ia64 m68k ppc ppc64 s390 sh sparc x86"
Stable for HPPA.
amd64/x86 stable
arm stable
alpha/ia64/m68k/s390/sh/sparc stable
ppc64 done
ppc stable
(In reply to comment #8) > This is a bug in stable HTML-Parser that triggers a vulnerability in unstable > Perl. Since the bug in Perl is being fixed (bug 293128), there is nothing to do > on this bug for security. > As a precautionary measure I am adding architectures to stable the latest > HTML-Parser release via this bug. Latest HTML-Parser is stable for all arches, the old versions were removed. I am closing this bug now.
This security bug was accidentally closed a while ago. I vote NO.
No, too. Closing noglsa.