So from the sounds of things we need to do evolution gcc java Put them in the testing area for a week or two, then push them out into the errata. Does that sound ok? Is there any other packages that I'm missing? Troy Jan Iven wrote: > On 20/03/07 20:36, Stephan Wiesand wrote: >> On Tue, 20 Mar 2007, Troy Dawson wrote: > >> Wouldn't this argument apply to the latest gcc updates as well? >> >> NB did the patch to remove that new warning on SL3 Jan told us about make >> it into this update as well? (looks like CERN rolled it out unmodified?) > > Yes, the issue with warnings on system header files appears to be fixed > (still wondering why nobody else complained about this, our users were > not too happy :-) > > * Fri Aug 11 2006 Jakub Jelinek <[log in to unmask]> 3.2.3-57 > - don't warn about unused parameters in system headers > (Alexandre Oliva, #201342, IT#99250) > >>> in the list as well too. It looks like if I push out the jdk 1.5 along with >>> the 1.4.2 only a small percentage of people will have the dissapearing java >>> problem. >>> But, the java has to get pushed out sometime. > > Hm -- for what it is worth (in case you need arguments for roll-out), in > our 1.4.2 JRE changelog, I have: > > * Wed Dec 20 2006 Jan Iven <[log in to unmask]> - 1.4.2_13-1 > - update to 1.4.2_13, sec fix for BID21673==FrSIRT5073, FrSIRT5074 > - move away broken Kerberos binaries (and manpages) > > and 1.5.0_08 fixes the same issue ( > http://sunsolve.sun.com/search/document.do?assetkey=1-26-102731-1 ) > as well as > http://sunsolve.sun.com/search/document.do?assetkey=1-26-102622-1 > > 1.5.0_09 fixes a GIF-based exploit: > http://sunsolve.sun.com/search/document.do?assetkey=1-26-102760-1 > > we went directly to 1.5.0_10 since that has some TZ fixes as well. > > Unfortunately the SUN release notes do not clearly highlight security > issues, so I don't know whether _11 is actually also a security update > (fixes one segv and has even more timezone stuff).. > > Regards > Jan > > > > >> I think it has caused problems on LCG nodes last time. It may be a good >> idea to allow sites to test this update for a while. >> >>> Troy >>> -- __________________________________________________ Troy Dawson [log in to unmask] (630)840-6468 Fermilab ComputingDivision/LCSI/CSI DSS Group __________________________________________________