Nigel Horne wrote: > Folks, > > 0.95 RC1 was published on Wednesday 25/2/09. > > It would help us a lot to see if this problem still exists in the new > ewlease if as many people as possible were to test this release > candidate by downloading it from www.clamav.net. If you have a test > machine/model/network please help us by loading ClamAV 0.95RC1 and > testing it. > > All bug reports should be filed at http://bugs.clamav.net. > > The full release of ClamAV 0.95 is scheduled for 16th March. > > Thank you for your continued support and help, > > -Nigel
Hi Nigel, ClamAV 0.95rc1 compiles and installs fine on Fedora 10. However, the "clamscan -d" test now appears to be more restrictive in what it finds and reports as "malformed". I have created a file of what was not detected with the ClamAV 0.94.2 version of "clamscan -d", but is now detected as malformed with the ClamAV 0.95rc1 version of "clamscan -d". The only file that reported errors was SecuriteInfo's securiteinfo.hdb signature database file. The reported malformed lines can be viewed at: http://www.inetmsg.com/pub/securiteinfo.hdb-malformed.txt and an updated version of the file with the malformed lines removed is available at: http://www.inetmsg.com/pub/securiteinfo.hdb Before I report this to Arnaud Jacques (of SecuriteInfo), could these now reported malformed lines possibly be the cause of the clamd crashes some of us have been experiencing? Bill ps, I've cross-posted this to both the SaneSecurity and ClamAV-Users lists. _______________________________________________ Help us build a comprehensive ClamAV guide: visit http://wiki.clamav.net http://www.clamav.net/support/ml
