On Fri, 16 Feb 2007, 07:36 GMT-08 Bill Landry wrote: (Rolf E. Sonneveld):
>> Did you see any performance or crash issues after the >> recompile-without-experimental? >> >> /rolf > No issues to report yet - hope it stays that way... Enabling the experimental code in 0.90 is dangerous because there has been introduced a critical bug after the last release candidate has been released, unfortunately without giving us a chance to test before releasing it as final (Bug 295). When I first tried 0.90 final via procmail on incoming mails, I had been affected immediately and had to revert back to rc3 shortly because of crashes, eating up CPU load and time, resulting in terminating the clamscan process by procmail after living something like 6 minutes without coming back. I will try to recompile a experimental-enabled ClamAV 0.90 with an previous entconv.c and see if that solves this problem. Unfortunately, there is no version control system currently available, being down since 4 days now, so the only change to get older source code is to look into http://www.clamav.net/snapshot/ regards, rob. _______________________________________________ Help us build a comprehensive ClamAV guide: visit http://wiki.clamav.net http://lurker.clamav.net/list/clamav-users.html
