On Wed, 2004-03-31 at 04:24, Miles Davis wrote: > > Have you ever made any progress with your problem? I'm pretty sure I'm > hitting the same thing; 0.70rc, RH 9, though I'm using exim with the > exiscan patch instead of sendmail. Clamd will run for anywhere from 1 > minute to an hour and segfault. > > I tried setting the LD_ASSUME_KERNEL env var as I saw in one of your > previous posts, but now clamd seems to lock up after a while instead of > segfaulting -- not sure if you were still using LD_ASSUME_KERNEL when you > described the same thing above. > > Attached is debug output from two different runs when clamd segfaulted.
Your crash output shows they both crashed due to a known bug in libclamav which Nigel has fixed in CVS. -trog
signature.asc
Description: This is a digitally signed message part