I responded in private to him, however fix should be simply to raise maxusers in your kernel and recompile. Use 'netstat -m' to check the status of the membufs and see if you are running out or how close you are during peak operation times both before and after you update the kernel. ------------------------------------------------------------ Stephen Comoletti - Network Engineer / Systems Administrator Delanet Inc. http://www.delanet.com Frontline Communications Corp. http://www.fcc.net phone: (302) 326-5800 fax: (302) 326-5802 x312 262 Quigley Blvd, New Castle, DE 19720, USA ------------------------------------------------------------ Tim Hunter writes: > Didn't someone already address this as a possible hardware/kernel/config problem? > > I have never seen any type of error message like this. > > ----- Original Message ----- > From: dengyj > To: [EMAIL PROTECTED] > Sent: Wednesday, October 18, 2000 6:33 PM > Subject: 1019 > > > hi: > we have two mail severs use qmail(+freebsd) to send mail(ezmlm) > but in this few days ,it often shutdown. > show: > db> > Fatal trap 12: page fault while in kernel mode > fault virtual address =0x18 > fault code =supervisor write,page not present > current proccess=19895(qmail-queue) > > why?? we use PC SERVER ,connect local 40,remote 400
