>Number:         8760
>Category:       kern
>Synopsis:       page fault trap while handling many incoming mails
>Confidential:   no
>Severity:       critical
>Priority:       high
>Responsible:    kern-bug-people (Kernel Bug People)
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Sun Nov  7 23:48:00 1999
>Last-Modified:
>Originator:     Martin J. Laubach
>Organization:
	
>Release:        1999-11-05
>Environment:
System: NetBSD cactus.emsi.priv.at 1.4M NetBSD 1.4M (CACTUS) #1: Sun Nov 7 23:47:26 CET 1999 mjl@asparagus:/home/temp/devel/cvs/src/sys/arch/i386/compile/CACTUS i386


>Description:
  While handling a larger amount of incoming mail, my mail hub crashed with

	fault(0xf025b9c0, 0xf0996000, 0, 1) -> 2
	kernel: page fault trap, code=0
	Stopped in sendmail at memcpy+0x1a: repe movsl (%esi), %es:(%edi)

		esi = f0996000
		edi = f2827000
		es  = f0380010

  A "show all procs" in ddb shows about 75 sendmail processes around,
most waiting on select, some on biowait, one on vflushb and one on wait.
And one running.

>How-To-Repeat:
  Have a larger number of mails queue up at a second machine, then
trigger a delivery "sendmail -v -q". After about 50 mails, the mail
hub crashes.

  This is probably related to kern/8743, but since this is a different
machine, and the symptoms are slightly different, I'm filing a separate
PR.

>Fix:
>Audit-Trail:
>Unformatted: