From owner-freebsd-questions  Wed Nov  8 07:58:04 1995
Return-Path: owner-questions
Received: (from root@localhost)
          by freefall.freebsd.org (8.6.12/8.6.6) id HAA09795
          for questions-outgoing; Wed, 8 Nov 1995 07:58:04 -0800
Received: from niceguy.isocor.ie (niceguy.isocor.ie [193.178.34.157])
          by freefall.freebsd.org (8.6.12/8.6.6) with ESMTP id HAA09780
          for <freebsd-questions@freebsd.org>; Wed, 8 Nov 1995 07:57:43 -0800
Received: (from alan@localhost) by niceguy.isocor.ie (8.6.11/8.6.9) id PAA02454; Wed, 8 Nov 1995 15:57:32 GMT
Date: Wed, 8 Nov 1995 15:57:31 +0000 (GMT)
From: Alan Byrne <alan@niceguy.isocor.ie>
To: freebsd-questions@freebsd.org
Subject: High NFS activity causing crash
Message-ID: <Pine.BSF.3.91.951108152824.1534C-100000@niceguy.isocor.ie>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-questions@freebsd.org
Precedence: bulk


Hi,

I am having severe system crashes on a FreeBSD 2.0.5-950622-SNAP system 
(I know, I should upgrade to a later release ! ).
I think they are related to NFS activity when the system is heavily loaded.
I first discovered the problem due to some corruption in some RCS 
controlled souce files (loads of null chars and/or truncated files).
I have been able to narrow it down to NFS writes from the client (SCO) to 
the server (FreeBSD). I can duplicate the problem by copying files from 
an NFS mounted disk and back again (it may take a few hours depending on 
the load).
I am about to try and fine tune my NFS client/server setup to slow things 
down a bit, just in case it's due to duplicate nfs requests not being 
handled correctly by the server, but that shouldn't cause the system to 
crash...

Is this a known problem with that verion of FreeBSD, will upgrading to 
latest SNAP solve it ??

System is a P-90, 32MB mem, SMC PCI network card, Seagate Barracuda drives.

I enabled kernel crash dumps and installed a debug kernel, so here is the 
backtrace from kgdb......


(kgdb) where
#0  boot (arghowto=256) at ../../i386/i386/machdep.c:870
#1  0xf0113683 in panic (fmt=0xf01755c5 "blkfree: freeing free block")
    at ../../kern/subr_prf.c:128
#2  0xf017579b in ffs_blkfree (ip=0xf0a68500, bno=8480, size=8192)
    at ../../ufs/ffs/ffs_alloc.c:1220
#3  0xf0177717 in ffs_truncate (ap=0xefbffd00) at ../../ufs/ffs/ffs_inode.c:334
#4  0xf017e192 in ufs_setattr (ap=0xefbffd4c) at ./vnode_if.h:827
#5  0xf014ccdd in nfsrv_setattr (nfsd=0xf0a3c400, mrep=0xf0be7380, 
    md=0xf0be7380, dpos=0xf08b4000 "", cred=0xf0a3c42c, nam=0xf0ca2e00, 
    mrq=0xefbffe5c) at ./vnode_if.h:188
#6  0xf015d673 in nfssvc_nfsd (nsd=0xefbffea4, argp=0x183cc "8\026", 
    p=0xf0a53700) at ../../nfs/nfs_syscalls.c:545
#7  0xf015cf7a in nfssvc (p=0xf0a53700, uap=0xefbfff94, retval=0xefbfff8c)
    at ../../nfs/nfs_syscalls.c:276
#8  0xf0199557 in syscall (frame={tf_es = 39, tf_ds = 39, tf_edi = 6, 
      tf_esi = 2, tf_ebp = -272638352, tf_isp = -272629788, 
      tf_ebx = -272638300, tf_edx = 0, tf_ecx = -10, tf_eax = 155, 
      tf_trapno = 646, tf_err = 646, tf_eip = 13141, tf_cs = 31, 
      tf_eflags = 646, tf_esp = -272638516, tf_ss = 39})
    at ../../i386/i386/trap.c:828
#9  0xf018f20b in Xsyscall ()
Cannot access memory at address 0xefbfde70.


Yours hopefully

Alan Byrne
Network Administrator 
ISOCOR Ireland