Date: Wed, 1 Dec 2004 19:41:22 -0400 (AST) From: "Marc G. Fournier" <scrappy@hub.org> To: freebsd-stable@freebsd.org Subject: page fault panic in 4.10-STABLE ... Message-ID: <20041201193615.T18428@ganymede.hub.org>
next in thread | raw e-mail | index | archive | help
Since it looks like a 4.11 is happening, here is a quick bug report about
a panic that I seem to be getting "out of the blue" on one of my servers
... I've got two *good* core dumps from it, here is a bt of one of them:
(kgdb) where
#0 dumpsys () at /usr/src/sys/kern/kern_shutdown.c:487
#1 0x801650bb in boot (howto=260) at /usr/src/sys/kern/kern_shutdown.c:316
#2 0x8016552d in panic (fmt=0x80299959 "%s") at /usr/src/sys/kern/kern_shutdown.c:595
#3 0x8025ba41 in trap_fatal (frame=0xb8319754, eva=0) at /usr/src/sys/i386/i386/trap.c:974
#4 0x8025b6ad in trap_pfault (frame=0xb8319754, usermode=0, eva=0) at /usr/src/sys/i386/i386/trap.c:867
#5 0x8025b20b in trap (frame={tf_fs = -1976500200, tf_es = -1995046896, tf_ds = -1995046896, tf_edi = -1386893312, tf_esi = 0, tf_ebp = -1204709440, tf_isp = -1204709504,
tf_ebx = 4096, tf_edx = -1386893312, tf_ecx = 1024, tf_eax = -1386893312, tf_trapno = 12, tf_err = 0, tf_eip = -2145016810, tf_cs = 8, tf_eflags = 66054, tf_esp = -1204709200,
tf_ss = -1204709228}) at /usr/src/sys/i386/i386/trap.c:466
#6 0x8025a416 in generic_bcopy ()
#7 0x80207a95 in ffs_write (ap=0xb8319858) at /usr/src/sys/ufs/ufs/ufs_readwrite.c:547
#8 0x801a3255 in union_write (ap=0xb831989c) at vnode_if.h:363
#9 0x8021ebd8 in vnode_pager_generic_putpages (vp=0xba70e240, m=0xb8319970, bytecount=4096, flags=12, rtvals=0xb8319940) at vnode_if.h:363
#10 0x801a30c6 in union_putpages (ap=0xb8319904) at /usr/src/sys/miscfs/union/union_vnops.c:1047
#11 0x8021e9fa in vnode_pager_putpages (object=0xc091d564, m=0xb8319970, count=1, sync=12, rtvals=0xb8319940) at vnode_if.h:1147
#12 0x8021b93f in vm_pageout_flush (mc=0xb8319970, count=1, flags=12) at /usr/src/sys/vm/vm_pager.h:147
#13 0x802188ab in vm_object_page_collect_flush (object=0xc091d564, p=0x83ddb5f0, curgeneration=73, pagerflags=12) at /usr/src/sys/vm/vm_object.c:806
#14 0x80218489 in vm_object_page_clean (object=0xc091d564, start=0, end=0, flags=4) at /usr/src/sys/vm/vm_object.c:605
#15 0x801959d8 in vfs_msync (mp=0x8b12e200, flags=2) at /usr/src/sys/kern/vfs_subr.c:2731
#16 0x80196b50 in sync (p=0x802d9720, uap=0x0) at /usr/src/sys/kern/vfs_syscalls.c:582
#17 0x80164e56 in boot (howto=256) at /usr/src/sys/kern/kern_shutdown.c:235
#18 0x8016552d in panic (fmt=0x80299959 "%s") at /usr/src/sys/kern/kern_shutdown.c:595
#19 0x8025ba41 in trap_fatal (frame=0xb8319bc4, eva=0) at /usr/src/sys/i386/i386/trap.c:974
#20 0x8025b6ad in trap_pfault (frame=0xb8319bc4, usermode=0, eva=0) at /usr/src/sys/i386/i386/trap.c:867
#21 0x8025b20b in trap (frame={tf_fs = -1975451624, tf_es = 1812135952, tf_ds = -1204748272, tf_edi = -1525469184, tf_esi = 0, tf_ebp = -1204708304, tf_isp = -1204708368,
tf_ebx = 8192, tf_edx = -1525469184, tf_ecx = 2048, tf_eax = -1525469184, tf_trapno = 12, tf_err = 0, tf_eip = -2145016810, tf_cs = 8, tf_eflags = 66054, tf_esp = -1204708064,
tf_ss = -1204708092}) at /usr/src/sys/i386/i386/trap.c:466
#22 0x8025a416 in generic_bcopy ()
#23 0x80207a95 in ffs_write (ap=0xb8319cc8) at /usr/src/sys/ufs/ufs/ufs_readwrite.c:547
#24 0x801a3255 in union_write (ap=0xb8319d0c) at vnode_if.h:363
#25 0x8021ebd8 in vnode_pager_generic_putpages (vp=0xc041aa80, m=0xb8319de4, bytecount=8192, flags=12, rtvals=0xb8319db0) at vnode_if.h:363
#26 0x801a30c6 in union_putpages (ap=0xb8319d74) at /usr/src/sys/miscfs/union/union_vnops.c:1047
#27 0x8021e9fa in vnode_pager_putpages (object=0xc09133f4, m=0xb8319de4, count=2, sync=12, rtvals=0xb8319db0) at vnode_if.h:1147
#28 0x8021b93f in vm_pageout_flush (mc=0xb8319de4, count=2, flags=12) at /usr/src/sys/vm/vm_pager.h:147
#29 0x802188ab in vm_object_page_collect_flush (object=0xc09133f4, p=0x83f147f0, curgeneration=20, pagerflags=12) at /usr/src/sys/vm/vm_object.c:806
#30 0x80218489 in vm_object_page_clean (object=0xc09133f4, start=0, end=0, flags=4) at /usr/src/sys/vm/vm_object.c:605
#31 0x801959d8 in vfs_msync (mp=0x8b12e200, flags=2) at /usr/src/sys/kern/vfs_subr.c:2731
#32 0x80195da6 in sync_fsync (ap=0xb8319f7c) at /usr/src/sys/kern/vfs_subr.c:2992
#33 0x8019408f in sched_sync () at vnode_if.h:558
(kgdb) up 7
#7 0x80207a95 in ffs_write (ap=0xb8319858) at
/usr/src/sys/ufs/ufs/ufs_readwrite.c:547
547 error =
(kgdb) list
542
543 size = BLKSIZE(fs, ip, lbn) - bp->b_resid;
544 if (size < xfersize)
545 xfersize = size;
546
547 error =
548 uiomove((char *)bp->b_data + blkoffset, (int)xfersize, uio);
549 if ((ioflag & (IO_VMIO|IO_DIRECT)) &&
550 (LIST_FIRST(&bp->b_dep) == NULL)) {
551 bp->b_flags |= B_RELBUF;
The second one looks to be exactly the same thing ...
So far, from testing, it seems to be triggered when I start up the Slony
Replication daemon for PostgreSQL, *but* ... I don't run that on a union
file system, it is on the main hard drive itself ... I'm running the
server right now without Slony, to see if it stays up longer, or if Slony
just *appears* to be the culprit ...
Anything I can provide out of those core files that might help any?
Thanks ...
----
Marc G. Fournier Hub.Org Networking Services (http://www.hub.org)
Email: scrappy@hub.org Yahoo!: yscrappy ICQ: 7615664
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20041201193615.T18428>