Subject: Locking panic - vmlocking2 fallout?
To: None <current-users@netbsd.org>
From: Paul Goyette <paul@whooppee.com>
List: current-users
Date: 01/05/2008 05:42:46
I installed a 4.99.48 kernel and userland a couple days ago.  Both were 
built from sources that were updated on 2008-01-04 at 01:26:51 UTC (I 
keep time-stamps!).

Since installing, the machine has "failed" twice, both times during the 
overnight running of /etc/daily processing.  The first time I didn't get 
a crash dump - the machine just hung and had to be forcefully rebooted. 
Last night, I got a crash dump with the following traceback:

(gdb) target kvm netbsd.0.core
#0  0xffffffff802d7bd5 in cpu_reboot (howto=260, bootstr=<value optimized out>)
     at /usr/src/sys/arch/amd64/amd64/machdep.c:652
652                     dumpsys();
(gdb) bt
#0  0xffffffff802d7bd5 in cpu_reboot (howto=260, bootstr=<value optimized out>)
     at /usr/src/sys/arch/amd64/amd64/machdep.c:652
#1  0xffffffff8025d418 in panic (fmt=0xffffffff803ae285 "lock error")
     at /usr/src/sys/kern/subr_prf.c:233
#2  0xffffffff80255ed3 in lockdebug_abort (lock=0xffffffff80579ee8,
     ops=0xffffffff80538620, func=<value optimized out>,
     msg=<value optimized out>) at /usr/src/sys/kern/subr_lockdebug.c:815
#3  0xffffffff80232b17 in mutex_vector_exit (mtx=0x0)
     at /usr/src/sys/kern/kern_mutex.c:716
#4  0xffffffff8023f7ba in sleepq_remove (sq=<value optimized out>,
     l=0xffff8000496a0840) at /usr/src/sys/sys/lwp.h:381
#5  0xffffffff8023fa50 in sleepq_unsleep (l=0xffff8000496a0840)
     at /usr/src/sys/kern/kern_sleepq.c:361
#6  0xffffffff80243fd2 in suspendsched () at /usr/src/sys/kern/kern_synch.c:731
#7  0xffffffff80295183 in vfs_shutdown () at /usr/src/sys/kern/vfs_subr.c:1790
#8  0xffffffff802d7baf in cpu_reboot (howto=<value optimized out>, bootstr=0x0)
     at /usr/src/sys/arch/amd64/amd64/machdep.c:639
#9  0xffffffff8025d418 in panic (fmt=0xffffffff803ac695 "%s\n")
     at /usr/src/sys/kern/subr_prf.c:233
#10 0xffffffff80295697 in vpanic (vp=<value optimized out>,
     msg=0xffffffff803b28e9 "list head mismatch")
     at /usr/src/sys/kern/vfs_subr.c:153
#11 0xffffffff80296546 in getcleanvnode () at /usr/src/sys/kern/vfs_subr.c:221
#12 0xffffffff80296736 in getnewvnode (tag=VT_UFS, mp=0xffff800004233000,
     vops=0xffff800003f53600, vpp=0xffff800049018a58)
     at /usr/src/sys/kern/vfs_subr.c:462
#13 0xffffffff801c81f9 in ffs_vget (mp=0xffff800004233000, ino=1020254,
     vpp=0xffff800049018b28) at /usr/src/sys/ufs/ffs/ffs_vfsops.c:1468
#14 0xffffffff801cf5df in ufs_lookup (v=<value optimized out>)
     at /usr/src/sys/ufs/ufs/ufs_lookup.c:626
#15 0xffffffff802a4b1c in VOP_LOOKUP (dvp=0xffff80004dd027d0, vpp=0x0, cnp=0x0)
     at /usr/src/sys/kern/vnode_if.c:131
#16 0xffffffff80292e95 in lookup (ndp=0xffff800049018d10)
     at /usr/src/sys/kern/vfs_lookup.c:696
#17 0xffffffff80293501 in namei (ndp=0xffff800049018d10)
     at /usr/src/sys/kern/vfs_lookup.c:332
#18 0xffffffff8029bbd0 in do_sys_stat (l=0xffff80004948c1c0,
     path=<value optimized out>, nd_flags=<value optimized out>,
     sb=0xffff800049018dd0) at /usr/src/sys/kern/vfs_syscalls.c:2414
#19 0xffffffff8029bc3a in sys___lstat30 (l=0x0, uap=0xffff800049018e80,
     retval=<value optimized out>) at /usr/src/sys/kern/vfs_syscalls.c:2456
#20 0xffffffff802d9191 in syscall (frame=0xffff800049018f20)
     at /usr/src/sys/arch/amd64/amd64/syscall.c:133
#21 0xffffffff801005e4 in Xsyscall ()
(gdb)

This machine has "only" 2GB of RAM, so I can keep the crash dump file 
around for a while.  If there's anything more that can be gathered from 
the crash, I'll be happy to provide it.


----------------------------------------------------------------------
|   Paul Goyette   | PGP DSS Key fingerprint: |  E-mail addresses:   |
| Customer Service | FA29 0E3B 35AF E8AE 6651 |  paul@whooppee.com   |
| Network Engineer | 0786 F758 55DE 53BA 7731 | pgoyette@juniper.net |
----------------------------------------------------------------------

!DSPAM:477f89561954589119237!