Re: VFS panic(?).

Hi,

On Jun 26, 2005, at 5:47 AM, Pawel Jakub Dawidek wrote:

> Hi.
>
> On a diskless machine (all file systems are NFS-mounted), when I start
> syslogd and run:
>
>     # tail -f /var/log/all.log
>
> I get this panic:
>
> panic: lockmgr: locking against myself
> cpuid = 1
> KDB: enter: panic
> [thread pid 697 tid 100159 ]
> Stopped at      kdb_enter+0x2b: nop
> db> tr
> Tracing pid 697 tid 100159 td 0xc249d600
> kdb_enter(c05d1267) at kdb_enter+0x2b
> panic(c05cf2f1,c249d600,0,c06030e0,ed63ba78) at panic+0x127
> lockmgr(c2545058,3001,c254507c,c249d600,ed63ba58) at lockmgr+0x3be
> vop_stdlock(ed63ba78,1001,c2545000,ed63ba94,c04f2d5c) at vop_stdlock 
> +0x1e
> VOP_LOCK_APV(c0608a40,ed63ba78) at VOP_LOCK_APV+0x87
> vn_lock(c2545000,1001,c249d600) at vn_lock+0xa8
> filt_vfsread(c2532000,6) at filt_vfsread+0x3a
> knote(c2870030,6,0) at knote+0x90
> VOP_WRITE_APV(c0608a40,ed63bc40) at VOP_WRITE_APV+0x18c
> vn_write(c23a4828,c286d380,c27a6c00,0,c249d600) at vn_write+0x1ea
> kern_writev(c249d600,a,c286d380,c286d380,0) at kern_writev+0x8e
> writev(c249d600,ed63bd04,3,1f,246) at writev+0x30
> syscall(bfbf003b,805003b,bfbf003b,80574d8,bfbfcf20) at syscall+0x22f
> Xint0x80_syscall() at Xint0x80_syscall+0x1f
> --- syscall (121, FreeBSD ELF32, writev), eip = 0x280c8d5b, esp =  
> 0xbfbfc83c, ebp = 0xbfbfcf48 ---
>
> This is HEAD from yesterday.

Can you try the patch at http://people.freebsd.org/~ssouhlal/testing/ 
knlist_locking-20050625.diff ?

--
Suleiman Souhlal     | ssouhlal_at_vt.edu
The FreeBSD Project  | ssouhlal_at_FreeBSD.org