Re: panic: ufs_dirbad: bad dir

On Tue, 8 Mar 2005, Jeremie Le Hen wrote:

> Hi,
>
> with a UP -CURRENT from sunday, I stressed my system and I got the
> following panic.  I know Jeff has just commited something in vfs_bio.c
> but I'm not sure this is related, so in doubt, I report it before
> upgrading.  This is hand-written, so I didn't reported numeric value,
> but I have a crash dump.  Please contact me off-list if you need it.

Unfortunately, the panic you transcribed is a secondary panic.  I'd like
to see the primary stack, but it sounds like that isn't available.  I'm
going to commit some more major fixes in a few days which address unsafe
code at vnode teardown time.  Hopefully this may have some effect.

>
> /usr: bad dir ino 413567 at offset 0: mangled entry
> panic: ufs_dirbad: bad dir
> KDB: stack-backtrace:
> panic: bremfree: buffer 0xc9304af0 not on a queue
> KBD: enter: panic
> [ thread pid 45 tid 100062 ]
> Stopped at       kbd_enter+0x30: leave
> db> trace
> Tracing pid 45 tid 100062 td 0xc18fb320
> kbd_enter()
> panic()
> bremfree()
> getblk()
> breadn()
> bread()
> ffs_update()
> ufs_inactive()
> VOP_INACTIVE_APV()
> vput()
> handle_workitem_remove()
> process_worklist_item()
> softdep_process_worklist()
> sched_sync()
> fork_exit()
> fork_trampoline()
> --- trap0x1, eip = 0, esp = 0xd50d2d7c, ebp = 0 ---
>
> My dmesg(8) as well as my kernel configuration files are attached.
>
> Regards,
> --
> Jeremie Le Hen
> jeremie at le-hen dot org
>