Hi.
With recent source tree (at least the fix for cursor tracking bug, f3a4893b,
is in) and the patch in the following message applied:
http://leaf.dragonflybsd.org/mailarchive/bugs/2009-11/msg00020.html
the same panic reproduced after running pbulk for about a day:
(kgdb) bt
#0 _get_mycpu () at ./machine/thread.h:83
#1 dumpsys () at /usr/src/sys/kern/kern_shutdown.c:657
#2 0xc01a1069 in boot (howto=260) at /usr/src/sys/kern/kern_shutdown.c:378
#3 0xc01a132e in panic (fmt=0xc030099e "assertion: %s in %s")
at /usr/src/sys/kern/kern_shutdown.c:813
#4 0xc028297e in hammer_ip_delete_range (cursor=0xd81ffc9c, ip=0xe2dcc8d0,
ran_beg=0, ran_end=9223372036854775807, truncating=2)
at /usr/src/sys/vfs/hammer/hammer_object.c:1949
#5 0xc0279619 in hammer_sync_inode (trans=0xd81ea114, ip=0xe2dcc8d0)
at /usr/src/sys/vfs/hammer/hammer_inode.c:2715
#6 0xc0277525 in hammer_flusher_flush_inode (arg=0xd6d332d8)
at /usr/src/sys/vfs/hammer/hammer_flusher.c:507
#7 hammer_flusher_slave_thread (arg=0xd6d332d8)
at /usr/src/sys/vfs/hammer/hammer_flusher.c:451
#8 0xc01a9c0a in lwkt_deschedule_self (td=Cannot access memory at address 0x8
)
at /usr/src/sys/kern/lwkt_thread.c:271
Backtrace stopped: previous frame inner to this frame (corrupt stack?)
My pbulk directory and $WRKOBJDIR is nohistory-flagged, and periodic
`hammer cleanup' has been turned off (manually invoked about once a week
when the machine is idle). Less than 40% of 1Tbyte harddrive is occuppied,
according to the output from df command.
The vmcore is on my leaf account, as ~y0netan1/crash/{kernel,vmcore}.13.
BTW, sshd on leaf dumps core when I tried to login unless I turn off
tty allocation (via -T). Is anything bad happening on it?
Best regards.