Re: occational system lock up in 6.0_BETA and 6.99.5

Hi, all

>> As D'Arcy I had a lock up on i386 too.
>
> NetBSD/amd64 is hanging itself up during heavy loads, too - there are
> several reports on this currently being discussed on port-amd64.  In my
> case, at least, it's obviously related to heavy disk I/O,

I can make system lockup easily.  With sandpoint NAS, running snapshot
dump causes an immediate trouble.

nas8245# df -i /store
Filesystem    1K-blocks       Used      Avail %Cap    iUsed   iAvail %iCap Mounted on
/dev/wd0d      73223688    3986786   65575718   5%   311375  8856047   3% /store

nas8245# /sbin/dump -X -0f - /store >/dev/null
  DUMP: Found /dev/rwd0d on /store in /etc/fstab ... got lockup in a few second ...
~# ... DDB break in works ...

db> bt
0x0060bd90: at comintr+0x590
0x0060bde0: at pic_handle_intr+0x198
0x0060be20: at trapstart+0x684
0x0060bef0: at sched_curcpu_runnable_p+0x2c
0x0060bf00: at idle_loop+0xe8
0x0060bf20: at setfunc_trampoline+0x8
saved LR(0x7ffffd) is invalid.
db> ... not a useful backtrace ... 

I was running telnet session in other window.  The command line shell
worked *after* the lockup happened, but running a new command
stucks right at the point.

db> ps
PID    LID S CPU     FLAGS       STRUCT LWP *               NAME WAIT
650      1 3   0         0            3c33be0               dump needbuf
162      1 3   0         0            3d90900                ksh needbuf
638      1 3   0        80            3179980              login wait
636      1 3   0        80            3d90380            telnetd select
38       1 3   0        80            32be120                ksh pause

Note that dump(8) could not reach "parallel dump session" phase.  the /store
filesys is WAPBL enabled.  It seems a scheduler issue.  The last operation
I did is invoking "ps xa" and DDB shows no ps process.
 
Toru Nishimura / ALKYL Technology