Re: Strange behaviour

From: Ken Smith <kensmith_at_cse.Buffalo.EDU>
Date: Thu, 22 Jan 2004 00:10:17 -0500
On Wed, Jan 21, 2004 at 10:43:07AM -0800, Kris Kennaway wrote:

> OK, sounds like it might be something to do with vnode locking.
> Turning on DEBUG_VFS_LOCKS might help to find it.

I turned that on but it doesn't seem to have given any more clues.
The machine is (recently...) attached to a serial console, no extra
messages.  At the time I did it the only net traffic would have
been my login session (watching with top) so odds are it's just
the disk I/O causing it.

Amongst other things I've got:

# Debugging for use in -current
options         DDB                     #Enable the kernel debugger
options         INVARIANTS              #Enable calls of extra sanity checking
options         INVARIANT_SUPPORT       #Extra sanity checks of internal structures, required by INVARIANTS
options         WITNESS                 #Enable checks to detect deadlocks and cycles
#options        WITNESS_SKIPSPIN        #Don't run witness on spinlocks for speed
options         DEBUG_VFS_LOCKS

Otherwise it's a more or less GENERIC config file.  As an earlier
attempt I did try turning off acpi with this in /boot/loader.conf:

hint.acpi.0.disabled="1"

The machine is up and running right now with several wedged find
processes:

perseus 13 % ps alxww | grep find
    0  1425     1  69  -4  0  1316  836 getblk D     p0-   0:13.09 find / -name
foo -print
    0  1426     1 115  -4  0  1316  836 ufs    D     p0-   0:11.74 find / -name
foo -print
    0  1427     1 105  -4  0  1316  836 ufs    D     p0-   0:11.77 find / -name
foo -print
    0  1428     1 112  -4  0  1316  836 getblk D     p0-   0:11.54 find / -name
foo -print
    0  1429     1  89  -4  0  1316  836 getblk D     p0-   0:11.22 find / -name
foo -print
perseus 14 %

Given the DDB config info above is there anything I can do to give you
more information?

-- 
						Ken Smith
- From there to here, from here to      |       kensmith_at_cse.buffalo.edu
  there, funny things are everywhere.   |
                      - Theodore Geisel |
Received on Wed Jan 21 2004 - 20:10:22 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:37:39 UTC