Subject: Re: News server SCSI disk usage problems.
To: Duncan Ferguson <d.ferguson@worc.ac.uk>
From: Mark Brinicombe <amb@physig.ph.kcl.ac.uk>
List: port-arm32
Date: 08/01/1997 00:41:12
On Thu, 31 Jul 1997, Duncan Ferguson wrote:

> Also, i am getting a lot of errors in /var/log/messages:
> 
> Jul 31 11:09:57 news /netbsd: inode vp: type VBLK, usecount 247,
> writecount 0, refcount 27, Jul 31 11:09:57 news /netbsd:   tag 
> VT_UFS, ino 3950, on dev 16, 0
> 
> Jul 31 11:12:21 news /netbsd: ffs_update: bad
> indirect addr (0): inode=1324529 5731096/00577318 ip=f1533200 adr
> =f15332b4
> 
> The log file is turned overe _every_ hour because i get soo many of 
> these messages.  I am using an own-compiled kernel based on source as 
> of July 22nd with 2 patches - one for ffs_inode and the other to 
> reduce the network error messages from my etherh card to speed up the 
> system.

Hi,
  OK that log message above comes from the ffs_inode patch you have
applied. Basically it is indicating that the patch has detected a posibly
problem and fixed it. I would suggest looking at ffs_inode.c and the
patch. You will see that when it fixes the inode it prints this message.
You can just comment out the printf to remove this.

Interestingly this problem appears to have an MI problem (some sort of
race condition probably) as it appears to have been fixed in
NetBSD-current. (well I have not seen the problem now for a week or two).

CHeers,
				Mark