Subject: Re: Process hang on LFS
To: YAMAMOTO Takashi <yamt@mwd.biglobe.ne.jp>
From: Gary Duzan <gary@duzan.org>
List: current-users
Date: 01/28/2003 22:41:01
In Message <1043801752.788727.453.nullmailer@yamt.dyndns.org> ,
   YAMAMOTO Takashi <yamt@mwd.biglobe.ne.jp> wrote:

=>> =>hi,
=>> =>
=>> =>can you provide a result of 'ps -axl'?
=>> =>(or ps/w on ddb)
=>> 
=>>    The result is attached. It is now to the point where trying to
=>> write anything to the file system hangs, so I just did an echo to
=>> a file there, which shows up as process 27038.
=>> 
=>> 					Gary Duzan
=>
=>thanks.
=>
=>how many buffers does your kernel have?

===========================================================================
using 6144 buffers containing 127 MB of memory
===========================================================================
options         BUFCACHE=25
===========================================================================
There are   6144 metadata buffers using           130900 kBytes of memory.
There are  16231 pages for cached file data using  64924 kBytes of memory.
There are   2184 pages for executables using        8736 kBytes of memory.

File System          Bufs used   %   kB in use   %  Bufsize kB   %  Util %
/usr2                     1393  22       11137   8       29764  22      37
/                          140   2         886   0        3020   2      29
/usr                        71   1         279   0        1508   1      18
/usr/home                    2   0          16   0          44   0      36

Total:                    1606  26       12318   9       34336  26      35
===========================================================================
Filesystem  1K-blocks     Used     Avail Capacity  Mounted on
/dev/wd1a    70558768 12635106  50867786    19%    /usr2
===========================================================================
Master Superblock at 0x8:
    magic    0x70162     version  0x2         size     78177792  
    ssize    1880064     dsize    74267956    bsize    8192      
    fsize    1024        frag     8           minfree  10        
    inopb    8           ifpb     409         nindir   2048      
    nseg     42580       sepb     341         cleansz  1         
    segtabsz 125         segmask  0x0         segshift 0         
    bmask    0x1fff      bshift   13          ffmask   0x3FF     
    ffshift  10          fbmask   0x7         fbshift  3
    sushift  0           fsbtodb  1           cksum    0x7efb    
    nclean   23588       dmeta    999115      minfreeseg 2129      
    roll_id  0xaf89a85   interleave 0         sumsize  1024      
    maxfilesize  0xfffff00000000000
  Superblock disk addresses:
     0x8        0x7749d8   0xee93b0   0x165dd88  0x1dd2760  0x2547138 
     0x2cbbb10  0x34304e8  0x3ba4ec0  0x4319898 
  Checkpoint Info
    free     175188      idaddr   0xd23e5b    ifile    1         
    uinodes  0           bfree    62321076    avail    41352804  
    nfiles   381923      lastseg  0xd23294    nextseg  0xd240ec  
    curseg   0xd239c0    offset   0xd23e74    serial   768405    
    tstamp   Tue Jan 28 13:13:49 2003
===========================================================================

=>because buffer reservation is just a estimate currently,
=>it usually over-estimate and can deadlocks if you doesn't have
=>plenty of buffers.

   I would hope that would be enough.

					Gary Duzan