nginx & FreeBSD 8.2 keglim

[madman]

Здравствуйте!

На FreeBSD 8.2-RELEASE-p2 amd64 nginx стал часто
находится в состоянии keglim и тормозить
при отдаче.

top:
last pid: 93140;  load averages:  7.65,  8.00,  3.90                    
                                                    up 0+20:21:25 
12:31:39
800 processes: 1 running, 799 sleeping
CPU:  0.0% user,  0.0% nice,  1.1% system,  0.0% interrupt, 98.9% idle
Mem: 4008M Active, 462M Inact, 3080M Wired, 172M Cache, 726M Buf, 146M
Free
Swap: 4000M Total, 268M Used, 3732M Free, 6% Inuse, 8K In

  PID USERNAME       THR PRI NICE   SIZE    RES STATE   C   TIME   WCPU
COMMAND
73145 www              1  44  -10 17816K  4220K keglim  2   0:19  0.20%
nginx
73103 www              1  44  -10 19864K  5836K keglim  7   0:25  0.10%
nginx

systat -v 1:
    1 users    Load  3.97  7.01  3.73                  Mar 28 12:32

Mem:KB    REAL            VIRTUAL                       VN PAGER   SWAP
PAGER
        Tot   Share      Tot    Share    Free           in   out     in 
 out
Act 1016156   12792 10225132    27476  356892  count    15            3
All 1631216   15172 1084906k   120172          pages    15           23
Proc:                                                           
Interrupts
  r   p   d   s   w   Csw  Trp  Sys  Int  Sof  Flt     91 cow   18041
total
             1k       16k  511 6207 2048  803  445    228 zfod       
ehci0 19
                                                          ozfod      
atapci1 23
 2.7%Sys   0.4%Intr  0.4%User  0.0%Nice 96.5%Idle        %ozfod    28
mfi0 irq33
|    |    |    |    |    |    |    |    |    |    |       daefr   138
atapci0 38
=+                                                        prcfr  2000
cpu0: time
                                         7 dtbuf      759 totfr  1882
bce0 256
Namei     Name-cache   Dir-cache   3000000 desvn          react  2000
cpu1: time
   Calls    hits   %    hits   %    101337 numvn          pdwak  2000
cpu3: time
    7336    7253  99                 51609 frevn          pdpgs  1994
cpu4: time
                                                        3 intrn  2000
cpu6: time
Disks   ad4   ad6 mfid0 mfid1                     3300556 wire   2000
cpu5: time
KB/t   9.89 12.53 32.00   101                     3960836 act    2000
cpu7: time
tps      79    60     3    15                      439212 inact  1999
cpu2: time
MB/s   0.76  0.74  0.09  1.51                      203748 cache
%busy     2     1     2     8                      153172 free
                                                   743504 buf

netstat -m:
126728/3712/130440 mbufs in use (current/cache/total)
9769/2961/12730/524288 mbuf clusters in use (current/cache/total/max)
9769/1368 mbuf+clusters out of packet secondary zone in use
(current/cache)
102400/0/102400/102400 4k (page size) jumbo clusters in use
(current/cache/total/max)
0/0/0/6400 9k jumbo clusters in use (current/cache/total/max)
0/0/0/3200 16k jumbo clusters in use (current/cache/total/max)
460820K/6850K/467670K bytes allocated to network (current/cache/total)
0/0/0 requests for mbufs denied (mbufs/clusters/mbuf+clusters)
0/0/0 requests for jumbo clusters denied (4k/9k/16k)
0/0/0 sfbufs in use (current/peak/max)
0 requests for sfbufs denied
0 requests for sfbufs delayed
3907967 requests for I/O initiated by sendfile
0 calls to protocol drain routines

Что можно сделать в этой ситуации? 
Спасибо

Posted at Nginx Forum: http://forum.nginx.org/read.php?21,224542,224542#msg-224542