виснет nginx-невозможно сделать рестарт

[maxim]

uname -a
CentOS Linux srv-4-0  2.6.18-194.8.1.el5 #1 SMP Thu Jul 1 19:04:48 EDT  
2010 x86_64 x86_64 x86_64 GNU/Linux

Не рестартует и не останавливается nginx
  из под рута kill и kill -9 по pid-у не убивает процессы(первый раз с  
таким встречаюсь)

в messages

Jul 27 05:25:22 srv-4-0 kernel: INFO: task nginx:3706 blocked for more  
than 120 seconds.
Jul 27 05:25:22 srv-4-0 kernel: "echo 0 > /proc/sys/kernel/ 
hung_task_timeout_secs" disables this message.
Jul 27 05:25:22 srv-4-0 kernel: nginx         D ffff81000103f1a0      
0  3706   3704          3707  3705 (NOTLB)
Jul 27 05:25:22 srv-4-0 kernel:  ffff81032f6cfc78 0000000000000086  
ffffffffffffffff ffff81032f6cfc58
Jul 27 05:25:22 srv-4-0 kernel:  0000000000000001 000000000000000a  
ffff8103304cb7a0 ffff81033fc38100
Jul 27 05:25:22 srv-4-0 kernel:  000023b513c7cd17 00000000000013ee  
ffff8103304cb988 000000078844191d
Jul 27 05:25:22 srv-4-0 kernel: Call Trace:
Jul 27 05:25:22 srv-4-0 kernel:  [<ffffffff80063c6f>]  
__mutex_lock_slowpath+0x60/0x9b
Jul 27 05:25:22 srv-4-0 kernel:  [<ffffffff80063cb9>] .text.lock.mutex 
+0xf/0x14
Jul 27 05:25:22 srv-4-0 kernel:  [<ffffffff8000cef3>] do_lookup 
+0x90/0x1e6
Jul 27 05:25:22 srv-4-0 kernel:  [<ffffffff80009bb2>] __link_path_walk 
+0x3a6/0xf5b
Jul 27 05:25:22 srv-4-0 kernel:  [<ffffffff8000e9bc>] link_path_walk 
+0x42/0xb2
Jul 27 05:25:22 srv-4-0 kernel:  [<ffffffff8000cce3>] do_path_lookup 
+0x275/0x2f1
Jul 27 05:25:22 srv-4-0 kernel:  [<ffffffff800127e8>] getname+0x15b/ 
0x1c2
Jul 27 05:25:22 srv-4-0 kernel:  [<ffffffff8002387c>] __user_walk_fd 
+0x37/0x4c
Jul 27 05:25:22 srv-4-0 kernel:  [<ffffffff800287b0>] vfs_stat_fd+0x1b/ 
0x4a
Jul 27 05:25:22 srv-4-0 kernel:  [<ffffffff8002b88e>] sys_recvfrom 
+0xd4/0x130
Jul 27 05:25:22 srv-4-0 kernel:  [<ffffffff800f87ae>] sys_epoll_wait 
+0x3b8/0x3f9
Jul 27 05:25:22 srv-4-0 kernel:  [<ffffffff800235ae>] sys_newstat 
+0x19/0x31
Jul 27 05:25:22 srv-4-0 kernel:  [<ffffffff8005d229>] tracesys+0x71/0xe0
Jul 27 05:25:22 srv-4-0 kernel:  [<ffffffff8005d28d>] tracesys+0xd5/0xe0

Свободное место на HDD есть, памяти 12 гиг, LA  меньше единицы в  
момент инцидента, сервер в GlsuterFS

начало nginx.conf

worker_processes  8;
user 212adm 212adm;

events {
     worker_connections  2048;
     use epoll;
}

http {
     passenger_root /opt/ree/lib/ruby/gems/1.8/gems/passenger-2.2.11;
     passenger_ruby /opt/ree/bin/ruby;

     passenger_max_pool_size 35;
     passenger_pool_idle_time 0;
     rails_spawn_method conservative;

     include       mime.types;
     default_type  application/octet-stream;

     client_max_body_size       100m;
     keepalive_timeout  65;
     gzip on;
     gzip_min_length 1000;
     gzip_proxied any;
     gzip_disable "msie6";
     gzip_types text/plain text/html text/xml application/xml  
application/x-javascript text/javascript text/css text/json;

     server {
         listen 80;
         server_name default;
.......
.......
----------- следущая часть -----------
Вложение в формате HTML было извлечено&hellip;
URL: <http://nginx.org/pipermail/nginx-ru/attachments/20100727/d0ac9691/attachment-0001.html>