A Debian has been crashed and I should found how it happened.
Memory, Swap and Harddisk wasn't full, but the Server was unreachable, no ssh, ftp, http, ping...
After a hardware reset I have analysed the log files.
The crash was 22:45 offline and I want to analyse log entries between 22:40 and 22:50 and greped them
cat syslog.2 |grep '22:4'
...
Apr 2 22:41:12 USER named[3493]: isc_socket_create: fcntl/reserved: Too many open files
Apr 2 22:41:12 USER named[3493]: isc_socket_create: fcntl/reserved: Too many open files
Apr 2 22:41:12 USER named[3493]: isc_socket_create: fcntl/reserved: Too many open files
Apr 2 22:41:12 USER named[3493]: isc_socket_create: fcntl/reserved: Too many open files
...
and seen a lot of the time the same error. The server has tried to open files, but couldn't and crashed.
number of the max open files was 1024
USER:/var/log# ulimit -a
core file size (blocks, -c) 0
data seg size (kbytes, -d) unlimited
scheduling priority (-e) 0
file size (blocks, -f) unlimited
pending signals (-i) 73728
max locked memory (kbytes, -l) 32
max memory size (kbytes, -m) unlimited
open files (-n) 1024
pipe size (512 bytes, -p) 8
POSIX message queues (bytes, -q) 819200
real-time priority (-r) 0
stack size (kbytes, -s) 8192
cpu time (seconds, -t) unlimited
max user processes (-u) 73728
virtual memory (kbytes, -v) unlimited
file locks (-x) unlimited
I've changed the value of max open files to 8192 with
ulimit -n 8192
and it works very well