netstat info weird

Max Laier max at love2party.net
Mon Feb 14 08:53:58 PST 2005


On Monday 14 February 2005 17:02, Varshavchick Alexander wrote:
> Hi all,
>
> I guess something is wrong here:
> > netstat -m
>
> 4294961156 mbufs in use
> ^^^^^^^^^^
> 4294967102/50176 mbuf clusters in use (current/max)
> ^^^^^^^^^^
> 0/3/12800 sfbufs in use (current/peak/max)
> 4192381 KBytes allocated to network
> ^^^^^^^
> 0 requests for sfbufs denied
> 0 requests for sfbufs delayed
> 0 requests for I/O initiated by sendfile
> 43 calls to protocol drain routines
>
>
> It is 5.3-RELEASE, 2-cpu server Xeon box with 4G of RAM.
>
> As the other evidence that something is wrong with the server, the
> system halts at random times (the only way is to press reboot button).
> The are no messages in the server logs to the subject.
>
> Any hints are welcome. Thanks in advance.

This is a statistic problem.  "$vmstat -z | grep Mbuf" gives accurate numbers.  
It is possible to have correct stats in netstat -m output, but at the cost of 
an additional atomic operation for every allocation.  As these are very 
expensive (esp. on P4) it was decided to not care that much about this 
special statistics.

If the command above gives normal numbers (third column) the cause must be 
somewhere else.  You can try to build a debugging kernel and see if that 
turns anything up.

-- 
/"\  Best regards,                      | mlaier at freebsd.org
\ /  Max Laier                          | ICQ #67774661
 X   http://pf4freebsd.love2party.net/  | mlaier at EFnet
/ \  ASCII Ribbon Campaign              | Against HTML Mail and News
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/freebsd-hackers/attachments/20050214/70c39beb/attachment.bin


More information about the freebsd-hackers mailing list