strange netstat -m output

Oliver Fromme check+ip0ml300rslkuxa6 at
Thu Oct 27 04:14:03 PDT 2005

Stefan Krueger wrote:
 > $ netstat -m
 > 193/1/13312 mbufs in use (current/peak/max):

I wonder how the peak can be less than the current value.
Normally, it should be current <= peak <= max.

 > 193 mbufs allocated to data
 > 4292490067/4292490067/6656 mbuf clusters in use (current/peak/max)
 > ^^^^^^^^^^^^^^^^^^^^^

Those numbers are definitely bogus.  An mbuf cluster is
about 2 Kbyte in size, so 4292490067 of them would require
about 8 Tbyte of memory.  I assume your machine doesn't
have that much memory.

 > 3434150 Kbytes allocated to network (220% of mb_map in use)

That's 3.3 Gbyte.  Also an impressive number.

 > could someone tell me why they're so high and if netstat -m's output is
 > correct

It's not correct.  Maybe there's an API problem -- Are you
sure that your userland is in sync with the kernel?

(I seem to remember that some things have been converted
from KVM to sysctl recently, but I think only vmstat was
affected, not netstat.  I'm not 100% sure, though.)

Best regards

Oliver Fromme,  secnetix GmbH & Co. KG, Marktplatz 29, 85567 Grafing
Dienstleistungen mit Schwerpunkt FreeBSD:

Any opinions expressed in this message may be personal to the author
and may not necessarily reflect the opinions of secnetix in any way.

More information about the Bugs mailing list