> Are you sure you aren't getting hit with system overhead? Some of the long
> times you give look reasonable for something being paged back in from
> disk.
I thought about system overhead, but the tests are all being run in the
same script, there is plenty of memory free, and I cant think of what sort
of overhead it could be. The very first time I run it in a while there is
some lag (possibly to a DNS lookup or a hostname entry), but that gets
worked out after one iteration of one test. Besides, when I run it with
an IP address, the delay persists. I would think 300 millisecond (300,000
microseconds) delay is awfully large for pulling data from a SCSI-2 disk.
Also, I would think the overhead would show up in the v2c polls. Perhaps
the v1 and v2u code written by a different person than the v2c stuff and
there is a different routine to establishing a connection.
-Steve
-- !! This message is brought to you via the `tkined & scotty' mailing list. !! Please do not reply to this message to unsubscribe. To subscribe or !! unsubscribe, send a mail message to <tkined-request@ibr.cs.tu-bs.de>. !! See http://wwwsnmp.cs.utwente.nl/~schoenw/scotty/ for more information.