Date: Tue, 25 Nov 1997 19:48:18 -0800 (PST) From: Jaye Mathisen <mrcpu@cdsnet.net> To: hackers@freebsd.org Subject: Maybe I'm missing something here. Message-ID: <Pine.NEB.3.95.971125193803.5824D-100000@mail.cdsnet.net>
next in thread | raw e-mail | index | archive | help
P6, 512MB RAM, 2.2.5-stable as of Monday. There's something like 1100 processes (web servers running on this), doing nothing. (no clients). maxusers is set to 128, and kern.maxfiles is set to 16384. Here's some vmstat output for a few seconds. 0 0 0198624124324 1 0 0 0 0 0 0 0 0 0 242 1610 525 0 11 89 0 0 0198624124324 1 0 0 0 0 0 0 0 0 0 242 1571 512 1 8 91 0 0 0198624124324 1 0 0 0 0 0 6 0 0 0 251 1838 602 2 15 83 0 0 0198624124324 1 0 0 0 0 0 0 0 0 0 240 1571 512 3 10 87 0 0 0198624124324 1 0 0 0 0 0 0 0 0 0 245 1571 511 1 10 89 0 0 0198624124324 1 0 0 0 0 0 0 0 0 0 242 1664 543 1 8 91 0 0 0198624124324 1 0 0 0 0 0 0 0 0 0 242 1574 514 0 13 87 0 0 0198624124324 1 0 0 0 0 0 0 0 0 0 241 1571 512 1 10 89 If nothing is going on, where is the system CPU time going? No disk activity, , no paging/swapping, no access, except my over-the-network xterm and vmstat -s. top shows nothing. everthing except top is sitting in select or lockf for as far down as I looked. lastpid hasn't changed for the 15 seconds or so when I checked. Top shows 6% of my time in interrupt, 3.8% in system. netstat -i shows nothing except what looks to be the correct amount of traffic for my xterm output. Do these numbers look right? Looking at a 2.2.2 server doing almost the same thing, things just seem lower all-the-way around, but of course, w/o identical loads, it doesn't mean anything.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.NEB.3.95.971125193803.5824D-100000>