From owner-freebsd-questions Fri Mar 6 18:22:02 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id SAA05658 for freebsd-questions-outgoing; Fri, 6 Mar 1998 18:22:02 -0800 (PST) (envelope-from owner-freebsd-questions@FreeBSD.ORG) Received: from shell6.ba.best.com (jkb@shell6.ba.best.com [206.184.139.137]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id SAA05603 for ; Fri, 6 Mar 1998 18:21:53 -0800 (PST) (envelope-from jkb@best.com) Received: from localhost (jkb@localhost) by shell6.ba.best.com (8.8.8/8.8.BEST) with SMTP id SAA11877 for ; Fri, 6 Mar 1998 18:21:49 -0800 (PST) X-Authentication-Warning: shell6.ba.best.com: jkb owned process doing -bs Date: Fri, 6 Mar 1998 18:21:49 -0800 (PST) From: Jan Koum X-Sender: jkb@shell6.ba.best.com Reply-To: Jan Koum To: questions@FreeBSD.ORG Subject: odd netstat statistic Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Hello all, I am getting odd netstat output. The machine is somewhat busy web server (it only does httpd, about 100000 hits per day) running apache 1.2.5 and 2.2.5-RELEASE with custom kernel which has: options NMBCLUSTERS=8192 #mbuf clusters at 8192 options CHILD_MAX=1024 #max number of child processes options OPEN_MAX=1024 Here is what do I mean by weird netstat outputs: % netstat -a Active Internet connections (including servers) Proto Recv-Q Send-Q Local Address Foreign Address (state) tcp 0 0 www.http ww-tm03.proxy.ao.41143 TIME_WAIT netstat: kvm_read: Bad address ??? udp 0 0 *.1304 *.* udp 0 0 *.1303 *.* udp 0 0 *.1302 *.* [snip] f45e8c00 dgram 0 0 0 f310a014 0 f310a794 f4552800 dgram 0 0 f4551480 0 f3179714 0 /var/run/log The "???" was in the output, I didn't stick it in there. :) Doing netstat -a again a second later worked just fine, but this time I got: (notice the negative numbers) tcp 0 0 www.http 207.235.168.3.10608 TIME_WAIT tcp 0 0 www.http 207.235.168.3.10541 TIME_WAIT tcp 320 -266393784 www.http 16.112.170.244.10526 -265120128 tcp 312 -266363128 www.http 144.94.170.244.10541 -266234176 tcp 239 -265502528 *.44020 144.134.170.244.37039 -265316220 tcp 2384 -266516456 *.44020 16.180.171.244.36893 CLOSED* tcp 28 -264778220 *.43764 16.74.169.244.39013 CLOSED* tcp 679 -266061684 *.44020 144.254.168.244.38924 -265958576 tcp 344 -266462408 *.43764 16.102.169.244.38938 CLOSED tcp 56 -264760852 *.43508 16.75.171.244.6372 CLOSED* tcp 0 -266389720 *.43764 144.156.170.244.39081 -266065488 [snip] However, netstat -m seem to be fine: % netstat -m 227 mbufs in use: 114 mbufs allocated to data 107 mbufs allocated to packet headers 5 mbufs allocated to protocol control blocks 1 mbufs allocated to socket names and addresses 111/408 mbuf clusters in use 844 Kbytes allocated to network (29% in use) 0 requests for memory denied 0 requests for memory delayed 0 calls to protocol drain routines % And almost every time netstat works right also: % netstat -a | wc -l 104 % But netstat -s is also doing negative numbers thingie *grin* : % netstat -s [snip] tcp: 4821743 packets sent 3459033 data packets (-967744068 bytes) 263312 data packets (296817086 bytes) retransmitted 192 resends initiated by MTU discovery 847250 ack-only packets (102774 delayed) 0 URG only packets 2790 window probe packets 1185 window update packets 248173 control packets 4760259 packets received 2726009 acks (for -1008981097 bytes) 612410 duplicate acks [snip] % Any suggestions? I assume this is of course not a good thing. Even though it only happened today for the first time in a month. :) I can upgrade to 2.2-STABLE or wait for 2.2.6, but is there any other way to fix it? And would upgrade help? -- Yan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message