From owner-freebsd-questions Tue Jan 11 4: 2:50 2000 Delivered-To: freebsd-questions@freebsd.org Received: from void.dynip.com (hobax2-059.dialup.optusnet.com.au [198.142.241.59]) by hub.freebsd.org (Postfix) with ESMTP id 7CFA215533 for ; Tue, 11 Jan 2000 04:02:38 -0800 (PST) (envelope-from death@southcom.com.au) Received: from windows (windows.void.net [192.168.1.2]) by void.dynip.com (Postfix) with ESMTP id 3D2AF55BA for ; Tue, 11 Jan 2000 23:02:30 +1100 (EST) Message-Id: <4.2.2.20000111213124.00a2c3d0@mail.southcom.com.au> X-Sender: death@mail.southcom.com.au X-Mailer: QUALCOMM Windows Eudora Pro Version 4.2.2 Date: Tue, 11 Jan 2000 23:02:28 +1100 To: freebsd-questions@FreeBSD.ORG From: james Subject: Telnet freezes with wait chan of 'ttywri'/'ttywai' Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Hi, I've just had another freeze on a telnet session from my Windows machine to my FreeBSD 4.0-CURRENT machine with a wait chan of 'ttywri'. I decided to actually investigate a little bit as somebody posted about the same problem on freebsd-stable just a few days ago. I was just doing a ping, and for no apparant reason it just froze. All that's left in telnet is: 64 bytes from 198.142.241.254: icmp_seq=2305 ttl=255 time=124.183 ms 64 bytes from 198.142.241.254: icmp_seq=2306 ttl=255 time=106.638 ms 64 bytes from 198.142.241.254: icmp_seq=2307 ttl=255 time=103.817 ms I waited for over and hour and still nothing. 'ps' shows (i formatted it a bit better so it's neater). Sorry if i missed some important columns, i didn't really know what to include: USER PID %CPU %MEM VSZ RSS TT STAT STARTED TIME UID PID PPID CPU PRI NI WCHAN COMMAND root 4259 0.0 0.8 428 240 p2 I+ 6:46PM 0:02.33 0 4259 4257 0 4 0 ttywri ping hobax2.optusnet.com.au Upon trying "echo "hmm" | write death ttyp2" from another telnet session, the echo hangs. tcpdump has been showing this over and over again since the ping froze (192.168.1.1 is freebsd, 192.168.1.2 is windows): 21:10:22.451388 192.168.1.1.23 > 192.168.1.2.1796: . 0:1(1) ack 10 win 17520 (DF) [tos 0x10] (ttl 64, id 5329) 21:10:22.451548 192.168.1.2.1796 > 192.168.1.1.23: . ack 0 win 0 (DF) (ttl 128, id 64882) 21:11:09.801469 192.168.1.1.23 > 192.168.1.2.1796: . 0:1(1) ack 10 win 17520 (DF) [tos 0x10] (ttl 64, id 5332) 21:11:09.801624 192.168.1.2.1796 > 192.168.1.1.23: . ack 0 win 0 (DF) (ttl 128, id 627) 21:11:57.151777 192.168.1.1.23 > 192.168.1.2.1796: . 0:1(1) ack 10 win 17520 (DF) [tos 0x10] (ttl 64, id 5335) 21:11:57.151943 192.168.1.2.1796 > 192.168.1.1.23: . ack 0 win 0 (DF) (ttl 128, id 1907) although there was a few small differences such as: 20:42:27.883546 192.168.1.2.1796 > 192.168.1.1.23: P 1:10(9) ack 0 win 0 (DF) (ttl 128, id 2413) 20:42:27.883737 192.168.1.1.23 > 192.168.1.2.1796: . ack 10 win 17511 (DF) [tos 0x10] (ttl 64, id 5200) 20:42:45.010565 192.168.1.1.23 > 192.168.1.2.1796: . 0:1(1) ack 10 win 17520 (DF) [tos 0x10] (ttl 64, id 5203) and: 21:26:09.483757 192.168.1.2.1796 > 192.168.1.1.23: . ack 0 win 0 (DF) (ttl 128, id 24435) 21:26:36.412563 192.168.1.2.1796 > 192.168.1.1.23: P 10:19(9) ack 0 win 0 (DF) (ttl 128, id 26227) 21:26:36.412739 192.168.1.1.23 > 192.168.1.2.1796: . ack 19 win 17511 (DF) [tos 0x10] (ttl 64, id 5400) 21:26:56.836448 192.168.1.1.23 > 192.168.1.2.1796: . 0:1(1) ack 19 win 17520 (DF) [tos 0x10] (ttl 64, id 5423) and this happens at the start of each tcpdump: 22:08:46.570425 192.168.1.1.23 > 192.168.1.2.1796: . 2330942662:2330942663(1) ack 60560415 win 17520 (DF) [tos 0x10] (ttl 64, id 5989) 22:08:46.570594 192.168.1.2.1796 > 192.168.1.1.23: . ack 0 win 0 (DF) (ttl 128, id 50550) I can't quite remember if pressing ctrl-c in the frozen telnet unhung the echo, but it killed the ping, and now i'm left with: USER PID %CPU %MEM VSZ RSS TT STAT STARTED TIME UID PID PPID CPU PRI NI WCHAN COMMAND root 4257 0.0 3.1 1220 932 p2 I+ 6:46PM 0:00.17 0 4257 4255 0 4 0 ttywai _su -m (tcsh) When i type into the telnet (nothing is echoed btw), i get the following incremental pattern in tcpdump: 22:48:50.234476 192.168.1.1.23 > 192.168.1.2.1796: . ack 1 win 17519 (DF) [tos 0x10] (ttl 64, id 6155) 22:48:50.450493 192.168.1.2.1796 > 192.168.1.1.23: P 1:2(1) ack 1 win 0 (DF) (ttl 128, id 32388) 22:48:50.450661 192.168.1.1.23 > 192.168.1.2.1796: . ack 2 win 17519 (DF) [tos 0x10] (ttl 64, id 6156) 22:48:50.450860 192.168.1.2.1796 > 192.168.1.1.23: P 2:3(1) ack 1 win 0 (DF) (ttl 128, id 32644) 22:48:50.451005 192.168.1.1.23 > 192.168.1.2.1796: . ack 3 win 17519 (DF) [tos 0x10] (ttl 64, id 6157) 22:48:50.451393 192.168.1.2.1796 > 192.168.1.1.23: P 3:4(1) ack 1 win 0 (DF) (ttl 128, id 32900) 22:48:50.451535 192.168.1.1.23 > 192.168.1.2.1796: . ack 4 win 17519 (DF) [tos 0x10] (ttl 64, id 6158) 22:48:50.512747 192.168.1.2.1796 > 192.168.1.1.23: P 4:5(1) ack 1 win 0 (DF) (ttl 128, id 33156) 22:48:50.512921 192.168.1.1.23 > 192.168.1.2.1796: . ack 5 win 17519 (DF) [tos 0x10] (ttl 64, id 6159) Any ideas what could be wrong? I can send any other output or config if you like.. I just don't want to include some massive email if the problem is really trivial or whatever.. Cheers To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message