From owner-freebsd-current Thu Oct 1 22:03:20 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id WAA29190 for freebsd-current-outgoing; Thu, 1 Oct 1998 22:03:20 -0700 (PDT) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from tor-dev1.nbc.netcom.ca (tor-dev1.nbc.netcom.ca [207.181.89.12]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id WAA29184 for ; Thu, 1 Oct 1998 22:03:19 -0700 (PDT) (envelope-from taob@tor-dev1.nbc.netcom.ca) Received: (from taob@localhost) by tor-dev1.nbc.netcom.ca (8.8.8/8.8.8) id BAA26635; Fri, 2 Oct 1998 01:02:46 -0400 (EDT) Date: Fri, 2 Oct 1998 01:02:46 -0400 (EDT) From: Brian Tao X-Sender: taob@tor-dev1.nbc.netcom.ca To: FREEBSD-CURRENT Subject: Processes stuck waiting for tty drain? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Installed a 3.0-BETA on a K6-2/300, supped at around 22:00 EDT on September 30. No devfs, ELF system. I used "pax -rwvpe" to duplicate /usr/{obj,src,sup} from /dev/wd0s2h to /dev/da0s1a (both with softupdates enabled). At some point during the copy, activity on that tty stopped. I telnetted back into the machine and found the pax process in SIGSTOP. Sending it a SIGCONT changed its state, but it didn't seem to be making any further progress: % ps lax UID PID PPID CPU PRI NI VSZ RSS WCHAN STAT TT TIME COMMAND 1000 338 337 0 18 0 452 264 pause Is p1 0:00.04 -csh (csh) 0 353 338 0 4 0 1312 1108 ttywai S+ p1 0:00.15 -usr/local/bin/tcsh 0 407 353 0 4 0 1112 980 ttywri S p1 0:09.00 pax -rwvpe obj ports src sup /depot Killing off pax and tcsh left just the login csh, which was unkillable: UID PID PPID CPU PRI NI VSZ RSS WCHAN STAT TT TIME COMMAND 1000 338 337 0 4 0 452 288 ttywai SEs+ p1- 0:00.00 (csh) Killing off the parent telnetd did allow the csh to exit. Last time I saw this sort of thing (probably three or more years ago now), the process was somehow blocked while waiting for its tty to drain before exiting. This time, I was ssh'd into a Solaris box from home over 64K ISDN, then telnetting into the -current box, if that makes a difference. Any ideas? I'll try reproducing the problem. [...] wdc0 at 0x1f0-0x1f7 irq 14 on isa wdc0: unit 0 (wd0): wd0: 8693MB (17803440 sectors), 17662 cyls, 16 heads, 63 S/T, 512 B/S wd0: ATA INQUIRE valid = 0007, dmamword = 0007, apio = 0003, udma = 0407 [...] ncr0: restart (scsi reset). de0: enabling 100baseTX port (probe6:ncr0:0:6:0): INQUIRY. CDB: 12 1 80 0 ff 0 (probe6:ncr0:0:6:0): ILLEGAL REQUEST asc:24,0 (probe6:ncr0:0:6:0): Invalid field in CDB sks:c8,1 pass0 at ncr0 bus 0 target 3 lun 0 pass0: Fixed Direct Access SCSI2 device pass0: Serial Number LN496865000019070BR8 pass0: 20.0MB/s transfers (20.0MHz, offset 15), Tagged Queueing Enabled pass1 at ncr0 bus 0 target 6 lun 0 pass1: Removable CD-ROM SCSI2 device pass1: 10.0MB/s transfers (10.0MHz, offset 16) da0 at ncr0 bus 0 target 3 lun 0 da0: Fixed Direct Access SCSI2 device da0: Serial Number LN496865000019070BR8 da0: 20.0MB/s transfers (20.0MHz, offset 15), Tagged Queueing Enabled da0: 4340MB (8888924 512 byte sectors: 255H 63S/T 553C) -- Brian Tao (BT300, taob@risc.org) "Though this be madness, yet there is method in't" To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message