Return to BSD News archive
Xref: sserve comp.unix.internals:7191 comp.unix.wizards:32322 comp.unix.bsd:13963 Path: sserve!newshost.anu.edu.au!harbinger.cc.monash.edu.au!msuinfo!netnews.upenn.edu!newsserver.jvnc.net!news.cac.psu.edu!news.pop.psu.edu!news.pop.psu.edu!not-for-mail From: craigr@netcom.com (Craig Robertson) Newsgroups: comp.unix.internals,comp.unix.wizards,comp.unix.bsd Subject: SunOS 4.1.1 debuggee hangs at PRI=-5 during ptrace attach Date: 17 May 1994 23:15:54 -0400 Organization: NETCOM On-line Communication Services (408 261-4700 guest) Lines: 28 Sender: barr@pop.psu.edu Approved: barr@pop.psu.edu Distribution: na Message-ID: <craigrCpz4rG.FJC@netcom.com> NNTP-Posting-Host: bosnia.pop.psu.edu Greetings. I'm occasionally having a problem attaching dbx to a reasonably large process that has been stopped (via SIGSTOP) and swapped out. Occasionally, dbx will not be able to attach to this process (the ptrace attach call will hang) and "ps -axlww" will show the process at PRI=-5 (from a previous? biowait) with a WCHAN of 0... I suspect a race in the special-condition of "if the debuggee is already stopped and you're attempting an attach" in ptrace(). I'm curious if this is Sun-specific code or if it exists in regular BSD unix... Machines on which I see this are: Both sun4c and sun4 architectures Have at least 32MB of physical memory Have *plenty* of swap space I am a SunOS source licensee but this stuff seems *really* tricky... I have a service call logged with Sun, but time is really of the essense on this issue. Any pointers or help would be appreciated. Thanks in advance, Craig Robertson craigr@netcom.com -- Craig Robertson craigr@netcom.com