Xref: utzoo comp.unix.questions:31115 comp.unix.aux:4652
Newsgroups: comp.unix.questions,comp.unix.aux
Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!think.com!barmar
From: barmar@think.com (Barry Margolin)
Subject: Re: record locking and portmapper error
Message-ID: <1991May8.054848.23491@Think.COM>
Sender: news@Think.COM
Reply-To: barmar@think.com
Organization: Thinking Machines Corporation, Cambridge MA, USA
References: <1991May7.234142.15065@ddtg.com>
Distribution: usa
Date: Wed, 8 May 91 05:48:48 GMT

In article <1991May7.234142.15065@ddtg.com> zhuo@ddtg.com (Victor Zhuo) writes:
>I am recently working on a very simple program and got a strange
>error back from the system. The program opens a disk as a raw device
>then locks a portion (64 bytes) of it (for read) through system call fcntl().
>Somewhere in fcntl(), the system prints out an error message on console
>"Portmapper not responding; still trying"
>and fcntl() never returns back. This disk has nothing to do with inet.

If your system supports a network lock manager, then *all* file locks must
be registered with it, in case a process on another host tries to lock the
same file via a network file protocol.
-- 
Barry Margolin, Thinking Machines Corp.

barmar@think.com
{uunet,harvard}!think!barmar