Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!bcm!lib!thesis1.hsch.utexas.edu
From: jmaynard@thesis1.hsch.utexas.edu (Jay Maynard)
Newsgroups: comp.unix.questions
Subject: Re: Help with shar needed!
Keywords: shar,help,unix
Message-ID: <4168@lib.tmc.edu>
Date: 8 Oct 90 19:54:26 GMT
References: <1990Oct8.181633.6838@ux1.cso.uiuc.edu>
Sender: usenet@lib.tmc.edu
Distribution: comp.unix.questions
Organization: University of Texas Medical School at Houston
Lines: 19
Nntp-Posting-Host: thesis1.hsch.utexas.edu

In article <1990Oct8.181633.6838@ux1.cso.uiuc.edu> gordon@osiris.cso.uiuc.edu (John Gordon) writes:
>	Hi.  I'm having some problems trying to run the shar program.  I have
>some files in a directory that I want to put in a shell archive, so I cd to the
>directory, then type "shar * > shar.file".  This produces a 5-minute wait, and
>then a "write failed, disk full" error, as shar keeps writing the same stuff
>over and over ad infinitum.  Anyone have any help for me?

I've seen this before. What apparently happens is that the shell creates the
output file before expanding the * into the arguments for shar. Shar goes
along and merrily stuff stuff into shar.file; when it gets to shar.file, it
starts reading from the beginning and writing to the end. You can avoid this
behavior from shar by doing "shar * > somewhere.else/file" - as in put the
file somewhere besides the current directory.

-- 
Jay Maynard, EMT-P, K5ZC, PP-ASEL | Never ascribe to malice that which can
jmaynard@thesis1.hsch.utexas.edu  | adequately be explained by stupidity.
"It's a hardware bug!" "It's a    +---------------------------------------
software bug!" "It's two...two...two bugs in one!" - _Engineer's Rap_