From xemacs-m  Mon Dec 16 15:33:26 1996
Received: from newman (root@newman.aventail.com [38.225.141.10]) by xemacs.cs.uiuc.edu (8.8.3/8.8.3) with SMTP id PAA26794 for <xemacs-beta@xemacs.org>; Mon, 16 Dec 1996 15:33:24 -0600 (CST)
Received: from kramer.in.aventail.com.aventail.com (root@newman [192.168.1.1]) by newman (8.6.12/8.6.9) with SMTP id NAA08243; Mon, 16 Dec 1996 13:32:20 -0800
Date: Mon, 16 Dec 1996 13:32:20 -0800
Message-Id: <199612162132.NAA08243@newman>
From: William Perry <wmperry@aventail.com>
To: Dave Mason <dmason@plg.uwaterloo.ca>
Cc: Benedikt Eric Heinen <beh@icemark.thenet.ch>, xemacs-beta@xemacs.org
Subject: Re: 0 length files on full disk (I'm glad this isn't a stupid question)
In-Reply-To: <199612162129.QAA25599@plg.uwaterloo.ca>
References: <199612161431.JAA21057@plg.uwaterloo.ca>
	<Pine.LNX.3.95.961216225608.2312C-100000@icemark.thenet.ch>
	<199612162129.QAA25599@plg.uwaterloo.ca>
Errors-to: wmperry@aventail.com
Reply-to: wmperry@aventail.com
X-Face: O~Rn;(l][/-o1sALg4A@xpE:9-"'IR[%;,,!m7</SYF`{vYQ(&RI1&EiH[FvT;J}@f!4kfz
 x_!Y#=y{Uuj9GvUi=cPuajQ(Z42R[wE@{G,sn$qGr5g/wnb*"*ktI+,CD}1Z'wxrM2ag-r0p5I6\nA
 [WJopW_J.WY;
Mime-Version: 1.0 (generated by tm-edit 1.1)
Content-Type: text/plain; charset=US-ASCII

Dave Mason writes:
>Benedikt Eric Heinen writes:
>> > It now turns out that C-X C-S doesn't even complain if the write
>> > fails.  auto-save also happily says that it has auto-saved, but
>> > there's a nice 0-byte autosave file!!!  C-X C-C asks to save dirty
>> > buffers, then writes 0 bytes and exits!!!  I know this used to work
>> > until about 19.13.
>> 
>> I just tried recreating the error on linux (Dave uses SUN-OS), and he's
>> right here. But, strangely enough, there IS a file system full error
>> message, but that occurs only, if I try and write a file with more than
>> 64K on a full partition. Anything below 64K, doesn't yield any error
>> message - not writing anything though.
>
>Ditto here. 65536 bytes - no error. 65537 - error.
>
>This is on a local disk.
>
>64K isn't a disk allocation unit, either.  If I fill up the disk so
>there's only 32K left, I get the same behaviour: 64K works (and writes
>32K!!!), 64K+1 fails.  But cat sees the error.  Here's an example
>(with the rc shell):
>
>; ls -l FOO
>-rw-rw-r--   1 dmason   pg_user    32768 Dec 16 16:19 FOO
>; x=`{cat FOO}
>; echo $x|wc
>       1       1   32769
>; echo $x|cat >FOO
>cat: output error (0/2049 characters written)
>No space left on device
>; ls -l FOO
>-rw-rw-r--   1 dmason   pg_user    30720 Dec 16 16:20 FOO
>
>Sorry, I don't have FSF emacs installed, and 19.13 is gzipped up
>because we're short of disk space (can you guess how I discovered this
>problem in the first case? :-) so I can't easily try other versions.

  Stallman recently mailed out a request to see if anybody else was seein
this problem to the emacs beta testers, so it seems to be a bug we all
have.

-bp

