If this isn't exactly what you wanted, please try our Search (there's a LOT of techy and non-techy stuff here about Linux, Unix, Mac OS X and just computers in general!):
From - Thu Aug 5 12:24:25 1999
Xref: world comp.unix.sco.misc:102422
Newsgroups: comp.unix.sco.misc
Path: world!newsfeed.mathworks.com!news.maxwell.syr.edu!news.msfc.nasa.gov!info.usuhs.mil!uky.edu!xenitec!news
From: "Radek Tomis" <rts@mediumsoft.cz>
Subject: Re: Disk write shell script
Resent-From: mmdf@xenitec.on.ca
Submit-To: scomsc@xenitec.on.ca
Content-Type: text/plain;
charset="iso-8859-2"
Cc: "Fabio Giannotti" <fabiog@venmar.com>
Organization: Medium Soft, a.s.
Date: Thu, 5 Aug 1999 13:01:54 GMT
Message-ID: <023b01bedf44$65984240$8e0314ac@rtsnt>
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
References: <37A8C3D4.25EFC00C@venmar.com>
Sender: news@xenitec.on.ca (xenitec.on.ca News Administrator)
Precedence: list
Lines: 84
X-Mozilla-Status: 8011
> From: Fabio Giannotti <fabiog@venmar.com>
> Sent: Thursday, August 05, 1999 12:51 AM
> Is there a way using bourne shell to write directly to the hard disk?
>
> That is to say, if you write a "regular" shell script that writes 100
> lines to a file, it may only hit the disk 1 or 2 times. I assuming it's
> buffering or caching a bunch, then writing a bunch all at once.
>
> What I want is for each line written to "go right to disk" and not be
> buffered/cached.
>
> Any ideas?
If you want transparent synchronous output, then download this 35 KB binary:
http://members.xoom.com/rts101/tmp/fildes
and use it inside (or outside) of your script in one or more of the
following ways:
1. Inside your script for individual output commands:
fildes 1=~,A+S : echo "line" >> /tmp/file
This will append "line" to "/tmp/file" synchronously (i.e. the line will
be physically written to the disk drive before the command 'echo'
returns).
Consider the following example using different output command:
fildes 1=~,S : tail -300 /usr/adm/messages > /tmp/file
This will synchronously append last 300 lines of "/usr/adm/messages"
to "/tmp/file. However, 'tail' (and almost all other commands/utilities)
uses user-level standard buffered I/O (from standard C library), and if
the output is directed to something other than a terminal device, then
the output is buffered by constant block size, usually 1 KB (depends on
implementation). So, the 'tail' above actually writes (calls kernel
routine write(S)) to "/tmp/file" only once per 1 KB (no matter whether
this happens to be in the middle of the read line or not). In this case,
each such 1K block gets written synchronously (the 1K block physically
appears on the disk drive before the write(S) call returns).
2. Inside your script for several output commands at once:
fildes 1=/tmp/file,A+S : sh -c '(
set -- comp unix sco misc
echo "`date`: arguments:"
for arg
do
echo " arg: ${arg}"
sleep 1 # some processing
done
)'
This will append several lines to "/tmp/file". Each line will be written
synchronously (the line will be physically written to the disk drive
before the command 'echo' returns).
3. Outside of your script for the entire script:
fildes 1=/tmp/file,S : <your_script>
This will run your script with standard output redirected to "/tmp/file".
Every standard output from your script (and all its children, i.e. all
external commands used within your script) will be synchronous, i.e. all
underlying write(S) calls completes after all data in the given write(S)
call have been physically written to the disk drive.
You can also use standard system command 'sync' after each line output,
however, it will write out not only this last line to the disk, but also all
other dirty file system blocks from system memory buffers for all file
systems. Depending on the recent write activity in the system, it may of
course take much longer than actually needed for your one-line write.
--
Radek Tomis
rts@mediumsoft.cz
/Bofcusm/77.html copyright 1997-2004 (various authors) All Rights Reserved
Have you tried Searching this site?
Unix/Linux/Mac OS X support by phone, email or on-site: Support Rates
This is a Unix/Linux resource website. It contains technical articles about Unix, Linux and general computing related subjects, opinion, news, help files, how-to's, tutorials and more. We appreciate comments and article submissions.
Many of the products and books I review are things I purchased for my own use. Some were given to me specifically for the purpose of reviewing them. I resell or can earn commissions from the sale of some of these items. Links within these pages may be affiliate links that pay me for referring you to them. That's mostly insignificant amounts of money; whenever it is not I have made my relationship plain. I also may own stock in companies mentioned here. If you have any question, please do feel free to contact me.
Specific links that take you to pages that allow you to purchase the item I reviewed are very likely to pay me a commission. Many of the books I review were given to me by the publishers specifically for the purpose of writing a review. These gifts and referral fees do not affect my opinions; I often give bad reviews anyway.
We use Google third-party advertising companies to serve ads when you visit our website. These companies may use information (not including your name, address, email address, or telephone number) about your visits to this and other websites in order to provide advertisements about goods and services of interest to you. If you would like more information about this practice and to know your choices about not having this information used by these companies, click here.
Click here to add your comments
Don't miss responses! Subscribe to Comments by RSS or by Email
Click here to add your comments
If you want a picture to show with your comment, go get a Gravatar