YATES@a.chem.upenn.EDU ("John H. Yates") (04/28/87)
Is there any way to retrieve the last accessed date for files? I know last revised is available, but haven't found if RMS keeps the last time a file was simply read. I at one time considered archiving user files that had not been accessed in n days, but gave up when the hook did not seem to exist. I know it would slow down i/o and for this reason may not be available, and yes, I know users could easily beat the archive system, by copying their files to nl: , etc. , but what they don't know will help me free up disk space. John
carl@CITHEX.CALTECH.EDU.UUCP (04/29/87)
> Is there any way to retrieve the last accessed date for files? I know > last revised is available, but haven't found if RMS keeps the last > time a file was simply read. To answer your question literally: no. However, you CAN get at a date that is related to the last accessed date; namely the expiration date. You can set two parameters for any given volume that determine how the expiration date is computed. For more information, type the command: $ HELP SET VOLUME/EXPIRATION
rick@NGP.UTEXAS.EDU (Rick Watson) (04/29/87)
> Is there any way to retrieve the last accessed date for files? I know >last revised is available, but haven't found if RMS keeps the last >time a file was simply read. Look at SET VOLUME /RETENTION=. It will probably do what you want. Rick Watson University of Texas Computation Center arpa: rick@ngp.UTEXAS.EDU rick@ngp.ARPA uucp: ...seismo!ut-sally!ut-ngp!rick rick@ut-ngp.UUCP bitnet: ccaw001@utadnx span: utspan::ccaw001 phone: 512/471-3241
dp@JASPER.PALLADIAN.COM (Jeffrey Del Papa) (04/29/87)
Date: Mon, 27 Apr 87 22:08 EST From: "John H. Yates" <YATES%a.chem.upenn.edu@cis.upenn.edu> Is there any way to retrieve the last accessed date for files? I know last revised is available, but haven't found if RMS keeps the last time a file was simply read. I at one time considered archiving user files that had not been accessed in n days, but gave up when the hook did not seem to exist. I know it would slow down i/o and for this reason may not be available, and yes, I know users could easily beat the archive system, by copying their files to nl: , etc. , but what they don't know will help me free up disk space. John the file system does not currently record that time. Every time someone asked Andy Goldstien for it he groaned and muttered about doubling the number of disk hits. He really doesn't want to. It has been on the decus wish list for years. <dp>
DHASKIN@CLARKU.BITNET (Denis W. Haskin, Manager, Technical Services) (04/29/87)
From: "John H. Yates" <YATES%A.CHEM.UPENN.EDU@CIS.UPENN.EDU> > Is there any way to retrieve the last accessed date for files? I know > last revised is available, but haven't found if RMS keeps the last time > a file was simply read. No, for reasons you mention below, among others. What about security and protection codes? Being able to update the last-accessed date would require modify access to the file (or is it the parent directory), and how does one handle that when the user looking at the file only has read priv and no modify priv? > I at one time considered archiving user files that had not been > accessed in n days, but gave up when the hook did not seem to exist. > I know it would slow down i/o and for this reason may not be available, > and yes, I know users could easily beat the archive system, by copying > their files to nl: , etc. , but what they don't know will help me free > up disk space. We thought about it too and gave up. When you start investigating, you will probably find that any system is highly unlikely to be worth the amount of time you are going to have to put into it. We purge files a few times per year and it's holding us (though admittedly only just -- remember that disk use will invariably expand to fill available disk space). Appeal to your user community's sense of camaraderie and the concept of *shared* resources and the like (how about posting a list of the worst offenders? Nothing like a little peer pressure!). I would add that commercial archiving products seem to be maturing to the point where we may start considering one (and what about the rumors that DEC has one that they use in-house all the time?). % Denis W. Haskin Manager, Technical Services % % ----------------------------------------------------------------------- % % DHASKIN@CLARKU.BITNET Office of Information Systems (617)793-7193 % % Clark University 950 Main Street Worcester MA 01610 % % % % "Anyone who _moves_ before Most Holy comes back out will spend the rest % % of eternity sipping lava through an iron straw." - Cerebus %
RAY@CRVAX.SRI.COM.UUCP (04/30/87)
I was just looking at this information last night...If you look in the System Manager's Reference, Vol. 5A, on page 7-32 in section 7.8.1, Setting File Expiration Dates, you will find "You can simulate the maintenance of "access dates," available in some other operating systems, by setting the retention periods to very small values (for example, 1 hour). Note, however, that doing so will incur substantial overhead in the file system in updating expiration dates so frequently." The go on further to describe the general technique of doing automatic file migration. Ray
RAY@CRVAX.SRI.COM.UUCP (04/30/87)
I was just looking at this information last night...If you look in the System Manager's Reference, Vol. 5A, on page 7-32 in section 7.8.1, Setting File Expiration Dates, you will find "You can simulate the maintenance of "access dates," available in some other operating systems, by setting the retention periods to very small values (for example, 1 hour). Note, however, that doing so will incur substantial overhead in the file system in updating expiration dates so frequently." The go on further to describe the general technique of doing automatic file migration. Ray #