Thanks for your reply - I understand your points ..   but darn, it seemed so
'easy'  :-)   I come from the days of heavy CMS application/data usage, so a
'reacc' made life nice -- and techniques like renaming files before copying
the new version of the file allowed you to get by 99% of the time and avoid
shared minidisks causing others grief..

I don't suppose such techniques apply here?

-  Rename a file - don't erase it until next 'IPL' so others sharing the
disk don't get corruption.

-  Rename a file before 'replacing' with a new one - so that 'old' users see
the 'old' copy and 'new' users see the new copy.

(TOOLSRUN used these methods IIRC..)

Would those techniques help with a 'shared' Linux filesystem the way they
did with a CMS one, do you think?

Thanks!

Scott

On Tue, Mar 10, 2009 at 10:16 AM, Edmund R. MacKenty <
[email protected]> wrote:

> On Tuesday 10 March 2009 11:44, Scott Rohling wrote:
> >Along these lines ..   does a Linux filesystem on a RO minidisk reflect
> any
> >changes at all if changes are made by a user with RW?
>
> Yes, but you *really* don't want to do that.  Your guest with the RO
> minidisk
> will get corrupted data.  You see, Linux caches blocks it has read from the
> filesystem in memory.  So imagine that it reads in a block containing a set
> of directory entries and caches that.  Now imagine that another guest with
> RW
> access to that filesystem removes that directory.  The RO guest won't know
> about it: it will still happily use that cached directory block when
> reading
> that directory, which contains references to files that no longer exist.
> What happens when it tries to read those files?  It reads those blocks from
> DASD, which may well have been overwritten by the RW guest with some other
> data, because those blocks were freed up when the directory was deleted.
> Oops!
>
> Another bad case is if the RO guest has cached some blocks from an
> executable
> file, and the RW guest has overwritten some or all of those blocks, perhaps
> with another executable.  The RO guest will read in blocks it hadn't cached
> and load it as code, but it has been overwritten by something else (other
> code, a text file, who knows?).  When that process executes whatever is in
> the newly-read block: boom!  It will seg-fault at best.  Or execute some
> other code even!
>
> >Is a deactivate/activate necessary?  re-LINK?   remount?   Anyone know the
> >minimum necessary action to see changes?
>
> You should just never do this.  Do not modify DASD while a Linux guest has
> it
> mounted read-only.  There is no way you can know what parts of that DASD
> are
> cached and what is not.
>         - MacK.
> -----
> Edmund R. MacKenty
> Software Architect
> Rocket Software
> 275 Grove Street · Newton, MA 02466-2272 · USA
> Tel: +1.617.614.4321
> Email: [email protected]
> Web: www.rocketsoftware.com
>
> ----------------------------------------------------------------------
> For LINUX-390 subscribe / signoff / archive access instructions,
> send email to [email protected] with the message: INFO LINUX-390 or
> visit
> http://www.marist.edu/htbin/wlvindex?LINUX-390
>

----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Reply via email to