HP3000-L Archives

August 1996, Week 5

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Jeff Vance <[log in to unmask]>
Reply To:
Jeff Vance <[log in to unmask]>
Date:
Tue, 27 Aug 1996 23:45:31 -0700
Content-Type:
text/plain
Parts/Attachments:
text/plain (41 lines)
On Aug 27,  8:15pm, Michael D. Hensley wrote:
> Subject: Re: Who's accessing a file (more)
>snip...
 
> BIG GIANT CAVEAT: the AIF manual says that the particular call which
> returns information about who has a file open doesn't work for
> "compatibility mode files".  No, they don't explain what that means.
 
Usually "CM file" refers to files that are still implemented in compatibility
mode (CM).  In 5.0 this includes CIR, RIO and (I think??) terminal files.
MSG files used to be in CM but are now native mode (NM).  I'm not sure why the
AIF manual says this since if you are using item 2065 in AIFSYSWIDEGET
all that is returned is the PID for each process that has a PLFD entry (i.e.
has "fopened" the file).  In the simple test that I did a CM CIR file has
a plfd entry and thus the AIF call should work fine.  Note: not all of the
fields are kept current for CM files but the plfd does seem to exist and
that is enough to determine the accessor.
 
>
> If I recall correctly, Eugene mentions somewhere in the MPEX manual
> that %LISTF,ACCESS was one of the hardest things he had to do.  I
> suggest, if you *really* need to know who has a file open, you use
> MPEX.  I *think* all of the free stuff uses the AIFs, which are no
> where near as reliable (for this function).
 
I thought that the MPEX %LISTF,access uses the AIF!?  I'm sure I'll be
corrected if this is wrong.  Anyway, I know there were problems in the past
wrt. deadlocks; however this has been fixed in 4.0? 4.5?  BTW the new CI
:LISTF,access feature uses the AIF method to obtain all file accessors, and
I checked it with CIR files and it does show who is accessing this CM file.
But the current record number is not accurate, which I guess is a "feature"
of it being a CM file...
 
snip...
 
regards,
Jeff Vance, CSY
 
 
--

ATOM RSS1 RSS2