Jeff--
listf,access looks really good. A few lingering thoughts...
Is this functionality going to be added to :LISTF, :LISTFILE or both? If
only one, I'd prefer :LISTFILE. Both is OK too.
Regarding sort order, file open vs. JSNo: are you considering making both
options available? If so, with what request syntax? If not, I assume you
want opinions on which is better - I vote for JSNo order for easy reading,
although there are some situations where it might be useful to see some
clues about who got to the file first.
I'd like to suggest two small changes which are mostly cosmetic but will
also make parsing easier. Change:
>#S12345 JOBNAME8,USER5678.ACCT5678 -lock- LDEV: 12345
>#S40 FRED,MGR.PAYROLL,PUB -lock- REM: clientmpe.cup.hp.com
>#J15 JSPOOL,RSPOOL.SYS,RSPOOL SPID: O7654321
To:
>#S12345 JOBNAME8,USER5678.ACCT5678 -lock- LDEV: 12345
>#S40 FRED,MGR.PAYROLL,PUB -lock- REM: clientmpe.cup.hp.com
>#J15 JSPOOL,RSPOOL.SYS,RSPOOL SPID: #O7654321
Note that REM: is now colon-aligned with LDEV: and SPID:, and the associated
data starts in the same column. It might also be a good idea to add the #
before the Onnn for SPID's, to make them more easily distinguishable from
node names without looking at the label (is # a valid character at the start
of a DNS name?).
About the -lock- flag-- I confess I'm somewhat ignorant about locking
levels, types, rules, etc., but the main thing I want this flag to help me
do is isolate the single culprit that is causing other processes to 'hang'
waiting for a file. In cases where more than one process has some sort of
lock on a file is there an easy way, like exclusive vs. shared, to
distinguish who the real 'bad guy' is? If so, I'd recommend that the
accessor be flagged differently, perhaps -LOCK- instead of -lock-.
Related to the lock issue, there are some times when it's not easy for me to
know *which* file is being fought over. Might there be a way to request a
:LISTFILE,ACCESS for only those files that are actually being accessed?
Something along the lines of...
:LISTFILE @[log in to unmask]@;seleq=[ACCESSED=TRUE];format=ACCESS
...well, there I go blue-skying again. I could probably draw up a whole
suite of additions I'd like to the seleq structure.
Thanks again, Jeff, for your hard work on this project and your continued
responsiveness to user input.
-----------------------------------------------------------------------
Jon Diercks -------- mail: [log in to unmask]
Programmer/Analyst /|| | talk: [log in to unmask]
Computing Services /_|| | WWW : http://rowlf.csv.anderson.edu/
Anderson University / ||__| tel : (317)641-4305
Anderson, IN 46012 -------- fax : (317)641-3851
o________________________________________________________ ___
_\_, \=`==^==>
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
