VMS question - HECnet

10 Aug 2015

Johnny Billquist <bqt at softjar.se> writes:

...
 On 2015-08-10 16:55, Brian Schenkenberger, VAXman-
wrote:
  Johnny Billquist <bqt at softjar.se>
writes:

  Well, if I wanted then, even easier is to just
set a version limit on=20
 the files. :-) 
 Both your solution (version limits) and the periodic batch job purge will
 not stop the unnecessary log file I/O, and both add to disk I/O and file
 system activity.  If this is a known problem with V5.1 and it's fixed in
 V5.3, I'd suggest that you obtain that patch/version/eco and install it.
 I, personally, loathe solutions to problems that elicit other problems or
 increase resource loads. 
Agree. But besides, I do not (so far) have excessive number of 
generations of log files in the first place. I just felt that if I 
wanted to keep the number of kept generations low, I would just put a 
limit on them instead of having a batch job run regularly that did a 
purge... 
Well, you're getting an implicit purge each time a new file is created with
a version limit.  Probably much better than filling up a directory with log
files and then, purging but you're still purging.

...
 But the known problem in 5.1 that I was referring to
was the total 
failure of the telnet daemon, by which is just spews 0xff after 
something manages to corrupt it. I also have a full disk, which I yet do 
now know for sure why.

  FYI, eventually, your version limit solution will
require intervention as
 you inevitably crash and burn when you hit ;32,767. 
That is also true. 
I had a client hit 32,767 on the day of the Mayan apocalypse.  It wasn't the
end of the world but it was the end of their production for that day. ;)

-- 
VAXman- A Bored Certified VMS Kernel Mode Hacker    VAXman(at)TMESIS(dot)ORG

I speak to machines with the voice of humanity.