bmacintyre@watcgl.waterloo.edu (Blair MacIntyre) (05/18/89)
One thing I noticed with my recently ( today here ) posted backup.rexx
was that is will not work if run more than once concurrently. This
has to do with the naming of the temporary pipes.
The reason I discovered this was because I had multiple cron statement
executing it, such as
rx backup dh0:rexx backup:dh0/rexx
rx backup dh0:s backup:dh0/s
rx backup dh1:c backup:dh1/c
and one or more of them hung ...
But, running just one of them ( the last one, for example ) creates
great backups of my c directory while I'm working.
So, a question:
is a pipe name like
pipe:dh0/dir/dir/dir/dir/blair/macintyre/yet/another/dir
valid? If something like that will work ( ie. I could test it, but
is it _guaranteed_ to work based on the way the 1.3 pipe device was
written ) I will fix it quite easily ...
All you have to do is name the pipefile differently, instead of
"backup_ls_" || depth, us something like x || depth ...
The other option I was thinking of was to create a file called
backup.dirlist in rexx: or s: and have it contain a list of pairs
dh0:rexx backup:dh0/rexx
dh0:s backup:dh0/s
....
which the one backup program would execute in sequence. That would
probably be better from the point of view of not having N rexx processes
running at once, killing the CPU ... :-)
--
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-///-=
= Blair MacIntyre, bmacintyre@watcgl.{waterloo.edu, UWaterloo.ca} \\\/// =
= now appearing at the Computer Graphics Lab, U of Waterloo! \XX/ =
= "Don't be mean ... remember, no matter where you go, there you are." BBanzai=