Re: DESCHALL runs wild when run under nohup?

Colin L. Hildinger (colin@ionet.net)
Mon, 05 May 97 14:15:11 -0500


On Mon, 5 May 1997 14:10:09 -0400 (EDT), Lee Sheridan wrote:

>Recently I have been running the DESCHALL client on the SPARC-1000 of my
>ISP (Baltimore County Public Library, who offers UNIX shell accounts) late
>at night. Because BCPL has a limit of 3 hours for any connection to the
>network, I haven't been able to "squeeze" the full potential out of my
>0-0700 timeframe when I would like to run the client. Last night I
>decided to check and see if DESCHALL would run correctly under nohup. I
>became distracted doing something else, and I thought that I had stopped
>DESCHALL, but apparently I didn't. This morning I received this note from
>the system administrator:
>
>> FYI:
>>
>> When I came in this morning I found the UNIX host running very slowly.
>> The culprit was your proccess "deschall-sparc keymaster.verser.frii.com",
>> which had been running since 11:40 PM Sunday. When I discovered it, it
>> had used a tremendous amount of processor time and was still growing
>> rapidly. I killed it off at about 8:35 AM.
>
>Of course, DESCHALL is *supposed* to use a `tremendous amount of processor
>time'. What I was concerned about was the fact that he mentioned that the
>amount of processor time was `growing rapidly', especially since the
>machine gets busy around that time of the morning and the client should be
>using LESS CPU time...

OK, tell me if this is way out of whack, I'm not familiar with SPARC
models and how many processors they have. If the machine has, say 2
processors, overnight, the CPU load would be about 50% (just one
deschall client running on one of the 2 processors, and not much else).
When the load began to pick up in the morning, the overall CPU load
would begin to increase from 50%, as the other tasks were scheduled on
the other processor. Thus, deschall would still be chewing up most of
one processor, and everything else would be running on the other. Even
when it reached 100%, the system wouldn't be slowed down because
deschall is so 'nice'

Colin L. Hildinger

Games Editor - OS/2 e-Zine!
http://www.os2ezine.com/

The Ultimate OS/2 Gaming Page
http://www.ionet.net/~colin/games.html

The Official Unofficial AWE32 and OS/2 Warp Page
http://www.ionet.net/~colin/awe32.html