[comp.os.eunice] nohup on Eunice?

km@emory.uucp (Ken Mandelberg) (12/03/88)

Does anyone know how to run a Eunice background job which
will survive the logout of the login shell that started it?
Normally these jobs die no matter how one tries to manipulate
the process group or hangup signals, presumably because the
Unix emulation is not faithful in this respect.

I imagine there is some way to exploit the VMS batch queue to
simulate the effect.
-- 
Ken Mandelberg      | km@mathcs.emory.edu          PREFERRED
Emory University    | {decvax,gatech}!emory!km     UUCP 
Dept of Math and CS | km@emory.bitnet              NON-DOMAIN BITNET  
Atlanta, GA 30322   | Phone: (404) 727-7963

jbuck@epimass.EPI.COM (Joe Buck) (12/06/88)

In article <3468@emory.uucp> km@emory.uucp (Ken Mandelberg) writes:
>Does anyone know how to run a Eunice background job which
>will survive the logout of the login shell that started it?
>Normally these jobs die no matter how one tries to manipulate
>the process group or hangup signals, presumably because the
>Unix emulation is not faithful in this respect.

The VMS security model makes this pretty much impossible.  The only
way a user can run a job that survives logout is to have DETACH
privilege.  But the way VMS defines DETACH privilege, a user with that
privilege can create a job running as any user, and therefore can do
anything.  Stupid, right?  But there's no way around this.
-- 
- Joe Buck	jbuck@epimass.epi.com, or uunet!epimass.epi.com!jbuck,
		or jbuck%epimass.epi.com@uunet.uu.net for old Arpa sites

<picture your favorite pithy quote here>

jeh@crash.cts.com (Jamie Hanrahan) (12/07/88)

In article <2703@epimass.EPI.COM> jbuck@epimass.EPI.COM (Joe Buck) writes:
>In article <3468@emory.uucp> km@emory.uucp (Ken Mandelberg) writes:
>>Does anyone know how to run a Eunice background job which
>>will survive the logout of the login shell that started it?
>The VMS security model makes this pretty much impossible.  The only
>way a user can run a job that survives logout is to have DETACH
>privilege.  But the way VMS defines DETACH privilege, a user with that
>privilege can create a job running as any user, and therefore can do
>anything.  Stupid, right?  But there's no way around this.

1.  VMS V4.0 introduced the "detached processes" quota.  This is set by 
the /MAXDETACH qualifier on the AUTHORIZE utility's MODIFY, ADD, etc., 
commands.  The default value of 0 represents an unlimited number.  With
this quota, a nonprivileged user can create the indicated number of 
detached processes with the same UIC as that of the creating process,
without needing the DETACH privilege (which does indeed permit the creation
of a process with any UIC).  

2.  Or, the interactive user can always start a batch job.  I don't know
how well this applies to Eunice, though....

sword@vu-vlsi.Villanova.EDU (Ronin) (12/08/88)

In article <3468@emory.uucp>, km@emory.uucp (Ken Mandelberg) writes:
...
> I imagine there is some way to exploit the VMS batch queue to
> simulate the effect.
> -- 

I'm not sure if it will simulate the effect, but I just
executed the command:

vms submit/noprint/nolog/notify/restart test.com

from the Eunice prompt and the batch job fired off and 
completed okay (running Eunice 4.2). I'm not sure if this
is what you meant.

..lar

-----
Larry Esmonde, UCIS Dept., Villanova University
UUCP   : {cbmvax,bpa,psuvax1}!vu-vlsi!excalibur!159628317 [preferred]
       : {cbmvax,bpa,psuvax1}!vu-vlsi!sword
BITNET : 159628317@vuvaxcom, larry@villvm
CSNet  : esmonde@villanova.edu