The problem we have with nohup jobs dying from sighups seems to be a bug in sh.
Using the "at" command instead of nohup is a good workaround.
Thanks to:
bf566z9@is000913.BELL-ATL.COM (Chipp Kuldell)
Fernando Frota Redigolo <fernando@caju.larc.usp.br>
Original post:
>
> Hi Managers,
>
> We are running semiconductor simulation software from Silvaco International on
> Sparc 10's under Solaris 2.5. When we start jobs nohup from X-Terminals
> rlogin'ed to the servers and then log out, sometimes the jobs die and the log
> file says they were killed by a SIGHUP. Other times the jobs continue to run
> fine.
>
> We can't seem to find a pattern in this. One time it runs fine. The next time
it
> dies when running the exact same files and commands.
>
> I can't figure out how the SIGHUP is getting past the shell when the programs
> are run in nohup.
>
> The user having the problem is using ksh.
>
> Anyone seen this behavior before? Are there any known bugs in Solaris' ksh
like
> this?
>
> The software runs a convoluted series of shell files to set up the run, but,
> sometimes running it nohup works, so I don't think that is the problem.
>
> Thanks in advance for any hints!
>
> Rick Fincher
> rnf@spitfire.tbird.com
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:11:43 CDT