Re: tlug: telnet: how to get a process to stay alive after disconnecting?

["Thomas O'Dowd" <tom@example.com>]

tluggers,

This is my first post (and also a repost, thanks to not being subscribed
to tlug-post the first time and my mail silently being rejected. (I must
read the full list intro details in future and not just save it as I'm
used to doing). My other first mail is below...

The second subscription is a curious thing though. Although I admit I'm not
on hundreds of mail lists, it is the first time that I've seen this 4 step
(including confirmations) approach to joining and posting to a mail list.

It seems a really wierd thing to do, just to stop spammers posting
to the list. Although I'm not aware of the full reasons behind the second
subscribe, I am curious as to why... Wouldn't it be the same (in spam terms)
to restrict posting to those who were on the main list? If a list member spams,
you knock them off the list... Of course, you may be assuming that people are
writing programs to programmatically join lists (including handling the 
confirmation mail) so that they can mass mail the Tokyo linux community
in particular.

When I didn't see my post go though, I assumed that my mail got stuck
somewhere, then after another while, I thought to myself that maybe first
posts are monitored, so all I had to do was wait for the person doing the
moderating to get to my post to see it. Then after a couple of hours I 
thought, hang on, I'll check the web site again...

ever curious,

Tom.

> You can use the command "nohup" see man nohup or info nohup for more info.
> 
> prompt$ nohup somecommand
> 
> should do the trick. stderr and stdout will go to a file nohup.out. If you
> want to just collect errors then you'll have to redirect stderr separately
> to a file. Something like...
> 
> prompt$ nohup somecommand 2>/tmp/somecommand.err (assuming bash or ksh)
> 
> or you could always just redirect stdout to /dev/null if you don't need
> it and leave errors go to nohup.out
> 
> prompt$ nohup somecommand 1>/dev/null (again it's different for csh, tcsh)
> 
> You'll have to log back in to collect errors unless you are running NFS
> and write the messages to a NFS'd directory. One other thing to keep in
> mind is if the command is long running and errors are far and few between,
> because you're redirecting to a file, there will be no auto flush to 
> the file, so you will have to wait a little to actually see the errors
> if you're doing a tail -f for example. To see errors, real time, then
> you would need to have your errors written to a pseudo terminal which
> autoflushed everything to a log file, but that is a bit complex me
> thinks. Just something to keep in mind that is all...
> 
> Tom.
> 
> On Fri, May 26, 2000 at 12:01:06PM +0900, Ulrike Schmidt wrote:
> > I want to get a process started via telnet and have it finish on its own
> > after I have disconnected and just collect the error messages. How do I do that? 
> > 
> > Thanks, Uli
--------------------------------------------------------------------
Next Nomikai Meeting: June 16 (Fri), 19:00   Tengu TokyoEkiMae
Next Technical Meeting: July 8 (Sat) 13:30   Topic: TBA
--------------------------------------------------------------------
more info: http://www.tlug.gr.jp        Sponsor: Global Online Japan