On 06.07.2016 18:42, Frank Ch. Eigler wrote:
> Alexander Lochmann <alexander.lochmann@tu-dortmund.de> writes:
> 
>> [...]
>> I chose the letters randomly. For me, it doesn't matter. I just need an
>> argument to tell staprun to create a pid file. :)
> 
> Would it be enough for staprun to have that pid option?
> That would mean the stap script would have to be compiled
> through to -p4 only, then staprun invoked manually.
> 
Yes, that should do.
>> Yeah, sure. Since I run stap on Android, I have a background service,
>> which periodically checks if every single stap instance is still
>> running. Yes, it might be possible having more than one stap script
>> running. :) [...]
> 
> Since your background service is already custom, maybe a custom shell
> script wrapper like this could serve your purposes without having to
> change staprun/stap at all:
> 
> % cat pidwrapper
> #! /bin/sh
> pidfile="$1"; shift
> program="$1"; shift
> echo $$ > $pidfile
> exec $program ${1+$@}  # exec to preserve $$ pid
> 
> and then you invoke staprun via:
> 
> % sh pidwrapper /path/to/this.pid staprun foo bar baz
> 
> 
That would be a nice workaround. I went through the source code. The
commandline option is used by stapio. Stapio creates the pidfile. If I
send a SIGKILL (or was it a SIGTERM? what so ever) to staprun, stap and
the module won't be shutdown properly. Hence, I have to send the signal
to stapio.

- Alex
> - FChE
> 


-- 
Technische Universität Dortmund
Alexander Lochmann                PGP key: 0xBC3EF6FD
Otto-Hahn-Str. 16                 phone:  +49.231.7556141
D-44227 Dortmund                  fax:    +49.231.7556116
http://ess.cs.tu-dortmund.de/Staff/al